KEYWORDS: Data modeling, Network architectures, Cameras, Visual process modeling, High power microwaves, Head, Feature extraction, Video surveillance, Surveillance systems, Surveillance
Person re-identification (Re-ID) aims at retrieving a person of interest across multiple cameras. With the advancement of deep network and increasing demand of intelligent video surveillance, it has gained significantly increased interest in the computer vision community. In this paper, we propose a simple yet effective Multi-Scale Horizontal (MSH) model for person Re-ID task. Firstly, the model consists of a novel Multi-branch network which adopted residual network ResNet50. There are two branches in our network: global branch and local branch. In local branch, the model slice a person into different parts in multi-scales. Secondly, we present a mix pooling method which considering both average and maximum pooling method. Finally, we employ triple loss and softmax loss as the loss function of the network. Experiments on two datasets (Market1501 and DukeMTMC-reID) demonstrate the advantage of the proposed model.
In order to discuss the error estimation of singular integral quadrature formulae, we construct new quadrature formulae with the anti-Gaussian quadrature, and the accuracy and remainder expression of quadrature formulae are given. The quadrature formulae are superior to the Gauss-Kronrod quadrature formulae in estimating the error.
In computer vision, pedestrian detection is a key problem. In this paper, we propose to speed up the HOG+SVM algorithm without sacrificing the classification accuracy. In order to eliminate the effects of aliasing phenomenon that products in the process of HOG extraction, we used trilinear interpolation to extract feature. This paper proposed HOG pedestrian detection method based on edge symmetry. In these experiments, we used INRIA dataset. Traditional HOG pedestrian detection is presence of slow detection speed and low detection rate. Experiments show that using trilinear interpolation and edge symmetry not only can improve the detection effect, but also can improve the detection rate.
A light field imaging system based on electrically tunable nematic liquid crystal (LC)-micro lens array (MLA) is proposed. In this system, a LC-MLA is utilized to replace the conventional glass MLA. The LC-MLA has an advantage of electrically controlled without any mechanical movements. This feature could extend depth of field (DOF) of the LCMLA. In the measurements, the relationship between the external applied voltage and the focal length of the LC-MLA is discussed. The point spread function (PSF) of the LC-MLA is given. An interesting experiment for measuring the direction of the incident light is presented. Those measurements show that the proposed LC-MLA has greatly potential in light field imaging system.
For image classification tasks, the region containing object which plays a decisive role is indefinite in both position and scale. In this case, it does not seem quite appropriate to use the spatial pyramid matching (SPM) approach directly. In this paper, we describe an approach for handling this problem based on region of interest (ROI) detection. It verifies the feasibility of using a state-of-the-art object detection algorithm to separate foreground and background for image classification. It first makes use of an object detection algorithm to separate an image into object and scene regions, and then constructs spatial histogram features for them separately based on SPM. Moreover, the detection score is used to rescore. Our contributions include: i) verify the feasibility of using a state-of-the-art object detection algorithm to separate foreground and background used for image classification; ii) a simple method, called coarse object alignment matching, is proposed for constructing histogram using the foreground and background provided by object localization. Experimental results demonstrate an obvious superiority of our approach over the standard SPM method, and it also outperforms many state-of-the-art methods for several categories.
In this paper, we will present a smart structure based on an electrically controlled liquid crystal (LC) Fabry-Perot to achieve terahertz (THz) filter, which has extremely potential in THz communication. This proposed structure doesn’t need any mechanical movements because of adapting LC as a key material to compose the Fabry-Perot device. The THz filter based on LC, which is smart, light and cheap, can be realized to solve that common problem of short of tunable devices in THz radiation. The chosen LC material is E7, which has very stable and good transmissions in THz range. Under the external applied voltage, the alignment of the nematic LC allows the refractive index of the device to be tuned. Because of this feature, the resonant peaks could be shifted by changing the applied voltage. Especially, when the alignment is changed from planar to phototropic, the maximum value of the shift could be realized. The simulation result of the proposed device could be got. And the optimal structural parameters could be also got. Numerical analyses results have shown that the proposed structure has a high narrow transmission band and very sharp edges. This THz filter is novel for compact and smart features, so this kind of proposed THz filter is very attractive in many applications, such as THz communication, and THz spectral imaging.
Based on a proposed electrically tunable liquid crystal (LC) micro-lens array (MLA) instead of a commonly used microlens array with fixed focal length in a conventional type, a new prototyped Shack-Hartmann sensor is reported. The LCMLA with 128 × 128 elements is fabricated by the methods of photolithography and hydrochloric acid etching. Composed of the proposed LC-MLA and a CCD, a new type Shack-Hartmann wavefront sensor is got. This kind sensor can solve problems of the tradition wavefront sensor that the larger measurement range and high measurement accurate can't be realized by the same device. Except for adaptive switching the two working modes, this wavefront sensor also has a dual-mode imaging feature with obtaining wavefront information of the target and it's two-dimensional optical intensity image at the same time. In order to verify it's characteristics, an extreme experiment is designed, which introduces a distortion wavefront. At this circumstanc, the traditional wavefront sensor can't get anything. However, with proposed wavefront sensor, this situation can be solved by adjusting the applied voltage of LC-MLA to change it's focal length. With a reconstruction method, the three-dimensional information of the wavefront can be got. At the same time, the two-dimensional optical intensity image is also got. From the experiments, we can prove that it can effectively improve detection sensitivity and dynamic measurement range of wavefront. Results of the prototype demonstrated qualitatively verify this feasibility. This kind new type wavefront sensor will have a wide variety of applications in adaptive optics.
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.