This PDF file contains the front matter associated with SPIE Proceedings Volume 6625, including the Title Page, Copyright information, Table of Contents, Introduction (if any), and the Conference Committee listing.

Landsat TM is the unique resource for global change research and applications in agriculture, geology, forestry, regional planning and so on. As the development of remote sensing applications, various water body extraction methods have been researched upon and developed in TM data. But there still are some limitations in the extracting water bodies which are partly confused with some residential districts. This paper concentrates on extracting water bodies by using decision tree classifiers based on TM images. Firstly, the formula TM2+TM3>TM4+TM5 was used to extract reservoirs, ponds and broad rivers. Then according to the mechanism and spectral characteristics of water body and other objects in TM data, the structure of the decision tree and the locations for each notes in the tree are determined. Finally, based on the model established using ENVI software, water area is extracted automatically, and the yield images are checked by visual and statistical accuracy assessment. The results show the application of decision tree is simple and could improve the accuracy of water body extraction. The decision tree, however, missed small water bodies at scales below the sensor resolution.

Different electronic display devices, like cathode ray tubes (CRTs), liquid crystal displays (LCDs) and plasma display panels (PDPs), show different performance levels with respect to motion blur and flicker visibility. Both motion blur and flicker visibility reduce the image fidelity and hence also the image quality. Developments in device design and industrial standardization require an evaluation system to characterize these properties for preferably all types of display devices. Starting from a photodiode, a measurement setup was built, which records the time varying luminance signal of these displays. Then a simulation tool was developed to predict the motion blur appearance and perceived seriousness of flicker, which tool was validated with perception experiments. A tradeoff between flicker visibility and motion blur was derived from this work. That is, the device with less motion blur shows more serious flicker, like a CRT, while in general a LCD with less flicker shows more motion blur. PDPs exhibit an intermediate performance. New technologies like scanning backlight LCD, and field emission displays (FEDs) meet the same problem. Explanations and analysis of such dilemma from the point of view of information and signal processing are presented. It shows that high frame rate is the best solution for the dilemma, where motion estimation and compensation is an effective alternative to increase the frame rate for the current video broadcasting system.

A novel non-zero order joint transform correlator (NOJTC) is designed based on wavelet transform. In this technique, joint transform power spectrum (JPS) is decomposed into four sub-bands by wavelet function and is reconstructed using only horizontal or vertical detail coefficients (rely on alignment of the joint input images). Thus, zero order term is removed and desirable correlation signals are obtained. To achieve high discrimination ability, fringe-adjusted filter(FAF) and edge detection methods are applied to this wavelet-based non-zero order JTC (WNOJTC) and perform rather well. Detailed analysis followed by simulation results is presented to demonstrate the effectiveness of the proposed technique.

In this work, we present some practical methods for analyzing the structure of spatial-temporal patterns with Matlab, including "Fourier transform", "obtaining the sketch of the pattern", "obtaining the brightness distribution in the pattern", "filtering the noise in the pattern", and so on. The types of the patterns can be determined by obtaining the spatial frequency spectrum of the patterns using Fourier transform. To obtain the sketches of the patterns can make the measurement of pattern structure parameters easier and more accurately. To obtain the brightness distribution in the patterns can help us to analyze the physics mechanism of the pattern formation systems. To filter the noise in the patterns can make the pattern pictures more clearly. This work can provide a beneficial reference for researchers who study pattern dynamics in different systems.

Medical image registration is one basic task in medical image processing. It can align multi images coming from different modes or time, and then it offer guarantee to images post processing. In this article we mainly research the mutual information matching method, and using image pixel grey to compute mutual information. We put forward improved registration algorithm. We compare improved mutual information registration method and classical first grade mutual information and second grade mutual information registration method. During aligning, by extracting image feature signs or edge information, using them as reference feature value of two images aligning information and improving multi parameters optimization algorithm, we get satisfactory experimental results. Regarding to the two-dimensional medical image registration, we utilized the improved method combining gradient information and mutual information, has solved the question of partial extreme value without spatial information when pure use the mutual information. The feasibility of this algorithm is proved by corresponding experiments.

The development of CT and MRI etc. technique offers the means by which we can research directly human internal structure. In clinic, usually various imaging results of a patient are combined for analysis. At present, in the most case, doctors make a diagnosis by observing some slice images of human body. As complexity and configuration diversity of the structure of human body organ, and as well unpredictiveness of focus location and configuration, it is difficult to imagine the cubic configuration of organs and their relationship from these 2D slices without corresponding specialty knowledge and practical experience. So it isn't satisfied with preferable requests of medical diagnosis that only aligning two 2D images to get one 2D slice image. As a result we need extend registration t problem to 3D image. As the quantity of 3D volume data are much more, it undoubtedly increases calculation quantity for aligning two 3D images accurately. It forces us to find some good methods that can achieve better effect on precision and satisfy the demand for time. So in this paper digitally reconstructed radiograph (DRR) image method is proposed to solve correlative problems. Ray tracking two 3D images and digitally reconstruct to create two 2D images, by aligning 2D data to realize to align 3D data.

The traditional tabletop AR systems general use head mounted display (HMD) that has some shortcomings, such as the imprecise precision and low flexibility. To solve these problems, a new design of video see-through tabletop system based on ADSP-BF561^[1,2] is presented, it can fix the display system on the end of the orientation system, that made up of mechanism and angle sensor, and move in a circular area measured by pole length, observe every orientation. These implemented system registration, image collection, transmission and display.

Virtual reality has played an increasing role in such areas as medicine, architecture, aviation, engineering science and advertising. However, in the art fields, virtual reality is still in its infancy in the representation of human movements. Based on the techniques of motion capture and reuse of motion capture data in virtual reality environment, this paper presents an assessment method in order to evaluate the quantification of dancers' basic Arm Position movements in Chinese traditional dance. In this paper, the data for quantifying traits of dance motions are defined and measured on dancing which performed by an expert and two beginners, with results indicating that they are beneficial for evaluating dance skills and distinctiveness, and the assessment method of digital Chinese dance movements based on virtual reality technology is validity and feasibility.

In order to detection and position the center of laser spot, the method based sub-pixel technology is set up. Under the conditions of no change on the dimensions of CCD detector, sub-pixel method can improve the position precision up to sub-pixel level. The basic principle of sub-pixel method is to precisely position the target with sub-pixel precision by applying the inner interpolation algorithm with the gray distribution characters of target spot in the pattern. At first, the algorithm of edge detection is used for original image and the precision is pixel level. Then, the algorithms to interpolation and approximation based on the first step are used to extract edge, the precision reaches sub-pixel level. Experiments show the position precision to center of laser spot is about 0.1μm.

The visual document image is the electronic image about newspapers, books or magazines taken by the digital camera, the digital vidicon etc. Whose getting is more convenient than got from the scanner. Along with the development of OCR technology, visual document images could be recognized by OCR. Affected by some factors, digital image will be degraded during its acquisition, processing, transmission. One of the main problems affecting image quality, leading to unpleasant pictures, comes from improper exposure to light. So preprocessing is becoming much more significant before recognition in order to get an appropriate image satisfied recognition requirements. For the low contrast images with underexposure, according to the visual document image's characteristic, a new algorithm, based on image background separation, for image object enhance is proposed, The proposed method calculate the threshold of separation firstly, And different processing be taken on foreground and background: Various gray values in image background will be merged into unitary gray value, whereas the contrast of foreground will be enhanced. The proposed algorithm implemented in Visual C++ 6.0, and compared the result of proposed algorithm with the results of Otsu's method and histogram equalization. The experimental results show clearly that this algorithm could enhance the details of image object adequately, increase the recognition rate, and avoid the block effect at the same time.

As a new type of aviation remote sensing earth observation system, the UAVRSS (Unmanned Aerial Vehicle remote sensing system) is used in civil remote sensing field more and more. In order to improve the efficiency of the remote sensing image processing and making the Orthophoto of the UAVRSS, in this paper one method is presented to improve the precision of the Orthophoto without the Ground Control Point(GCP) and the high precision sensors, such as the POS and the IMU. Through some real flying experiments of the UAVRSS, the data and the images were obtained. These data and the images were analyzed by the method. The result shows that the precision of the Orthophoto.

Virtual Endoscope is a method to emulate cavity checking visually combined row volume data obtained from CT and MR with three-dimensional image technologies in virtue of navigation, flythrough and pseudo-color technologies. The application on virtual endoscope is developed in recent several years, and the software realization needs the support of multiple complicated algorithms, including internal surface reconstruction, center path automatic extraction, lens setting-up, multi-case processing, collision detection, and corresponding algorithm computation and realization, which cause the application software development for Virtual Endoscope is rather complex and difficult. It puts forward volume rendering for rapid three-dimensional reconstruction, introduces highly active path planning algorithm of three-dimensional space path algorithm, improved path smooth algorithm and lens entry point auto-detecting algorithm, illustrates three-dimensional scene establishment by VTK development toolkit, and discusses the key technologies in virtual endoscope realization in this paper, based on which the virtual endoscope system has graceful performance.

Three-dimensional image reconstruction by volume rendering has two problems: time-consuming and low precision. During the diagnosis procedure, some detailed organ tissue is the interest to doctors, so the reconstructed two-dimensional images are pre-processed before three-dimensional reconstruction including disturbance removing and precise segmentation, to obtain Region-Of-Interest (ROI) based on which three-dimensional reconstruction carries through, that can decrease the complexity of time and space. By this, Live Wire segmentation algorithm model for medical image is improved to gain exact edge coordinate for the image segmentation with interior details by improved filling algorithm. Segmented images with object details only are regarded as input to realize volume rendering by ray casting tracking algorithm. Because the needless organs have been filtered, the disturbance on interested objects for doctors is reduced. Meanwhile, generally speaking, these needed organs left are less proportion in images. So it reduces data amount of volume rendering, and improves the speed of three-dimensional reconstruction.

LiDAR data interpolation is an important step in LiDAR data processing. Interpolation arithmetic involves high time complexity. Now it is difficult to tackle it by existing centralized processing because of increasing datasets. Emerging Grid technologies is a cost-effective solution for solving scientific problems that involve large datasets and complex analyses. In this paper we present a grid-based architecture to coordinate various resources as LiDAR data processing platform, utilizing centralized task scheduling and parallel computing technologies. LiDAR data is partitioned into appropriate slice and neighboring slice boundary is overlap. These slices are send to each node by scheduler. Computation speed is improved by parallel computing of each node. Grid service is the interface of interpolation arithmetic. In addition, it is beneficial to make these available to the community in a unified framework through a portal, so scientists can focus on their scientific work and not be concerned with the implementation of the underlying infrastructure.

In general image restoration, the point spread function (PSF) of the imaging system, and the observation noise, are known a priori information. The aero-optics effect is yielded when the objects ( e.g, missile, aircraft etc.) are flying in high speed or ultrasonic speed. In this situation, the PSF and the observation noise are unknown a priori. The identification and the restoration of the turbulence degraded images is a challenging problem in the world. The algorithm based on the nonnegativity and support constraints recursive inverse filtering (NAS-RIF) is proposed in order to identify and restore the turbulence degraded images. The NAS-RIF technique applies to situations in which the scene consists of a finite support object against a uniformly black, grey, or white background. The restoration procedure of NAS-RIF involves recursive filtering of the blurred image to minimize a convex cost function. The algorithm proposed in this paper is that the turbulence degraded image is filtered before it passes the recursive filter. The conjugate gradient minimization routine was used for minimization of the NAS-RIF cost function. The algorithm based on the NAS-RIF is used to identify and restore the wind tunnel tested images. The experimental results show that the restoration effect is improved obviously.

The fusion techniques have been developed quickly in recent years and become an important remote sensing research topic. This paper systematically discusses the technique used in pixel level image fusion including IHS transform, YIQ transform, HLS transform, HSV transform, PCA transform, HPF transform and wavelet transform image fusion method. Then a fusion experiment of IKONOS image is made to compares the different merging methods from spectral quality and the spatial quality in order to choose the suitable method for the high resolution image.

Image segmentation is one of the difficult research problems in the machine vision industry and pattern recognition. Thresholding is a simple but effective method to separate objects from the background. A commonly used method, the Otsu method, improves the image segmentation effect obviously. It's simpler and easier to implement. However, it fails if the histogram is unimodal or close to unimodal. Under studying the principle of the Otsu method, an improved threshold image segmentation algorithm based on the Otsu method is developed. Because the optical threshold should near the cross where the object and the background intersect, the probability of occurrence at the threshold value should divide into two parts. Its half belongs to object and half belongs to background. Then we apply a new weight to the Otsu method, this weight can make sure that the result threshold value will always reside at the valley of the two peaks or at the bottom rim of a single peak. Moreover, it ensures that both the variance of the object and the variance of the background keep away from the variance of the whole image. Comparing with the Otsu method, the improved method can get satisfactory results both for the image with histogram of bimodal and unimodal distributions. The experiments indicate that this segmentation algorithm has advantages of real time and certain anti-noise abilities, the target can be extracted more precisely. Therefore, the target recognition in the next step will be simple and reliable.

In image-forming laser locating system, irregular diffuse laser spot is usually generated by the factors as follow: orientation, intension and multi-path of the laser beam. A precise algorithm for laser spot locating based on Gradient Hough Transform is presented in this paper. First of all, the gradient gray image of the diffuse laser spot is produced by improved Sobel operators. Then the gradient image is segmented by certain gradient threshold. And then the center of the spot is localized by an improved Hough transform algorithm. The experiment shows that the invalid accumulation in Randomized Hough Transform (RTH) and the effects by orientation and intension are well resolved, and the centre of the laser spot can be located accurately by this algorithm.

Image segmentation is one of the most difficult and important steps in image analysis. Segmentation accuracy determines the eventual success or failure of computerized analysis procedures. The thresholding method is considered as one widely used method in image segmentation. It is considered by Kittle and Illingworth that the optimum threshold of segmentation can be calculated by the method of probability statistics, if the mixture density function composed of object and background is known and estimated. Suppose that the gray histogram of image obeys Poisson distribution, a new method of segmentation based on the maximum relationship principle of conditional distribution is proposed in this paper. The new method is compared with Ostu method and the minimum error method, the result shows that the proposed one is good and the performance is stable.

An effective scheme for stereoscopic video object segmentation was presented based on depth and edge information. Disparity estimation algorithm using region information was firstly proposed to get reliable depth information. In the proposed method, a new adaptive-size window approach was introduced to stereo matching in order to overcome problems with fixed-size window. Then, an improved watershed algorithm was exploited to achieve initial segmentation of depth image and coarse object areas were obtained. For each depth segments, edge detection was implemented to obtain accurate contours of object. Finally, watershed algorithm was utilized to segment object again. Using the extracted objects, an efficient motion tracking algorithm based on region was performed to track the video objects of the rest frames. The experimental results show that the proposed method can get good disparity map and segment video objects successfully.

The L-curve and its curvature are often applied to determine a suitable value of the regularization parameter when solving ill-conditioned linear systems of equations in super-resolution image reconstruction. However, the computation of the L-curve and its curvature is quite costly. In this paper both L-curve and its curvature can be computed fairly inexpensively by partial Arnoldi process applied to the matrix of the given linear system of equations in super-resolution image reconstruction. Through the Arnoldi process the techniques can generate orthogonal bases for the Krylov subspaces, which is a small and condensed Hessenberg matrix. The paper presents the simple solution in super-resolution image reconstruction by the Hessenberg matrix and presents the method for quickly computing L-curve and its curvature.

A closed-form maximum entropy image restoration method is proposed in this paper. First, the least square solution in the presence of measurement error is found by Weiner filter. Then a small perturbation from the solution is determined in the null space of the blur matrix by maximum entropy criterion. The closed-form least square solution and its perturbation are deserved which means no iterative, less complexity and rapid processing. Compared with Weiner filter, the method proposed is also a stabilizing algorithm, and more details can be deserved due to the introduced perturbation.

The images of square pattern, square superlattice pattern, and hexagonal pattern observed in dielectric barrier discharge are processed to get the exactitude binary images for further investigation. By using spatial correlation function, it is found that both of square and hexagonal patterns show perfect structures, and the square superlattice pattern exhibits an interleaving of two perfect squares with the same wavelength including large and small spots respectively. Furthermore, angular spectral distribution is used to study pattern development. It shows that, square and hexagonal patterns are formed with single modes respectively, while square superlattice pattern is governed by a three wave resonance.

The furnace flame images captured by the CCD are temperature fields inside the furnace. The flame images have some characteristics, such as noisy areas, low contrast and weak edges. The traditional edge detection methods are quick, such as Roberts, Canny etc, but the edges detected are often discontinuous and these methods are sensitive to noise. The level set is a new method to extract geometrical curve which can deal with the geometrical topological changes. It is effective on segmentation of non-rigid objects. But the level set method has some problems: firstly, the computation is very complex; secondly, if the initial curve is given arbitrarily, there would be an increase in iteration times and computational complexity. In view of these features, this paper puts forward a technique which uses fast mean shift and level set methods to segment flame images. Making use of fast mean shift to get the original contours for level set method so that it can decrease iteration times of the level set function. Experimental results indicate that the proposed algorithm can identify the combustion state of a flame more effectively and improve the evolution of the level set function. So it is of extensive practical use.

Multi-frame infrared image restoration is concerned with the improvement of imagery acquired in the presence of varying degradations. The degradations can arise from a variety of factors: common examples include undersampling of the image data, uncontrolled platform or scene motion, system aberrations and instabilities, noise characteristic of the infrared detector. In this paper, the mathematic models of infrared image blur and sampling and noise models are discussed. The multi-frame infrared image restoration problem is discussed, too. We show the origin and restoration infrared images which are used in the application of multi-frame infrared image restoration. By assessing subjectively and objectively to restoration images, we have verified this kind of model and the feasibility of the multi-frame infrared image restoration.

The problem of extracting continuous structures from images is a difficult issue in early pattern recognition and image processing. Tracking with contours in a filtering framework requires a dynamical model for prediction. Recently, Particle filter, is widely used because its multiple hypotheses and versatility within framework. However, the good choice of the propagation function is still its main problem. In this paper, an improved particle filter, EM-PF algorithm is proposed which using the EM (Expectation-Maximization) algorithm to learn the dynamical models. The EM algorithm can explicitly learn the parameters of the dynamical models from training sequences. The advantage of using the EM algorithm in particle filter is that it is capable of improve tracking contour by having accurate model parameters. Though the experiment results, we show how our EM-PF can be applied to produces more robust and accurate extracting.

At present, the large telescopes use segmented primary mirror. Segmentation and edge effects are closely related to system image quality. Generally there are four kinds of edge effects including gaps between segments, turned down/up segment edges, piston and random tip-tilt errors. This paper concentrates on a primary mirror structure with a central octagonal segment and eight marginal approximately annular-sector segments. We will give a Strehl Ratio formula that contains edge effect parameters. In addition, both the derivation and simulation are based on the assumption that edge effects exist uniquely.

The traditional interpolation could achieve a good correcting result when the resolution is almost homogeneous in the whole image. But the resolutions of the images captured by side-looking digital cameras are not uniform, and the resolutions of the image pixels should be taken into consideration. The imaging model of side-looking camera is founded, and the resolution distribution is given by analyzing the imaging equation of side-looking camera. And the resolution along x and y direction is calculated with the parameters of MADC(Multi-model Airborne Digital Camera). For the image processing for the side-looking camera, a new interpolation method is proposed based on the resolution distribution. The new interpolation equations are given for the linear and array CCD images.

A virtual reality flight procedure vision system is introduced in this paper. The digital flight map database is established based on the Geographic Information System (GIS) and high definitions satellite remote sensing photos. The flight approaching area database is established through computer 3D modeling system and GIS. The area texture is generated from the remote sensing photos and aerial photographs in various level of detail. According to the flight approaching procedure, the flight navigation information is linked to the database. The flight approaching area vision can be dynamic displayed according to the designed flight procedure. The flight approaching area images are rendered in 2 channels, one for left eye images and the others for right eye images. Through the polarized stereoscopic projection system, the pilots and aircrew can get the vivid 3D vision of the flight destination approaching area. Take the use of this system in pilots preflight preparation procedure, the aircrew can get more vivid information along the flight destination approaching area. This system can improve the aviator's self-confidence before he carries out the flight mission, accordingly, the flight safety is improved. This system is also useful in validate the visual flight procedure design, and it helps to the flight procedure design.

In the process of the multi-sensors target recognition fusion, focused on the problem that it is difficult to determine the reliability of each sensor and how the data measured by different sensors are fused, a multi-sensor target recognition fusion method based on fuzzy theory is proposed. The mutual supportability of multiple sensors is obtained from the correlation function. Then by the membership function, the reliability of information provide by each sensor is gained. Finally, the supposed fusion result of multi-sensors target recognition can be produced on the basis of fuzzy integration function. The method is simple computationally and can objectively reflect the reliability of each sensor and interrelationship between these sensors. By applying the method to the target recognition, the simulation experiment shows that it can identify the target accurately and is an effective and feasible multi-sensors target recognition fusion method.

A method of difference image to inspect real-time defects for cold rolled steel strip is proposed, which is based on subtract arithmetic of image. That is, shooting a scene at different time subtraction or image of the same scene at different waveband subtraction. This paper outlines a subtraction operation between the gathering images and the standard images. The standard image selection utilizes sequence extraction technique, which is to extract background image as a standard image from the multi-frame continuous real-time processing of images, and the standard image is self adaptive update. In the course of image defect inspection, we divided the difference image into small regions and inspected them respectively with the character of defects being remarkable in part image. Through the experiment analysis, conditions can be obtained to judge if any defects exist. Experiment on five typical defects (wrinkles, inclusion, weld, holes and serrated edges) were done. The results show that this method can meet the requirements of defect inspection and a higher rate of correct identification can be achieved.

Moderate Resolution Imaging Spectroradiometer (MODIS) is one of the most important sensors onboard the Terra and Aqua satellites and can be used in observing the air, the sea and the land quickly, timely and multi-temporally, which makes it an attractive and powerful tool for ecosystem pattern and process study. Carnegie-Ames-Stanford Approach (CASA) model is one of the most quick, convenient and accurate models to estimate the NPP (Net Primary Productivity) of vegetation. In this paper, the estimation and analysis of spatial and temporal distribution of vegetation NPP of the Haihe River Basin, the most man-interfered area in China, was made by using the CASA model based on the data of MODIS sensor onboard on the Terra satellites from the beginning of 2000 (the first year of delivering data) to the end of 2006. The result shows that the total NPP of the study area ranged from 1.25 × 10¹⁴ to 1.54 × 10¹⁴ gC/a during 2000 to 2006. The crop yields measured in the study area and the NPP value of the agriculture area from the CASA model were compared for testing the model's applicability in this study. The testing result shows that the NPP forecasted by using MODIS data and CASA model is acceptable. The NPP of the study area in all the years accumulated mainly from May to September, in which the maximum NPP appears in July and in August. The average NPP of the seven years of the main four vegetation types varied differently: values of two ripe crops a year or three ripe crops two years (519 gC/m²•a) > value of temperate deciduous shrub (428 gC/m²•a) > value of crop of one ripe a year and value of cold tolerating economic crop (293 gC/m²•a) > value of temperate grassland (195 gC/m²•a). From the results of the time series, it shows that the NPP varied irregularly with the time, but the NPP of all vegetation types in 2001 is lower than those in any other years. It relates to the drought climate in 2000 and 2001.

We propose a super-resolution resolution algorithm on the basis of maximum likelihood (ML) method and edge-orient diffusion. By using Hammerseley-Clifford theorem, an image field assumed to be a Markov random field is Gibbs distributed. An edge-orient diffusion function is introduced and employed in the Gibbs prior. According to Bayesian theorem, the solution to the maximum likelihood function is equal to that to maximum a posterior function. Therefore we incorporate ML with a prior distributed function. Experimental results illustrate that our method has a powerful super-resolution restoration performance. Compared with traditional ML method, our approach can not only obtain super-resolution images, but also eliminate noise artifacts effectively without smoothing edges.

A new similarity function for region based image fusion is proposed incorporating with Gabor filters and FCM clustering in this paper. First, the fuzzy c-means clustering algorithm (FCM) is used to segment the image in the feature space formed by multi-channel Gabor filters. Second, wavelet decomposition is performed on the source images, and then the weighting factors are constructed based on the local energy and the new similarity function defined by Gabor filters. Finally, the fused image is obtained by taking inverse wavelet transform. The performance of the image fusion method is evaluated using five criteria including root mean square error, peek-to-peek signal-to-noise ratio, entropy, cross entropy and mutual information. The evaluation results indicate that the proposed image fusion method is effective.

This article proposes an effective method for removing the thin cloud and mist in the remote sensing images using Mallat algorithm by analyzing the frequency distribution characteristics of the remote sensing images influenced by cloud and mist. Based on the characteristics of relatively lower frequency of the cloud and fog, relatively higher frequency of the scenes, and the multi-resolution of the wavelet function, we analyze the characteristics of the wavelet transformation in both the theory and the practical application, and conclude that the detail coefficient at the lower level of the wavelet represents relatively high-frequency of the image, the detail coefficient at its higher level represents the relatively low frequency band of the image. Like this, we can effectively strengthen the high-frequency components and weaken the low frequency components in the images, achieve goal for removing cloud and mist. The experimental results of the proposed algorithm are found to be satisfactory.

Ultrasonic inherent characteristic and underwater environment make underwater ultrasonic image have speckle noise, and image contrast is low. So need a method to reduce speckle noise and enhance image contrast at the same time. In certain applications, however, the removal of speckle and image enhancement may be two contradictory problem each other, so we difficult to gain an ideal image processing result. For this reason, considerations in characteristic of the underwater ultrasonic image, this paper provides a new speckle reduction and image enhancement anisotropic diffusion method based on wavelet technology. An anisotropic diffusion model has been established based on wavelet transform. We analyze the characteristic of the model and discuss the model's mechanism of action for removing speckle and enhance image edge of the underwater ultrasonic image. A compare experiment for real underwater ultrasonic image has been done using the method and other traditional methods. The experimental result indicates that the method proposed have strong speckle reduction and enhancement image ability. The purpose of removing speckle noise and enhancing edge at same time has been reached.

The image segmentation is often an important step in the analysis of images. In this paper, an image segmentation method based on cultural algorithms is presented. The method performs the image segmentation by selecting the optimal threshold values. The multi-threshold values are used. First of all, an entropy function corresponding to an image is defined. The optimal threshold values are obtained by making the entropy function reach the maximal value. Secondly, an algorithm based on the principle of cultural algorithms is presented for the computation of the optimal thresholds. The algorithm consists of three major components: a population space, a belief space, and a communication protocol that describes how knowledge is exchanged between the first two components. The designs and implementations of the three components are given in detail. The experimental results show that the segmentation method proposed in this paper can obtain the near optimal threshold for image segmentation.

The paper introduced a real-time pattern recognition method for animal posture. It based on digital image processing, BP neural network and DSP techniques. 2D movement projective image of mice was captured and processed. Some features were extracted, such as morphological parameters and motion parameters. A BP neural network was set to distinguish the posture with these parameters. According to the experiments, the recognition rate of special posture reached 90.5%. The method can be applied in pharmacology experiments and other fields for classification of animal posture.

The advanced multi scale Retinex (AMSR) algorithm was put forward: after enhanced by multi scale Retinex the image of the truncate interval that it was a range near the mean brightness μ in k multiple standard deviation σ had been elected and stretched. According to the ideas of D. J. Jobson et al. if the production of the brightness multiply with the contrast degree for the stretched image was in the range of [6 000, 10 000], the stretched image has high quality. As the same time, the experiments proved the conclusion: the highest quality image in enhanced images for an underexposure color image was that it selected about k=1 multiple standard deviation range near the mean brightness and stretched after enhanced by multi scale Retinex algorithm.

Current research on computer vision often needs specific techniques for particular problems. Little use has been made of high-level aspects of computer vision, such as three-dimensional (3D) object recognition, that are appropriate for large classes of problems and situations. In particular, high-level vision often focuses mainly on the extraction of symbolic descriptions, and pays little attention to the speed of processing. In order to extract and recognize target intelligently and rapidly, in this paper we developed a new 3D target recognition method based on inherent feature of objects in which cuboid was taken as model. On the basis of analysis cuboid nature contour and greyhound distributing characteristics, overall fuzzy evaluating technique was utilized to recognize and segment the target. Then Hough transform was used to extract and match model's main edges, we reconstruct aim edges by stereo technology in the end. There are three major contributions in this paper. Firstly, the corresponding relations between the parameters of cuboid model's straight edges lines in an image field and in the transform field were summed up. By those, the aimless computations and searches in Hough transform processing can be reduced greatly and the efficiency is improved. Secondly, as the priori knowledge about cuboids contour's geometry character known already, the intersections of the component extracted edges are taken, and assess the geometry of candidate edges matches based on the intersections, rather than the extracted edges. Therefore the outlines are enhanced and the noise is depressed. Finally, a 3-D target recognition method is proposed. Compared with other recognition methods, this new method has a quick response time and can be achieved with high-level computer vision. The method present here can be used widely in vision-guide techniques to strengthen its intelligence and generalization, which can also play an important role in object tracking, port AGV, robots fields. The results of simulation experiments and theory analyzing demonstrate that the proposed method could suppress noise effectively, extracted target edges robustly, and achieve the real time need. Theory analysis and experiment shows the method is reasonable and efficient.

This research aimed for development multi-spectral imaging technique for green tea categories discrimination based on texture analysis. Three key wavelengths of 550, 650 and 800 nm were implemented in a common-aperture multi-spectral charged coupled device camera, and images were acquired for 190 unique images in a four different kinds of green tea data set. An image data set consisting of 15 texture features for each image was generated based on texture analysis techniques including grey level co-occurrence method (GLCM) and texture filtering. For optimization the texture features, 5 features that weren't correlated with the category of tea were eliminated. Unsupervised cluster analysis was conducted using the optimized texture features based on principal component analysis. The cluster analysis showed that the four kinds of green tea could be separated in the first two principal components space, however there was overlapping phenomenon among the different kinds of green tea. To enhance the performance of discrimination, least squares support vector machine (LSSVM) classifier was developed based on the optimized texture features. The excellent discrimination performance for sample in prediction set was obtained with 100%, 100%, 75% and 100% for four kinds of green tea respectively. It can be concluded that texture discrimination of green tea categories based on multi-spectral image technology is feasible.

A kind of real time image processing system for target recognition and track is discussed. The system adopts master-slave DSP embedded structure in hardware, combines general DSP with special image processing circuit, and video data collection is controlled by program. This real time image processing system is designed with the high speed DSP TMS320C6201 while FPGA for image pretreatment. This paper emphasizes on introduction of real time image processing system based on DSP from the view of hardware. Experiments have been done by the system and the results are satisfactory.

We propose a new architecture of optical encryption technique using the fractional Fourier transform with three statistically independent random phase masks. Compared with the existing double-phase encoding method in the fractional Fourier-domain, the proposed extra phase mask in the last fractional Fourier domain makes the architecture symmetrical, and additive processing to the encrypted image can be turned into complex stationary white noise after decryption, and enlarge the key space without any degradation of its robustness to blind decryption. This property can be utilized to improve the quality of the recover image. Simulation results have verified the validity.

This paper proposes a new image super-resolution restoration algorithm. The development of the algorithm is based on the improvement of the classical projection on convex set (POCS) algorithm and wavelet fusion to restore a super-resolution image from a series of low resolution (LR) images. At first, the POCS iteration is used to restore high-resolution (HR) image from every LR image. Then several different rules are chosen to fuse HR images in wavelet domain, and a HR image is reconstructed by inverse wavelet transform. The reconstructed image is evaluated by entropy, cross entropy, definition and the peak signal-noise ratio. The experimental results of the processed CT images showed that this method can improve the ability of fusing different image information, and the texture of the image is more prominent, the image quality is higher.

In order to achieve a spatial resolution comparable to a monolithic telescope of the same diameter, the segmented mirror surface must be phased with an accuracy about λ/20 wavefront rms, which is comparable with the accuracy of the segment figure. During the phasing procedure, piston, tip-tilt are controlled to achieve the best image quality of the telescope. The residual rms wavefront error of whole aperture taken as the merit function reaches minimum analytically when the least-square-fit plane of tested segment meets together with that of reference mirror, which means that "phasing" is a statistical mood of the whole aperture, not to match some small areas of the segments. The impact of segment aberration on the whole aperture residual rms wavefront error is simulated using the whole segment aperture sensing and edge sensing respectively. By the analytical and numerical simulation, it is demonstrated that the whole segment aperture phase sensing can make a better image quality than the edge sensing.

The major studied in this dissertation is the non-destructive testing method of silkworm cocoon's quality, based on the digital image processing and photoelectricity technology. Through the images collection and the data analysis, procession and calculation of the tested silkworm cocoons with the non-destructive testing technology, internet applications automatically reckon all items of the classification indexes. Finally we can conclude the classification result and the purchase price of the silkworm cocoons. According to the domestic classification standard of the silkworm cocoons, the author investigates various testing methods of silkworm cocoons which are used or have been explored at present, and devices a non-destructive testing scheme of the silkworm cocoons based on the digital image processing and photoelectricity technology. They are dissertated about the project design of the experiment. The precisions of all the implements are demonstrated. I establish Manifold mathematic models, compare them with each other and analyze the precision with technology of databank to get the best mathematic model to figure out the weight of the dried silkworm cocoon shells. The classification methods of all the complementary items are designed well and truly. The testing method has less error and reaches an advanced level of the present domestic non-destructive testing technology of the silkworm cocoons.

For the interference fringes' photodiode detection, its SNR can be improved by two ways to a great extent, one is space arrangement of photodiodes, another is that photodiode operates without external reverse voltage. Despite application of reverse voltage to photodiode is very useful in improving linearity and high frequency, it has the accompanying disadvantage of increasing dark current and noise levels along with the danger of damaging the device by excessive applied reverse voltage. In order to keep the high SNR and high frequency response of photodiode, we analyze the relation between movement frequency and photoelectric conversion of interference fringes when space arrangement of photodiodes are designed as a kind of high SNR detection method proposed in our previous article and photodiode operates in no reverse voltage state. On basis of the analysis, a combinative amplifier of DC (direct current) and AC (alternative current) is present to amplify the photoelectric difference signal. The input and feedback resistance of the amplifier are all replaced by the resistance and capacitance in parallel connection. When this amplifier is compared to the common one, it can greatly broaden the frequency response range of fringe signal and ensure the instrument can work normally at low and high frequency.

Computed tomography (CT) is a non-invasive imaging technique, which is widely applied in medicine for diagnosis and surgical planning, and in industry for non-destructive testing (NDT) and non-destructive evaluation (NDE). So, it is significant for college students to understand the fundamental of CT. In this work, A CT imaging system named CD-50BG with 50mm field-of-view has been developed for experimental teaching at colleges. With the translate-rotate scanning mode, the system makes use of a 7.4×10⁸Bq (20mCi) activity ¹³⁷Cs radioactive source which is held in a tungsten alloy to shield the radiation and guarantee no harm to human body, and a single plastic scintillator + photomultitude detector which is convenient for counting because of its short-time brightness and good single pulse. At same time, an image processing software with the functions of reconstruction, image processing and 3D visualization has also been developed to process the 16 bits acquired data. The reconstruction time for a 128×128 image is less than 0.1 second. High quality images with 0.8mm spatial resolution and 2% contrast sensitivity can be obtained. So far in China, more than ten institutions of higher education, including Tsinghua University and Peking University, have already applied the system for elementary teaching.

To achieve the landing site locating, the proof-of-concept strategy demands a reference image captured beforehand. This strategy adopts the way of wide-baseline matching between images from the airborne camera and the reference image. First of all, a proof-of-concept strategy of vision-aided landing area locating of unmanned aerial vehicle is presented. Then, there are mainly two algorithms under discussion. One is landing site finding using SIFT based wide-baseline match, and the experiment results show that is qualified to finding landing site. The other is landing site locating by camera position measurement based on the feature point with known world coordinates. Assuming that the airborne camera has been calibrated off-line, with known intrinsic camera parameters and world coordinates of at least 4 feature points, the position of camera can be roughly estimated. The SIFT feature points of reference image have been selected and their position in world coordination have been marked before wide-baseline matching, so, when matching having been put into practice, world coordinates of matched points in each image captured by airborne camera can be gotten directly. Assuming when landing site locating task is carrying out, the UAV is fairly high, so the landing site could be regarded as planar. And due to the camera is fixed to UAV, the relative position of UAV to the landing site can be measured according the algorithm in this paper, that is to say, providing that at least one image of UAV landing site is offered, the landing site can be successfully located.

Aerosol model is a major obstacle for passive remote sensing of aerosol properties. For the continent or urban aerosol model does not fit the needs for more accurate retrieval, and the user defined aerosol model may be more near to the realistic condition in Taihu region. A method has been developed for retrieving the aerosol model and optical properties including polarization, i.e., scattering coefficient, asymmetry factor, single scattering albedo, scattering phase function and polar phase function. In Lake Taihu, the study area of our two combined remote sensing observations in the winter and summer 2006, we got water surface spectral data of ASD (Analytical Spectral Devices) by the ship, atmospheric data of CE318 sunphotometer on the shore, and the MODIS (Moderate Resolution Imaging Spcectroradiometer) image data on the TERRA Satellite. By using the nearly synchronous measurements of the data from surface spectrum and the sunphotometer with the image, and by use of the radiative transfer model 6S(Second Simulation of a Satellite Signal in the Solar Spectrum), varying the components of the aerosol type, a LUT (look up table) is made for the radiance on the satellite. When the total relative error of the new defined parameter for relative error is getting to the least, the aerosol type will be decided. Then, based on the determination of aerosol model, the atmospheric aerosol properties over Lake Taihu have been computed by using Mie theory and analyzed with the typical continental and urban aerosol models available in 6S. These results show that the user-defined aerosol model is a mix model of continent and urban which corresponds with previous studies. Moreover, they may be useful for resolving the vector RTE (radiative transfer equation). In this paper we tried to provide a method with the combination of remote sensing data to obtain the optical properties of atmospheric aerosol over inland water in different seasons. We respect it will be helpful for accurate atmospheric correction in the future.

With the advantages of non-invasiveness, uniqueness, stability and low false recognition rate, iris recognition has been successfully applied in many fields. Up to now, most of the iris recognition systems are based on PC. However, a PC is not portable and it needs more power. In this paper, we proposed an embedded iris recognition system based on ARM. Considering the requirements of iris image acquisition and recognition algorithm, we analyzed the design method of the iris image acquisition module, designed the ARM processing module and its peripherals, studied the Linux platform and the recognition algorithm based on this platform, finally actualized the design method of ARM-based iris imaging and recognition system. Experimental results show that the ARM platform we used is fast enough to run the iris recognition algorithm, and the data stream can flow smoothly between the camera and the ARM chip based on the embedded Linux system. It's an effective method of using ARM to actualize portable embedded iris recognition system.

Content-based video retrieval is a hot topic on video research subject today. To avoid the drawback of traditional retrieval methods which depend on only one feature such as color, texture, shape, movement, this paper put forward a new method which uses both color layout descriptor and homogeneous texture in description MPEG-7. The method extracts color layout descriptor and homogeneous texture descriptor, afterward these two descriptors are normalized. According to the result of retrieving video feature database, it can be concluded that using both features can increase the accuracy and the result suits more to the human vision need.

The millimeter-wave images have low resolution and heavy noise. Hence it is hard to detect the edges in such images. A detection scheme based on curvelet transform is proposed. The idea is to suppress noise through Wrapping algorithm of curvelet transform at first, then determine gradient amplitude of pixels. At last the non-maximum and double threshold method are used to obtain the edges. The experiments show that clear edges of human and object images in millimeter-wave images can be detected efficiently, and the scheme implements fast.

The report provides the principle and advances of range-gated Underwater Laser Imaging Technology. A prototype Underwater Laser Imaging System has been designed and developed. The range-gated technique is one of the effective methods to decrease the backward scattering of the water media. This report provides the study and implementation for range-gated Underwater Laser Imaging System (ULIS). A careful analysis of the novel Underwater Laser Imaging System (ULIS) shows that a range-gated system will have a significantly improved viewing performance over conventional systems. The system uses a 2-kHz diode-pumped frequency doubled Nd:YAG laser as an illumination source. The light is collected by a 5-cm-diam zoom lens. The detector is a gated image CCD detector with a 10-ns gate. In the strongly scattering waters, this system has a range of from three to five times that of a conventional camera with floodlights. By the use of the lower attenuation window for the water media in the visible light spectrum, pulse laser with the blue-green wavelength has been adopted as an illumination source and CCD as the receiver. The range-gated underwater laser imaging system has been successfully fulfilled. The system can decrease the influence of backward scattering and at the same time the system can improve target luminance. The detail schematic principle and realization scheme have been analyzed. It has been found that the range-gated underwater laser imaging system could improve the imaging quality and the imaging distance by conveniently adjusting the pulse delay and the pulse width. It has been become one of the key techniques in underwater laser imaging research field.

Cone-beam computed tomography (CBCT) is a technique for imaging cross-sections of an object using a series of X-ray measurements taken from different angles around the object. It has been widely applied in diagnostic medicine and industrial non-destructive testing. Traditional CT reconstructions are limited by many kinds of artifacts, and they give dissatisfactory image. To reduce image noise and artifacts, we propose a statistical iterative approach for cone-beam CT reconstruction. First the theory of maximum likelihood estimation is extended to X-ray scan, and an expectation-maximization (EM) formula is deduced for direct reconstruction of circular orbit cone-beam CT. Then the EM formula is implemented in cone-beam geometry for artifact reduction. EM algorithm is a feasible iterative method, which is based on the statistical properties of Poisson distribution. It can provide good quality reconstructions after a few iterations for cone-beam CT. In the end, experimental results with computer simulated data and real CT data are presented to verify our method is effective.

Terahertz electromagnetic radiation has been generated successfully with semi-insulating Gallium Arsenide(GaAs) photoconductive switches in linear mode, but haven't detected any radiation under nonlinear mode of photoconductive antenna. Nevertheless, compared with linear mode, the photoconductive switches in the mode of nonlinear has a sharper current rise. In this condition, the possibility of generating terahertz wave via the mode of nonlinear of photoconductive switches has discussed. We based on Drude-Lorentz theory and optically activated domain theory of nonlinear mode of photoconductive antenna, instantaneous response of photoexcited carriers in the mode of nonlinear is analyzed, and carrier dynamic characteristic connected with terahertz radiation in semi-insulating GaAs is discussed. The amplitude of THz radiation originating from impact ionization and acceleration in domain electric field is calculated. According to theory of far field radiation of terahertz wave, it can arrived at that the electric field intensity of far field of THz wave has a direct proportion to the derivative of current density in photoconductive antenna. For reason given above, the intensity of single THz wave with single triggering under the mode of nonlinear stronger than intensity of linear mode can be derived.

Multi-channel scanning radiometer, on boarding FY-2 geostationary meteorological satellite, plays a key role in remote sensing because of its wide field of view and continuous multi-spectral images acquirements. It is significant to evaluate image quality after performance parameters of the imaging system are validated. Several methods of evaluating imaging quality are discussed. Of these methods, the most fundamental is the MTF. The MTF of photoelectric scanning remote instrument, in the scanning direction, is the multiplication of optics transfer function (OTF), detector transfer function (DTF) and electronics transfer function (ETF). For image motion compensation, moving speed of scanning mirror should be considered. The optical MTF measurement is performed in both the EAST/WEST and NORTH/SOUTH direction, whose values are used for alignment purposes and are used to determine the general health of the instrument during integration and testing. Imaging systems cannot perfectly reproduce what they see and end up "blurring" the image. Many parts of the imaging system can cause blurring. Among these are the optical elements, the sampling of the detector itself, post-processing, or the earth's atmosphere for systems that image through it. Through theory calculation and actual measurement, it is proved that DTF and ETF are the main factors of system MTF and the imaging quality can satisfy the requirement of instrument design.

Laser ranging is to measure the distance of a target with laser beam. Having the advantages of narrow beam and strong resistance to electromagnetic interference, it has been developed and researched by many countries. Because of the complexity of backgrounds and targets, it is necessary to research the related laser echo characteristics. The power, waveform and other identities of laser pulse echo from targets are analyzed based on the experimental results of the laser ranging system. Valuable conclusions were drawn for the laser echo characteristics for different materials.

The composition of modes in double-clad multimode active fiber is studied. The bending influence on numerical apertures of fibers is analyzed by means of geometrical optics. When the optical fiber is bent, different modes have different bending losses, A numerical simulation was used to calculate the bending losses of LP₀₁ and LP₁₁ in different bending radii. Optimum bending radius was chosen to inhibit the high-order modes so that the fundamental mode signal amplification was assured. A 10 meters long Yb³⁺-doped double-cladding fiber with core diameter of 30μm and core numerical aperture of 0.06 was used to verify our theoretical calculation. The fiber amplifier was pumped by a diode laser array emitting at 976nm. A non-planar ring oscillator with cw single frequency output at 1064 nm was used as signal source. By coiling the double-cladding fiber, the amplified signal of fundamental mode was obtained. When the pump power was 49.6 W and the signal power was 200 mW, 16.1W of single frequency output power was obtained, the amplification factor was 80.5. The spectra of output signals before and after amplification were measured with Fabry-Perot interferometer. Fundamental mode output is verified.

Three different chemometric methods were performed for the determination of sugar content of cola soft drinks using visible and near infrared spectroscopy (Vis/NIRS). Four varieties of colas were prepared and 180 samples (45 samples for each variety) were selected for the calibration set, while 60 samples (15 samples for each variety) for the validation set. The smoothing way of Savitzky-Golay, standard normal variate (SNV) and Savitzky-Golay first derivative transformation were applied for the pre-processing of spectral data. The first eleven principal components (PCs) extracted by partial least squares (PLS) analysis were employed as the inputs of BP neural network (BPNN) and least squares-support vector machine (LS-SVM) model. Then the BPNN model with the optimal structural parameters and LS-SVM model with radial basis function (RBF) kernel were applied to build the regression model with a comparison of PLS regression. The correlation coefficient (r), root mean square error of prediction (RMSEP) and bias for prediction were 0.971, 1.259 and -0.335 for PLS, 0.986, 0.763, and -0.042 for BPNN, while 0.978, 0.995 and -0.227 for LS-SVM, respectively. All the three methods supplied a high and satisfying precision. The results indicated that Vis/NIR spectroscopy combined with chemometric methods could be utilized as a high precision way for the determination of sugar content of cola soft drinks.

For the measuring demand of surface coating quality, and to keep the surface to be measured from being damaged, laser is adopted as a optical probe. When contactless measurement of the object surface coating uniformity is carried out, a laser beam illuminates the object surface to be tested. By analysis of the reflected light signal from the random surface, relation between the intensity of reflected light and surface coating uniformity is obtained. To decrease the error caused by the dithering of source, laser source is fitted at a certain height above the surface to be measured. By moving the optical table, light spot scans over the object surface. Real-time statistic of the data collected is made by computer, and then deviation report of surface coating uniformity is acquired. Experiment results prove that the measurement system is simple, stable and reliable, with high speed of response and good interference killing feature. Measurement accuracy of the system achieves 5μm, and error of measurement value is less than ±1% within an operation scope from 1 μm to 50 μm. The method has the advantage of simple structure and intuitionistic principle, and it can detect online conveniently.

The design of segmented mirror is an effective way to increase the resolution of space optical remote sensor. We analyze the gaps and piston that affecting the system imaging performance. The corresponding Point Spread Function (PSF), modulation transfer functions (MTF) and SR (Strehl Ratio) are presented. The General Image Quality Equation (GIQE) is used to Linking NIIRS to those effects to provide engineering the criteria for system performance. We show an example of how NIIRS is applied to measure the influence of these effects on system performance. In the example, the Δ NIIRS is computed by GIQE. Δ NIIRS is performed to measure small loss in image quality. When the Δ NIIRS is more than 0.2, the loss of image quality are non-negligible. The experiment show the NIIRS can provide a simple, yet powerful, tool for assessing and communicating image quality and sensor system requirements.

It is very important to detect the protein and fat content in milk powder fast and non-destructively. Near-infrared (NIR) and mid-infrared(MIR) spectroscopy techniques have been compared and evaluated for the determination of the protein and fat content in milk powder with the use of Least-squares support vector machines (LS-SVM). LS-SVM models have been developed by using both NIR and MIR spectra. Both methods have shown good correlations between infrared transmission values and two nutrition contents. MIRS provided better prediction performance over NIRS. It is concluded that infrared spectroscopy technique can quantify of the protein and fat content in milk powder fast and nondestructively. The process is simple and easy to operate than chemistry methods. The results can be beneficial for designing a simple and non-destructive instrument with MIRS or NIRS spectral sensor for the determination of the protein fat content in milk powder.

As a rapid and non-destructive methodology, near infrared spectroscopy technique has been paid much attention recently. This paper presents an automatic recognition scheme for the fine wool fiber and cashmere fiber by Vis/NIR spectroscopy technique, aim at the characteristics of Vis/NIR spectra on cashmere and fine wool. One mixed algorithm was presented to discriminate cashmere and fine wool with principal component analysis (PCA) and Artificial Neural Network (ANN). Preliminary qualitative analysis model has been built: We adopt Vis/NIRS spectroscopy diffuse techniques to collect the spectral data of cashmere and fine wool, two kinds of data pretreatment methods were applied: the standard normal variate (SNV) was used as scatter correction. Savitzky-Golay with the segment size 3 was used as the smoothing way to decrease the noise processed. Followed the pretreatment, spectral data were processed using principal component analysis, 6 principal components (PCs) were selected based on the reliabilities of PCs of 99.8%, the scores of these 6 PCs would be taken as the input of the three-layer back-propagation (BP) artificial neural network (BP-ANN). Trained the BP-ANN with samples in calibration collection and predicted the samples in prediction collection. Experiments demonstrate that the system works quickly and effectively, and has remarkable advantages in comparison with the previous systems, The result indicted a model had been built to discriminate cashmere from fine wool using Vis/NIR spectra method combined with PCA-BP technology. The model works well, which indicates that this kind of approach is effective and promising, can raise resolution of cashmere and fine wool.

The principle of parallel light projection method for diameter measurement is to image the workpiece on the photosensitive units of CCD by parallel light. The object size can be obtained according to the shadow area projected on CCD. Considering the limitation of the lens' caliber and the effective length of CCD, the single light path system is incapable to measure a large size. Hence the double light paths method is presented. The left and right edges of the workpiece are respectively imaged on two CCDs. The diameter is achieved by calculating the length of the shadow field on two CCDs, plus to the distance between two parallel light paths. In order to make the object illuminance uniform, and to diminish the influence of optical errors on the measurement accuracy, the Kohler telecentric illumination system and the object plane telecentric imaging system are applied. The parallel light projection method with double light paths can amplify the measurement range while ensuring the measurement accuracy. The optical measurement theory, the parameter design and adjustment of the system will be introduced in detail.

All-optical wavelength conversion plays a major role in providing the wavelength flexibility in optical communication networks. All-optical wavelength converters (AOWCs) based on cross-gain modulation (XGM) and cross-phase modulation (XPM) in semiconductor optical amplifiers (SOAs) have attracted considerable research interest. In this paper, we propose a novel scheme for cascaded wavelength conversion based on cross-gain modulation and cross-phase modulation in SOAs. The wavelength conversion operation in the proposed scheme includes two stages, that is, XGM in the first stage followed by the stage of XPM, and thus is expected to have a high ER and a large input power dynamic range simultaneously.

Electronic Rolling Shutter (ERS) was usually considered to be the shortcoming of low-cost and low-power CMOS image sensors, because ERS will cause distorted and/or blurred images when the target is moving. In this paper, we propose a new method to measure roller's velocity by using ERS's distortion as a response of moving target rather than a drawback. In the presented method, dimension expanding method is improved to generate a repeated isosceles triangle pattern as the reference object. 2D velocity is simultaneously calculated from the distorted isosceles triangle pattern. Some experimental results are given to clarify the feasibility of the proposed method.

A novel hybrid-integrated 3D acceleration geophone, based on Mach-Zehnder interferometer, has been developed. Mach-Zehnder waveguide interferometer based on photoelastic effect of LiNbO₃ waveguide was researched and its optimization structure design was accomplished The Mach-Zehnder waveguide interferometer is realized by symmetrical double Y-branching waveguide and three polarizers, which are monolithically integrated on an X-cut, Y-propagating LiNbO₃ substrate. The simulation results performed in FD-BPM agrees with the theoretical analysis wel1. The Mach-Zehnder interferometer is fabricated successfully and the lighting test is proceeding to check the quality of the Mach-Zehnder interferometer Experimental results matches well with the theoretical prediction.

Based on the near-infrared absorption theory, a new carbon monoxide concentration measurement system is devised. The near-infrared absorption line of carbon monoxide is chosen as the measured spectrum and a distributed feed-back laser diode (DFBLD) is used as the light source. The wavelength of the light source is locked at the carbon monoxide gas absorptive peak which is chosen at 1.567μm by light source modulation technology, and the signals are processed by lock-in amplifier. The ratio of the second-harmonic to fundamental signal is used as the output of the system to eliminate optical interference. Optical fiber is used as transmission medium, instead of traditional complex optical system, so the structure of the system would become miniaturization. Proved by examinations, high measurement sensitivity of carbon monoxide gas can be achieved in this study.

The morphology and mechanism of lithium niobate (LiNbO₃) crystal ablated by a femtosecond laser pulse have been investigated. The ablated spots have been examined by means of scanning electron microscopy (SEM) and atom force microscopy (AFM). Sub-diffraction limit spots in LiNbO₃ crystal ablated by femtosecond laser have been obtained. The diameter of ablation spot is 400 nm and 800 nm when 170 nJ single pulse and 100 nJ, 17 pulses were used, respectively. There were about 200nm periodic ripples on the bottom surfaces of the spots at low laser energy. The result shows that the sub-diffraction limit structures may be formed by the multi-photon excitation, and the femtosecond laser ablation is an innovative tool for manufacturing LiNbO₃-based optical devices.

Micro-vibration or displacement measurement is very important in modern industrial fields, especially for the micro electrical and mechanical system (MEMS). An optical fiber sensor was proposed to measure the micro-vibration or displacement based on a novel fiber Bragg grating pair, in which one of the gratings is written in the fiber cladding and the other is written in the fiber core, however the positions of the two grating are staggered. Because the reflection spectrum of this structural grating is much sharper than that of ordinary grating, the special reflection spectrum of the fiber Bragg grating pair is used, resulting in a high measurement sensitivity. Theoretical analysis of the fiber Bragg grating pair and micro-vibration or displacement measurement principle with the experimental setup (Micro-vibration is measured by this grating pair, while this grating pair is demodulated by ordinary grating) are introduced in this paper.

This measuring machine is specially designed to solve many parameters of crankshaft for some factory. According to the laser slot scanning principle, non-contact measuring of two axes' parallel degree for crankshaft is realized. The machine use laser technique, photoelectric measuring technique, fine mechanical technique, modern photoelectric sensor technique, electronics technique and computer-technique. And the difficulty which include complex shape, large weight, large size, high request of the measured parameter precision and difficult measuring, is solved.

Based on the Kirchhoff's Law, a practical dual-wavelength fiber-optic colorimeter, with the optimal work wavelength centered at 2.1 μm and 2.3 μm is presented. The effect of the emissivity on the precision of the measured temperature has been explored under various circumstances (i.e. temperature, wavelength) and for different materials. In addition, by fitting several typical material emissivity-temperature dependencies curves, the influence of the irradiation (radiant flux originating from the surroundings) and the surface reflected radiation on the temperature accuracy is studied. The results show that the calibration of the measured temperature for reflected radiant energy is necessary especially in low target temperature or low target emissivity, and the temperature accuracy is suitable for requirements in the range of 400-1200K.

Based on the solution of the diffusion equation for photon density wave in rectangular coordinates, a model of a three dimensional cubic tissue with an abnormity for the inverse problem is established. By use of the analytic expression of the photon density and the least squares (LS) method, the objective function for solving the inverse problem is constructed. Then, an adaptive Harmony Search (HS) is presented and employed to seek the possible solution. In the improved algorithm, the harmony memory considering rate (HMCR) and the pitch adjusting rate (PAR) are adjusted in the searching process to obtain more accurate solution of the detection problem. The numerical experiments for comparisons of the conventional HS and the adaptive HS are given. The results show that the improved algorithm can be used to localize the abnormity embedded in the tissue with a good agreement with the presupposed location parameters. It can be concluded that AHS algorithm is more effective than the conventional HS for the inverse problems similar to the detection using photon density wave.

The paper introduces the design idea, basic structure, testing conditions, and using methods of the portable quality monitor of TCM with near infrared diffuse reflectance. Characterized by small cubage, light weight, and low cost, etc., the spectrometer is a convenient tool for the on-the-spot quality control of the Traditional Chinese Medicine (TCM).