Due to the supervised model data labeling is more complex and cumbersome, this paper adopts unsupervised Auto- Encoder network for defect detection. In addition, for the problem that the convolutional neural network does not have high accuracy in detecting defects on metal surfaces, a defect detection method based on the PoolFormer reconstruction model is proposed. Firstly, according to the type of defects, the defect-free image with superimposed similar morphological noise is used to train the reconstruction model, so that the model has the ability to repair; secondly, the defective image is input into the model for repairing; finally, the residuals between the reconstructed image and the defective image are calculated, and the defect detection and localization can be realized. The experimental results show that the method is able to detect many types of metal surface defects, and the detection accuracy reach more than 98%.
In this paper, an uncalibrated photometric stereo network model is proposed. Traditional photometric stereo has precise requirements on light source and object surface reflectance, which greatly limits the usability of photometric stereo, and the ideal network model should be able to handle arbitrary objects under uncalibrated light sources to complete the reconstruction. To solve this problem, we propose an uncalibrated photometric stereo neural network, First, an arbitrary number of images are input to the first-stage neural network system to estimate the light source information and complete the calibration to eliminate the dependence on the light source information. Then the calibration result is input to the second-stage neural network together with the image, and the multi-scale and attention mechanism is used to obtain the object surface texture information to achieve the surface normal reconstruction of the object. Our network (MAPS-Net) is compared with other uncalibrated photometric stereo methods on the DiLiGenT benchmark, and the average MAE achieves an excellent performance of 9.20.
Accurate calibration of image acquisition and projection device in fringe projection measurement system is the basis of system work, and the calibration accuracy of the parameters directly affects the measurement accuracy of the system. An improved projector calibration method is proposed. Using a high-precision circular calibration plate, the camera is calibrated by Zhang Zhengyou's calibration method and nonlinear optimization; using the vertical and horizontal two sets of grating images projected by the projector, the corresponding relationship between them and the camera image is established, and then the center of the mark point in the projection is obtained. Then use the camera model with radial distortion to calibrate the projector; analyze the phase error when the projector is calibrated, and use the improved multifrequency fringe phase shift method to reduce the error; finally, through the calibration and the three-dimensional measurement experiment verifies the method in this paper. The experimental results show that the calibration method in this paper is simple in operation, high in precision and wide in application.
In order to investigate the penetration depth of silver nanoparticles (Ag NPs) inside the skin, porcine ears treated with Ag NPs are measured by two-photon tomography with a fluorescence lifetime imaging microscopy (TPT-FLIM) technique, confocal Raman microscopy (CRM), and surface-enhanced Raman scattering (SERS) microscopy. Ag NPs are coated with poly-N-vinylpyrrolidone and dispersed in pure water solutions. After the application of Ag NPs, porcine ears are stored in the incubator for 24 h at a temperature of 37°C. The TPT-FLIM measurement results show a dramatic decrease of the Ag NPs’ signal intensity from the skin surface to a depth of 4 μm. Below 4 μm, the Ag NPs’ signal continues to decline, having completely disappeared at 12 to 14 μm depth. CRM shows that the penetration depth of Ag NPs is 11.1±2.1 μm. The penetration depth measured with a highly sensitive SERS microscopy reaches 15.6±8.3 μm. Several results obtained with SERS show that the penetration depth of Ag NPs can exceed the stratum corneum (SC) thickness, which can be explained by both penetration of trace amounts of Ag NPs through the SC barrier and by the measurements inside the hair follicle, which cannot be excluded in the experiment.
Aspheric mould design includes the top-down design and reversal design. In this paper, a new framework of reversal
design is proposed combining with cloud supported collaborative design (CSCD) based on aspheric measurement. The
framework is a kind of collaborative platform, which is composed of eight modules, including the computerized aspheric precision measurement module (CAPM), computer-aided optical design of aspheric lens system (CAOD),
computer-aided design of lens mould (CADLM), FEM(finite element method) simulation of lens molding module
(FEMLM), computer-aided manufacture of lens and moulds (CAMLM), measurement data analysis module (MDAM),
optical product lifecycle management module (OPLM) and cloud computing network module (CCNM). In this framework, the remote clients send an improved requirement or fabrication demand about optical lens system through CCNM, which transfers this signal to OPLM. In OPLM, one main server is in charge of the task distribution and
collaborative work of other six modules. The first measurement data of aspheric lens are produced by clients or our proposed platform CAPM, then are sent to CAOD for optimization and the electronic drawings of lens moulds are generated in CADLM module. According the design drawings, the FEMLM could give the lens-molding simulation
parameters through FEM software. The simulation data are used for the second design of moulds in CADLM module. In this case, the moulds could be fabricated in CAMLM by ultra-precision machine, and the aspheric lens could be also
produced by lens-molding machine in CAMLM. At last, the final shape of aspheric lens could be measured in CAPM and the data analysis could be conducted in MDAM module. Through the proposed framework, all the work described
above could be performed coordinately. And the optimum design data of lens mould could be realized and saved, then shared by all the work team.
A new design and fabrication of flat optical lens by use of 90-degree volume holographic grating recording geometry is presented. It is recorded by the interference of a plane wave and a spherical wave in a Fe-doped LiNbO3. The flat optical lens can laterally transfer and focus the plane wave which perpendicularly incidents on the planar crystal surface. The optimal switching from recording to thermal fixing is taken into consideration in order to obtain the nonvolatile hologram with maximum fixing efficiency. The flat optical lens with different recording and reconstructing wavelength has also been discussed. The flat optical lens has the advantages of light weight, laterally transferring and focusing, small duty ratio, and easy for the microstructure integration. The testing results measured verify that the flat optical lens can successfully be used for the free space optical communication.
Lens molding has become the promising technique to conduct mass produce of aspheric glass lens. It overcomes some
disadvantages of traditional grinding or turning methods, such as high cost, low efficiency, unstable accuracy, and so on.
Up to now, the lens molding process has been looked on as one of the reliable methods in fabrication of aspheric glass
lens. However, in real production, one has found that it’s hard to control the molding parameters, e.g. molding
temperature, molding period, molding speed and pressing pressure, etc. Therefore it’s necessary to develop the specific
molding processes for a certain glass material. In this paper, SCHOTT P-SK57 is adopted to carry out the lens molding
analysis in order to achieve the relative processing parameters. The molding cases are analyzed based on three different
temperatures of 510°C, 520°C, and 530°C, higher than transition point 493°C of P-SK57. Through continuous heating
and pressing simulation, the results show that the best pressing temperature could be about 530°C, at which the residual
stress is only 5.22MPa (with the molding speed of 0.1mm/s).
The paper overviews various supporting ways for large-size movement mirrors, and the
advantages and disadvantages of the support methods are summarized. Some valuable
optimization methods to improve support effects are also introduced. As a case study, a radial
segment-face contact support method is proposed to solve the support problem for a large-aperture
rotating prism, and a two-step optimization method is implemented to improve the support effects.
The surface deformations under different support separated angles are evaluated. The overview
can be as good references for large-size mirror support design in similar opto-mechanical systems
especially under movement conditions.
Success of optical measurement methods such as interferometry, holography and fringe projection methods
mainly depends on the availability of robust phase recovery (PR) algorithms. In presence of some problems such as
strong noise, abrupt phase changes, low fringe contrast and discontinuities, it’s necessary to introduce a highly
reliable algorithm to overcome them. As for the experimental phase map from holographic interferometry of China
coin’s micro-surface, some typical PR algorithms including the proposed (weighted) DCT (Discrete Cosine
Transform) are studied to try to recover the satisfactory continuous phase. The experimental phase map
(13921040 pixels) exhibits the local surface characteristic of one China coin. It features excessive noise and many
bad data points. It’s very difficult to achieve the continuous phase by the general algorithms. So some robust
algorithms are studied to solve this problem. These algorithms include the quality map guided method (QG),
(weighted) least square method (LS or WLS), mask cut method (MC) and (weight) DCT method (DCT or WDCT).
According to the tested results, the fastest algorithm is the LS method, and it only needs about 10s. The clearest
pattern profile results from the DCT and WDCT algorithms that need about 30s and 31s in the consuming time,
respectively. However, it’s very hard to clearly and completely reflect the smooth top of imprinted letters for all the
above algorithms. This could be due to the excessively much noise, the limited resolutions of CCD setup, and so
on.
A novel measurement of laser coarse-fine coupling tracking is proposed for robot trajectory errors, which can not only
meet the requirements of large range, rapid response and dynamic tracking, but also achieve the high accuracy of
submicroradian magnitude. The mathematic model of robot parameters is deduced according to the motion definition. An
experiment platform together with the test system is built to complete the robot trajectory test. The circular and linear
trajectory, as well as the harmonious motion parameters, is tested respectively. Some error factors affecting the test
uncertainty are given to be considered according to the test experiment results.
A new kind of 3D measurement system has been developed to achieve the 3D profile of complex object. The principle of
measurement system is based on the triangular measurement of digital fringe projection, and the fringes are fully
generated from computer. Thus the computer-generated four fringes form the data source of phase-shifting 3D
profilometry. The hardware of system includes the computer, video camera, projector, image grabber, and VGA board
with two ports (one port links to the screen, another to the projector). The software of system consists of grating
projection module, image grabbing module, phase reconstructing module and 3D display module. A software-based
synchronizing method between grating projection and image capture is proposed. As for the nonlinear error of captured
fringes, a compensating method is introduced based on the pixel-to-pixel gray correction. At the same time, a least
square phase unwrapping is used to solve the problem of phase reconstruction by using the combination of Log
Modulation Amplitude and Phase Derivative Variance (LMAPDV) as weight. The system adopts an algorithm from
Matlab Tool Box for camera calibration. The 3D measurement system has an accuracy of 0.05mm. The execution time of
system is 3~5s for one-time measurement.
A novel car camera lenses have been designed. To reduce the cost, the system gives up the glued lenses. The whole
optical system is made of six separated lenses and an IR-Cut Filter, among which there is two aspheric lenses and four
spherical lenses. All lenses are made of glass. The resolutions of car camera system is three million pixels and the field
angle is 150°.The Tangential MTF is 0.2 at spatial frequency of 300 cycles/mm when field angle is equal to 150°. The
axis MTF is 0.7 at spatial frequency of 150 cycles/mm. The overall length of system is 18.23mm. The F-number is 2.8
and the effective focus length is 1.6mm. The new-designed car camera lenses can provide high-resolutions and very wide
field angle. At the same time, the glass structure of lenses enjoys better thermal and mechanical stability than polymer
lenses. The novel car-camera lenses will be then tested in the experiment.
KEYWORDS: Digital holography, Holograms, 3D image reconstruction, Phase shifts, Digital imaging, Demodulation, Digital recording, Charge-coupled devices, CCD image sensors, Holography
A novel one-shot in-line digital holography based two-dimensional Hilbert demodulation is proposed. By weakening the
object wave compared with the reference wave and applying natural logarithmized operation on the in-line digital
hologram, the real part of object wave can be well extracted. Then utilizing two-dimensional Hilbert transform to
digitally realize π / 2 phase-shift makes it possible to reconstruct the object wave front from single-exposure in-line
digital hologram. Preliminary experimental results are presented to demonstrate the proposed method. This technique
can be used for real time imaging and monitoring moving objects.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.