最新刊期
卷 6 ， 期 7 ， 2001
- 摘要：Triangular mesh is widely used in 3D computer vision because meshes can model objects of arbitrary shape and are easily constructed from sensed 3D data. However the number of the triangules in the mesh is often very huge which makes later process lostly. As a result, mesh simplification becomes a hotspot in computer vision.The Mesh simplification of 3D models for computer vision applications should preserve the shape, the topology and other attribute values (e.g. colors, textures, surface normals, edges, etc) of the object, which makes vision systems work effectively in the presentation, description, recognition and understanding of the object. This paper presents a mesh simplification algorithm with texture preservation by applying edge operations (edge collapse, edge split) to the whole surface mesh. The maximum asymmetric distance of the two meshes is computed as the shape change measure and the maximum texture distance of the two vertices in local neighborhood of the edge is computed as the texture change measure. This algorithm can reduce a great number of faces while preserving the object shape, the topology and texture features. The algorithm is also distributing the vertices of the mesh evenly over the surface of the object.3933|366|0更新时间：2024-05-07
- 摘要：Images may be blurred due to diffusion. This paper investigates the inverse diffusion algorithms for restoration of such images. The conventional One Shoot(OS) algorithm and the Successive Recursion (SR) algorithm are analysed, their performance are compared, and the limits to their validity scopes are revealed and formulated in terms of the blur scales; all these lead to propose a novel "Search and One Shoot" (SOS) algorithm, where the unknown blur scale of a diffused image is detected in a successive approximation manner and the inverse diffusion computing is accomplished in One Shoot with an as small accumulative error as possible. Then, both natural and artificial blurred images are used for tests on the restoration algorithms, including their sensitivity to blur scales of images, their validity scopes , and the effects of SOS algorithm applied to blurred images of incoherent textural structure or scene depth, etc.The experimental results provides evidence that the digital image restoration using inverse diffusion technique can make images of high clarity in scene depth that is beyond the capability of common optical` photographic techniques.2593|330|0更新时间：2024-05-07
- 摘要：The surfaces of many industrial products are formed from the textural primitives by placement rules on the image field. It is an important task of automatic vision inspection to detect the defects in these regular textures. In the paper, two types of matched Gabor filters, that is normal texture matched Gabor filters and defect matched Gabor filters, are designed based on power spectrum of the regular normal texture and the abnormal texture. The normal texture matched Gabor filters, whose band pass area is positioned on the dominant frequency components of the normal texture, are designed to uplift the normal texture while suppress the defect texture. On the other hand, the defect matched Gabor filters, the band pass area of which guarantees the predominance of frequency components of the defect texture, have the effects of uplifting the defect texture while suppressing the normal texture. Taking advantage of the recent progress of study on principle of human vision, a novel adaptive filter design method is presented. Experiments on edge enhancement for defect detection using these matched Gabor filters have yielded satisfactory results, as far as both the precision and the speed are concerned.2769|358|0更新时间：2024-05-07
- 摘要：Some imaging systems employ detector arrays that are not sufficiently dense to meet the Nyquist criterion during image acquisition. This is particularly true for infrared digital imagers. Therefore, the spatial resolution afforded by the optics can't be fully utilized in such imaging systems. This paper presents a new technique to reconstruct a high resolution image, with reduced aliasing, from a sequence of under sampled rotated and shifted frames of the same object. Firstly, the continuous and digital models of an imaging system are defined to explain image acquisition. Secondly, the image registration algorithm and the ordered subset conjugate gradient (OSCG) reconstruction algorithm are designed. Finally, Simulated experiment results are provided to test our algorithm, and we obtain the image whose resolution is 5 times high than the 25 under sampled frames. The results show that OSCG reconstruction algorithm is robust, rapid convergent, and memory saved.2438|331|0更新时间：2024-05-07
- 摘要：This paper firstly made a brief introduction of several methods for information fusion. and also the paper raised a comparison of the fused images which were merged by wavelet transform, PCA an HIS methods, in spatial quality improving and spectral information preserving respectively. Then, the paper introduced several new fusion methods of Multisensor Remote Sensing data, Using algebra calculation and linear calculation,avoid lost spectral information of traditional method, and develop resolution of image, and author performed a discussion on the relationship between several methonds and the fused results, specially made useful advantage of image and overcome disadvantage of image in the fused process. Finally the article introduced a fusion framework of Multisensor Remote Sensing, the author particularly studied every process, include image match, image sharpness enhancement, image fusion, image color enhancement process. The framework is used for a big amount data and a large range image. The article introduced the application of image fusion.2218|303|0更新时间：2024-05-07
- 摘要：A digital watermark is an invisible mark embedded in digital images, video or audio documents, which may be used for a number of different purposes including copyright protection, authentication and captioning. In this paper, we proposed a watermarking scheme that embeds watermarks in the local region of the image. This new idea that we name it local watermarking scheme, inserts watermarks into the wavelet domain of local regions of the image instead of entire image. We select some stable feature points in the image, and then use these points to identify the locations where we insert watermarks. When the image is cropped, the feature points in the remaining piece of the image will not change. We can extract watermark with the help of the remaining feature points. A special quantization procedure is used when we embed watermark bits into the wavelet coefficients of the sub image, and the watermark can be detected without knowing the original image. Experimental results show that the watermark scheme proposed in this paper is robust to cropping, and also has good robustness to compression, filtering, noise adding, and StirMark attack.2947|284|0更新时间：2024-05-07
- 摘要：In this paper we propose a novel quadtree decomposition coding algorithm for residual frames in video coding. First we employ overlapped block motion compensation to reduce blocking artifacts, this leads to smoother motion prediction frames which helps to encode the residual frames more efficiently; based on the nature of human visual system, we regard each 2×2 subblock as a basic coding unit, compute its mean value, and use a new quadtree decomposition method to label the subblocks with larger magnitude than a chosen threshold, then code them with scalar quantizer and adaptive arithmetic coder. Compared with transform-based methods, such as EZW and H.263, the proposed algorithm is more efficient and has better performance, which indicates that it is a very effective and computationally simple technique especially for low bit rate coding. Moreover, the coding and decoding procedures are so fast that it can satisfy real-time applications.2690|231|0更新时间：2024-05-07
- 摘要：In the situation of image synthesis for image based rendering(IBR), we need not only precise matching but also a dense disparity map. In order to meet the needs of this application, a feature constrained stereo matching algorithm is presented. It integrates feature based and area based process. First we obtain a sparse precise disparity map of edge points using feature based approaches, next we fill in the disparity data of featureless points by using area based algorithm under the constraint of the acquired disparity map. In this way, not only can we reduce the search for correspondences but also ensure the validity of matching. During the matching phase the two images play a symmetric role to ensure the validity of matching. At the same time, the technique of stereo matching with adaptive window based on the gray value information of nearby points is adopted in the area based process. The experimental results show this algorithm is effective and it is of great value to use.2645|311|0更新时间：2024-05-07
- 摘要：Because the number of domain blocks is not sufficient in the Jacquin's fractal image coding method, the matching between domain blocks and range blocks is not very effective. To improve the matching efficiency, this paper puts forward a new fractal image coding method, which increases the number of domain blocks through linear rotation transforms and gets the optimal matching. Unlike in the ordinary methods, in order to exploit fully the self similarity in the image, this method chooses domain blocks from multiple directions in the image, then generates new sets of domain blocks through a symmetric transform and linear rotation transforms. The Isometries transforms proposed by Jacquin can be viewed as a special case of this method when the rotation direction number is fixed to 4. This method modifies the number of rotation directions, thus amends for the inadequacy of domain blocks in the Isometries transforms, and gets a better compression result. An acceleration algorithm based on a lookup table is also established for the linear rotation transforms, which effectively shortens the encoding time.关键词：Fractal image coding;Linear rotation transforms;Isometries transforms;Range block;Domain block2623|324|0更新时间：2024-05-07
- 摘要：CL multiwavelets have some characteristics different from that of other multiwavelets such as the commonly used GHM multiwavelets. After the multiwavelet transformation of an image, most energy of the image is mainly concentrated on its lowest resolution subimage. As for CL multiwavelets, most energy of the lowest resolution subimage is further concentrated on its first component. This implies that the number of most significant coefficients to be encoded in CL multiwavelet image coding is approximately one fourth as many as that in other multiwavelet image coding. However this characteristic of CL multiwavelets as well as its potential possibility of application to image coding have not been mentioned yet. In this paper, we first reveal this characteristic of CL multiwavelets from theoretical aspect and experimental data and then apply this characteristic in CL multiwavelets image coding. We have achieved the results better than that of GHM multiwavelet image coding published so far in terms of compression ratio as well as PSNR.2394|295|0更新时间：2024-05-07
- 摘要：In order to improve quality of fractal image coding and reduce coding time, a fast polynomial fractal image coding method based on wavelet decomposition is presented in this papre. Jacquin's fractal image coding method needs searchign for the optimum domain block in the image. It must take so long time to do this searching work. This character is a fatal flaw of the method which confine this method in using. Monro's fractal image coding method doesn't meed searching for the optimum domain block, so his fractal image coding is very rapid and simple in reference. His coding method is zero searching fractal coding method and its coding time is very short. But his reconstructed image has blocking effect. In this paper. in order to solve this question, a new fractal image coding based on discrete wavelet transformation is presented. The new method also doesn't need searching for the optimum domain block, so its coding time is very short. This algorithm utilizes the character of wavelet analysis which is unbalanced distributions of energy in subbands among wavelet trasform image. The algorithm based on discrete wavelet transformation is:First, the image is decomposed into different channels by discrete wavelet transformation. Second, wavelet coefficients are encoded by fractal image coding method. Simulation shows that the quality of the reconstructed image is improved greatly and coding time is only 1.48 seconds.2565|288|0更新时间：2024-05-07
- 摘要：Medical imaging techniques produces a lot of digital images. Conventionally, lossless compression methods are used to make store and transmission to be more effective. To achieve high compression ratio, lossy compression must be exploited. However, lossy compression will distorted the original images and so have to be applied carefully. A medical image composes of two categories of regions. The regions in the first category include important information for diagnosis. The compression with very good quality of reconstructed image is necessary to these regions since the cost due to error representation will be high. On the other hand, information in the other category of regions may be less important and so compression to these regions should achieve high ratio. Based on this idea, we proposed a task oriented compression algorithm for medical images in this paper. We integrate the lossless compression and lossy compression via wavelet transform. Lossless compression is applied to the regions of interest and lossy compression to other regions. The proposed algorithm performs well in both compression ratio and quality of reconstructed image.2147|309|0更新时间：2024-05-07
- 摘要：This paper presents a novel face detection algorithm, which consists of two parts of research work. The first one is a frontal view upright face detection algorithm which is based on the well known singular value feature (SVF) and hidden Markov models (HMM). A trained HMM is employed to classify the SVF of the sub-image at every location in the image as a face or nonface. The algorithm couples the virtues of both the SVF and HMM and produces excellent detection results. It is tested on a collect photo album and has detected the 85.1 percent of its 484 people, while 97 false alarms are also reported. The second part of our algorithm is the extension of the first one to rotation invariant face detection. Several HMMs are utilized to recognize the SVF of the sub image at the same time to obtain the angle of the "face" image. Then the HMM for detecting the upright faces is employed to verify the faceness of the rotated test pattern. The rotation invariant algorithm is tested on another image set where there are 173 persons. The detection rate is 72 2%,which seems not too high,but the false alarm rate is also low, only 34 out of the 6 100 000 windows,scanned.2625|404|0更新时间：2024-05-07
- 摘要：The new video&audio coding standard MPEG-4 is enabling content-based functionalities. It takes advantage of a decomposition of sequences into semantically meaningful video object (VO). A snapshot of a VO is named video object plane(VOP) at a given time and a series of VOPs represent one moving object. This is a very challenging task, because physical objects are normally not homogeneous with respect to low-level features such as color, luminance, or optical flow. Hence, conventional segmentation algorithms will fail to obtain meaningful partitions. In this paper, a new automatic VO segmentation algorithm based on modified Hausdorff object tracker is presented. A binary model for moving object is automatically derived and tracked in subsequent frames using the modified Hausdorff distance. First initial model is extracted and tracked based on the proposed object tracker against subsequent frames in the sequence. Then the model is updated every frame to accommodate for rotation and changes in shape. The video object is extracted by a series of binary models in the end. Furthermore, to improve the quality of segmentation and to reduce the computational complexity , stationary background is filtered by a novel technique. Experimental results demonstrate the performance of our proposed algorithm.2431|354|0更新时间：2024-05-07
- 摘要：The rapid development in multimedia technology and the possibility of sharing and distributing image data through large bandwith computer networks have contributed to an increase of visual data in the global information exchange but also posed a tough task of retrieving image efficiently before researchers. In this paper, an approach of image retrieval using combined fuzzy histogram is proposed. The proposed approach uses combined color and texture features in this way: we first get the color fuzzy histogram in HSI color space in a bulk processing manner, then we use the texture features to expand the color fuzzy histogram and get the combined fuzzy histogram. In particular, the method of extracting color features and texture features of image and the computation process of constructing combined fuzzy histogram is introduced. Experimental results of this approach on an image library consisting of 200 color images are presented to indicate the good performance of the proposed approach.2591|317|0更新时间：2024-05-07
- 摘要：The conversion of drawing image into vector form is of important in many fields, especially in the field of electronic and mechanical applications. In this paper, a new approach for extracting determinant contour points of lines in scanned engineering drawings is proposed. This method can not only obtain determinant contour points placed in line contour of the drawing image but also process these points automatically. The basic idea of proposed algorithm is totally different from that of traditional algorithms, it can remedy some deficiencies which are suffered from traditional algorithms and executes successfully without thinning. This algorithm detects black runs on scanning or consecutive scanning lines automatically, and then extracts determinant contour points from scanned engineering drawings, the main idea of the detecting algorithm is that:first,black runs on current scanning line are detected automatically, and determinant contour points are extracted by detecting the edge points of both current black run and the next black run which is on consecutive scanning line if the relevant condition holds. Second,based on the determinant contour points,the extraction of lines from the original drawing image will be done easily by using B spline theory. Experimantal results indicate that the algorithm has higher speed and other advantages.2905|392|0更新时间：2024-05-07
- 摘要：Polygon simplification and aggregation play an important role in automatic cartographic generalization. Design of this kind of operator algorithm requires to consider geometric and topological characteristics, and needs to find a spatial data model to support extraction of interesting adjacent area, such as bend feature, gap region between polygons, neighbor area between islands within polygon. In this process, detection of neighbor relation is a key step. Delaunay triangulation structure is a powerful tool of spatial neighbor analysis due to its circumcircle principle and closest to equilateral properties. Applying Delaunay triangulation principles, this paper presents a spatial data model and constructs formal selection strategies. On the basis of constructed model, the study systematically investigates the problem of polygon bend simplification, exaggeration as well as neighbor polygon aggregation. The paper divides polygon simplification into steps: islands selection, concave part fill, shallow area skeletonizing, concave part exaggeration and convex part clip. As for polygon aggregation, it is classified as three cases: absorbing, enclosing and separating and three methods are discussed respectively. Some of operator algorithms are presented in the paper and corresponding experiment results are illustrated.4423|347|0更新时间：2024-05-07
- 摘要：Geometry based rendering(GBR) is a subject of traditional computer graphics, Along with the development of computer graphics, Image based rendering(IBR) becomes an important approach. GBR and IBR have advantage and disadvantage independently. This paper gives a novel approach of geometry and image based hybrid rendering(GIBHR) aiming at the status quo of computer graphics, analyses the research contents of GIBHR, and , gives a rendering pipeline of GIBHR. GIBHR can overcome the disadvantage of GBR and IBR, which can render geometry object and scene freely like GBR, and, has a high rendering speed like IBR, by texture map, it can keep high reality also. This paper expands algorithm of 3D WARP of IBR into geometry based rendering, proves the feasibility and validity, and, gives an algorithm of hybrid rendering based 3D WARP. This algorithm can overcome pixel discrete phenomenon of 3D WARP algorithm. At last, this paper gives an experiment result of hybrid rendering algorithm, it proves that the algorithm of hybrid rendering based 3D WARP can speed the rendering of geometry models obviously, and it has the ability of space anti aliasing.3095|362|0更新时间：2024-05-07
- 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