最新刊期
卷 5 ， 期 11 ， 2000
- 摘要：In this paper, Research background of automatic face recognition and its relation to human vision system are briefly reviewed. Then current face recognition technologies are roughly introduced and classified according to different recognition features. Four main algorithms are analyzed and compared. The first is eigenface, which is extraction of global features using the PCA. In this approach, a set of faces is represented using a small number of global eigen vectors, which encode the major variations in the input set. The second is flexible model, which separate shape and gray parameter. The third is wavelet-based elastic graph matching, in which memorized faces are represented by regular graphs, whose vertices are labeled by a multi resolution description in terms of localized spatial frequencies. Spatial relationships within the object are labeled by geometrical distance vectors. The last method is traditional analytical techniques. Based on the analysis and comparison, key factors in face recognition technologies are concluded and distilled as suggestion to future research.关键词：face recognition;eigenface;Wavelet-based feature;Shapeless model7219|274|0更新时间：2024-05-08
- 摘要：The general optimal set of discriminant vectors is the extension of the Foley-Sammon optimal set of discriminant vectors. First, this paper gives the definition and existed calculating method in theory, through which it is found that the existed method has two principal problems:(1)The general optimal discriminant vectors are calculated step by step, which can not make sure that the corresponding general Fisher discriminant function can reach the maximum; (2) When the popular scatter matrix is singular, it is possible that there exists one discriminant vector on which the between-class distance of the projected set of the training sample set is equal to zero, which is meaningless for classification. To solve the above two problems, a new method for calculating the general optimal set of discriminant vectors is presented. In the end, our method is applied to human face recognition. Experimental result shows that the new method is superior to the existed method in terms of correct classification rate and stability.关键词：General optimal set of discriminant vectors;Foley-sammon optimal set of discriminant vectors;feature extraction;pattern recognition2191|334|0更新时间：2024-05-08
- 摘要：We can get much useful information by remote sensing sensor. Wake disturbances contain information about the generating vessels speed, course, length. Space-based sensor system could provide naval force with an enhanced capability for surveillance of ship traffic in open ocean areas if the information content of signatures associated with the hydrodynamic wake of a moving ship can be exploited. In this paper, the detection of wakes obtained form satellite images using synthetic aperture radar techniques is discussed. First we analyze the patterns of ship wake in SAR images. Then it is concluded that the problem can be considered a detecting line problem, and we describe the theory and the characteristics of Radon Transform briefly. Finally, we apply Radon Transform to detect narrow-V ship wake from some ERS-1 SAR images. The results show that the algorithm is effective.关键词：synthetic aperture radar;Detection of ship wake;RADON transform2507|189|0更新时间：2024-05-08
- 摘要：This paper proposes that the way of imaging on retina in virtual environment(VE) is the combination of both the imaging way of the real world and that of planar images. Since the binocular parallax is important for the three-dimension depth perception, we found that the problem with conventional VE in vision lies in the inconsistency of divergent degree of light with the binocular status. It is also emphasized that the reason of that conventional VE easily causes some persons dizziness and eyestrain is because virtual objects usually make binocular vergence and accommodation conflict. At last, this paper lists and analyzes some factors which might bring discomfort such as delayed effects, resolution, interpupillary distance and so on.关键词：virtual environment;Visual physiology;parallax;Three-dimension depth perception3023|184|0更新时间：2024-05-08
- 摘要：Edge detection efficiently and accurately indicates the boundary points of image areas, so it is popularly used in various computer vision applications. However edge detection alone is not a whole image segmentation process, because usually the detected edges are not continuous and many loose edge points exist in high detail areas. In this paper we present a novel approach called edge growing to attack edge discontinuity after edge point detecion. Every salient edge point in an edge end would grow forward based on the edge structures in its neighborhood. All the edge end points grow simultaneously until closed edge contours are presented and the image is segmented into closed regions. After that, salient regions can be identified by its horizontal and vertical spans and extracted by contour tracking. Therefore high detail areas enclosed by the adjacent salient regions can be indirectly extracted as a large area, without grouping these high detail areas to use some complicated algorithms. As a procedures after edge detection, the algorithms can be applied in diverse applications, and can be embedded in other complicated segmentation procedures to incorporate edge information. Experimental results show that the algorithms proposed in this paper achieve excellent performance in color image segmentation.关键词：Image segmentation;edge detection;Edge growing2115|164|0更新时间：2024-05-08
- 摘要：This paper is focused on image reconstructions from incomplete projecting data. Based on the assumption of continuity among image pixels, the linear system which the optimal solution satisfies is derived for incomplete projecting data. An artifitial parameter is then introduced and a new linear system is formed from the original one. First, the new system is solved by perturbation method and the expansion cofficients are computed by an iteration; second, the solution to the original system is achieved by applying rational approximation of vector-valued funcation to the perturbation solution. The difficulty of the high computation amount of the direct solution of the original linear system is overcome by the proposed method. Numerical examples indicate that when only three items in the perturbation expansion are used, satisfactory approximations to the original images can be achieved by the rational approximation method.关键词：Image reconstruction;Imcomplete projecting data;Rational approximation2178|159|0更新时间：2024-05-08
- 摘要：A new fractal predictive image coding scheme based on zerotrees of wavelet coefficients is presented. We first pick out the range blocks to be predicted based on the zerotrees from subimages of the same orientation but different resolutions. These range blocks will form a range tree, something like zerotrees. For each range block, we then look for its optimum fractally-matched domain block in the subimage of the same orientation but lower resolution than that of the range block. Each domain block obtained this way can extend through different resolutions along the same orientation to form a domain tree. Among all these domain trees, one closest to the range tree will be its predictive tree. Theoretical analysis and experimental results presented in the paper show that our scheme can obtain higher compression rate at the little cost of signal-to-noise ratio.关键词：Fractal image coding;Wavelet image coding;Zerotrees of wavelet coefficients2599|191|0更新时间：2024-05-08
- 摘要：To further improve the performance of H.263' s encoder in the real time H.324 system, a motion search algorithm using zero block detection technology, is proposed in this paper. This algorithm exploits the fact that many residual blocks become all zero blocks after DCT and coarse quantization in very low bit rate video coding. To further speed up the encoder implementation, this zero detection technology is combined with motion estimation. In the motion search process, if an all zero macro-block is detected using the distortion value of the macro-block, the search process is stopped. The choice of the threshold for zero detection is critical. Simulation results show that the algorithm can dramatically decrease the motion search points. And also many blocks' quantization, DCT and other reconstruction steps are skipped. Under typical quantization parameter of 12, the average search points decrease is 30%~50%, and 25%~30% blocks' DCT and quantization is skipped due to the zero detection in the motion search. With the gain of encoder speed, the algorithm does not suffer the coding efficiency of the H.263 baseline. The algorithm does not affect the reconstructed image quality either. After the algorithm is combined into a real~time H.324 system, the performance of the whole system is improved a lot.2594|178|0更新时间：2024-05-08
- 摘要：This paper presents an existent and constructive proof of image fractal compression mappings and application in fractal image coding. In addition, based on the procedure of the existent and constructive proof of image fractal compression mappings, a new fractal image coding algorithm is presented. Some beneficial theory about these constructive fractal mappings are also discussed. The theoretical result shows that the theory is feasible. The experimental results prove that the new algorithm can improve reconstructing image while reducing encoding time.关键词：fractal;Constructive proof;Image coding;Compression map1925|153|0更新时间：2024-05-08
- 摘要：In the conventional iterated function system(IFS) image compression, the threshold of distortion between range block and domain block is prefixed which does not take the statistics characteristics of current range block into account. Thus the encoding scheme can not adapt to different input images and sometimes the encoding performance relies on the skill of the operator. In order to avoid that problem, a new scheme to compress still grayscale image is presented. The difference between this new scheme and the conventional one is that the threshold in the new scheme is calculated from the variance of the range block. Therefore the encoding system can adapt to the given image. The threshold is analyzed and a conclusion is drawn that the threshold is a positive portion of the variance of current range block. Furthermore, this new encoding scheme is proved to be in accordance with features of human visual system. The adaptive encoding scheme improves the effectiveness of encoding. This new scheme also makes the IFS encoding algorithm more practical.关键词：Fractal image coding;Iterated function system(IFS);Quadtree partitioning;Adaptive threshold(AT)2028|163|0更新时间：2024-05-08
- 摘要：Images are lifelike, specific and directly perceived through the sense. They contain a great deal of information to be regarded. However, a lot of time is spent on transporting image-information, when you scan on Internet for information. This paper introduces a novel image scaling technique by multiscale mean values. The images of various spatial resolution are generated by the mean values of blocks in the original image. Combined with wavelet transform for image-layered coding gives high compression ratio and increase flexibility. The speed of information downloaded will be accelerated based on this scheme. It will also be convenience for users.关键词：Image scaling;Image coding;Multiscale mean values;wavelet transform;Internet;Information transport2110|159|0更新时间：2024-05-08
- 摘要：Point-pattern matching is an important problem in the fields of computer vision and pattern recognition. Its main task is to pair up the points in two images of a same scene when there is a geometric transformation relating the two images. Once the correspondences between the two sets of points are set up, we can recognize objects and locate their poses in many optical sensor applications. However it is well known that the imaging geometry of a TV camera is a nonlinear perspective transformation. To simplify the difficult solving process, many researchers approximate the perspective-transformation relation with an affine transformation. Obviously it will increase the possibility of fault matchings. In this paper, a new algorithm is proposed to solve the problem of matching two point sets with the different cardinality under a perspective transformation without simplifying the original perspective-transformation relation. Supposing two sets of three points have been matched as a whole beforehand, based on projective coordinates andp2-invariant theory, a new concept named generalized distance is introduced. Theoretical analysis and simulation results show that the new algorithm is fast and effective.关键词：Point-Pattern matching;perspective transformation;Projective coordinates;p-invariant;Generalizeddistance2676|146|0更新时间：2024-05-08
- 摘要：This article presents problems coming out with the aspect of visual effect quality control of the digital film making process. The various factors contributing to effects of visual quality during image reproduction from analog to digital and from digital to analog are analyzed and the practical methods and instruments to measure them are introduced. To construct an effective model for comparing the visual effect quality of images recorded or represented on different media such as film, videotape, hard disk and monitor, we separate a series of functions, which reflect the abilities and characters in image reproducing of different devices. Those non-linear functions are identified as curves that can not been converted into simple formulas precisely. Referring to the Color Profile System used in desktop press industry, we adopt lookup tables for each function and device. Those lookup tables represent the characters of each device and the cross-influence are calculated by software. Based on the studies of those lookup tables, we developed a visual effect management system for digital film making studio, which combines the device calibration, system adjustment and image quality measurement together to parallel the digital film making procedure.关键词：Digital film;Visual effect;Color profile system;Image reproduce2210|173|0更新时间：2024-05-08
- 摘要：This paper introduced mode of collecting、technology of filtering and measuring process about image of three dimension composite preform surface. After image of preform surface which was filtered by based direction adaptive filter(BDAF), we can calculate spectrun of image of preform surface that had been manipulated. We can measure average angle using method of texture analysis. It has been proved to be correct by sandard image. The experiment shows the system will be accurate. Practice shows that it is one of the effective and new technique about measuring average angle of three dimension composite preform.1913|140|0更新时间：2024-05-08
- 摘要：Affected by the wearing of frequent usage and variation of environmental brightness, binaried numerical character images of freight train become deformed and incompleted, which are sampled by freight train mark recognizing system under various weather conditions. As traditional methods, for example pattern plate matching, are not effective to extract the character features exactly, a new method called fuzzy-statistic is presented according to which the numerical character' s figure features inside and outside them are extracted by computing their direction routes. By the mean time the fuzzy decision method is drawn on to overcome the indefiniteness during feature extracting. After utilize in some power plant, it appeared that the method is available for character's deformed and incompleted, also high recognized-rate was obtained. The method might be referred to recognize bold printed numerical characters and other courses.关键词：character recognition;Fuzzy decision-making;Statistic method;Figure;projection2891|254|0更新时间：2024-05-08
- Postal code：100190
- Tel：010-58887035/58887030/58887418 Email：jig@aircas.ac.cn
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备05080539号-4
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0