最新刊期
卷 9 ， 期 1 ， 2004
- 摘要：DCT(Discrete Cosine Transform) transforming coding has been widely used coding-standards such as JPEG,MPEG and H.26x exclusive of JPEG2000 at present. The compressing codings for image/video signals have been advanced to a new stage since the EZW(Embedded Zerotree Wavelet) had been introduced by Shapiro and the SPIHT(Set partition in hierarchical trees) applied by Said et al,which are a competitive challenge against the conventional DCT's coding. However,DCT's coefficients have a similar multiresolutional attributes after reorganization its datum like wavelet transforming. There are many excellent performances in the based-DCT such as arbitrarily shaped DCT,region support DCT and morphological representation of DCT coefficients are proposed.Digital watermarking,image resizing and pattern cognition can be done through utilizing DCT's coefficients,and so on.The capability and performance of the DCT have further been promoted to a higher phase. For instance,object-based coding for DCT is a promising technique for next-generation coding.Besides,the experiments show that MR-DCT(Morphological representation of DCT) is among the state-of-the-art DCT-based image coders reported in the literature.What mentioned above have been studied and analyzed deeply in this paper.Hereby,the future development of DCT is also conjected.2921|211|0更新时间：2024-05-08
- 摘要：Segmentation on weak edged medical image is a difficulty in segmenting technology. In this paper, a simplified snake algorithm for image segmentation is proposed. This proposed model introduces the idea of simplified snake to improve the traditional snake model especially by adding an area energy term with variable coefficients into the internal energy term. This area energy term does well in improving the initialization problem, furthermore, it keeps the low time complexity of original simplified model. And besides, this paper also introduces simulated annealing algorithm to this improved simplified snake model and this algorithm makes a better effects on image segmentation. In this paper, the author discusses the choice of adjacent region, mechanism of generating random variables and the acceptance principles, etc. which are all playing an important role in searching the ideal optimum solution in simulated annealing algorithm. This simulated annealing based simplified snake model proposed in the paper has been tested on medical images. Enough experiments and the results comparing with traditional snake have proved that this proposed algorithm shows a significant improvement in segmenting weak edged medical images with a low time complexity.关键词：Simplified snake;Weak edge segmentation;simulated annealing;global optimization;medical image3013|204|0更新时间：2024-05-08
- 摘要：Image processing has to deal with much information of an image. Maximum entropy theorem of information theory is one of the useful tools to treat with this kind of information. Based upon the maximum fuzzy partition entropy principle, a novel approach for edge detection is presented. After the concept and the principle of the fuzzy probability and fuzzy partition are introduced briefly, a definition of fuzzy partition entropy is proposed. Using the relation of the probability partition and the fuzzy 2-partition of the image gradient, the algorithm is based on conditional probabilities and fuzzy partition. First, a gradient operator is performed and the gradient image is produced. Second, the problem of edge detection is to find a fuzzy partition of the gradient image, which is considered as being composed of edge region and smooth region, and the automatic optimal threshold is searched from gray-level histogram through maximizing the entropy of fuzzy partition. At last, an edge-enhancing procedure is executed on the edge image. The experiment is conducted on various test images and the results show that the proposed approach has better performances than some classical edge detection methods based on gradient do.关键词：edge detection;fuzzy entropy;image segmentation;Fuzzy partition2851|234|0更新时间：2024-05-08
- 摘要：The Hausdroff distance, a measure defined between two point sets, has been used to search for objects in images for its robustness and reliability. In this paper, a highly efficient approach, which uses the Hausdorff distance to locate an object that has undergone an affine transformation, is presented. Prior to our study, the most efficient methods for solving such a problem are two of Rucklidge' s methods:“without zoning" and“4×4 zoning". The proposed approach shares some similarities with Rucklidge' s methods but is mainly based on the“smallest square box distance transform" introduced in this paper. The idea of“smallest square box distance transform" is described in detail in this paper along with its implementation. By replacing the“box distance transform" of Rucklidge' s methods with our“smallest square box distance transform" in the search process, a much higher search speed can be achieved while maintaining the search success rate and target location accuracy. In all, two experiments are given in this paper. One experiment is to test the proposed approach on edge images and the other experiment is based on feature point images. It can be seen from the experimental results of both experiments that the proposed approach apparently outperforms Rucklidge's fast methods.2917|230|0更新时间：2024-05-08
- 摘要：The images matching algorithm is the core of the scene matching aided navigation system, whose performance decides the total performance of the system. In order to improve the navigation accuracy and real-time performance of the scene matching aided navigation system, the choice of images matching algorithm must be suitable. The paper has given a quick and effective images matching algorithm based on image features, which has chosen the part Hausdorff distance as the similarity measuring, and utilized the binary images acquired by the image features abstraction to match images, and simultaneous introduced the neighborhood technology to the searching termination condition of images matching based on the combination of three kinds of images matching speed technology, which has improved the effective of the images matching greatly. The results of simulation show that the algorithm can prove the robustness and validity of the images matching and overcome the effect of noises and geometric distortions. The scene matching based on the algorithm can finish within five seconds, which is equivalent to the global position system output frequency (one second). So it can meet the accuracy and real-time performance needs for matching navigation in the scene matching aided navigation system.关键词：Scene matching aided navigation system;Part Hausdorff distance;Images matching;Image features3074|227|0更新时间：2024-05-08
- 摘要：Iris recognition is an emerging biometric technology for personal identification, whereas iris localization is a crucial part in the process of iris recognition,thus obtaining the iris localization precisely and fleetly is the prelude of effective iris localization . For the purpose of localizing iris precisely, this paper puts forward a novel algorithm of iris localization using block statistic while based on introducing some prevailing algorithms for iris localization. The boundaries that delimit iris can be modeled in a simple way with circular contours. Therefore ,the first step in the paper consists of thresholding the iris image intensity to build two binary bitmaps for the succeeding image procession, one for the whole iris and the other for the pupil. The second step is to search for the centroid of the largest block in the iris binary bitmaps by means of Run Length Encoding (RLE), and calculate the average distance from each point of the boundaries to the centroid obtained before. Experiments show that the algorithm is efficient and successful for the purpose of iris localizing.2778|221|0更新时间：2024-05-08
- 摘要：In the paper, the formation conditions and the characteristics of kernel functions are researched and analysed which are used in kernel principal component analysis algorithm. Kernel principal component analysis algorithm is a new statistic signal processing technique which can extract nonlinear features of images. Kernel functions are key elements for improving it's performance. A new kernel function-combination kernel function is proposed. The new kernel function combines a local kernel function with a global kernel function. The local kernel is conditionally positive definite kernel which can extract local features of images. The global kernel function is polynomial kernel function which can extract global features of images. So the new kernel function can extract not only local features but also global features of images. The new kernel function is applied in kernel principal component analysis for extracting features of images. The test images are MNIST handwriting digits and ORL face database. Features of images are extracting by kernel principal component analysis firstly. Then performing classification by using linear support vector machines, the effect of the new kernel and that of other kernel on kernel principal component analysis are compared. The experiment results indicate the new kernel function certainly improves the performance of kernel principal component analysis.2363|94|0更新时间：2024-05-08
- 摘要：CBIR(Content-based image retrieval) is an image retrieval method that exploits the feature vector of the image as the retrieval index, which is based upon the content, including colors, textures, shapes and distributions of objects in the image etc. The implementation of the image feature vector and the searching process take a great influence upon the efficiency and result of the CBIR. In the paper, a prospective retrieval method based upon the CBIR and the IGA (Interactive Genetic Algorithm) are proposed. Firstly, each image in the collection is segmented into a constant number of sub-images, and the content in each sub-image is computed to make up the feature vector of the image. Then the feature vector of every image is reshuffled in a helix way, to produce the image chromosome, which is used in the following genetic operations. The image chromosome not only characterizes the image, but makes a difference among different parts in the image. During the image retrieval process, the user can employs the IGA to find the image he/she needss for by means of the interaction with the system. In order to promote the efficiency of the IGA, 'Non-uniform Inheriting Operator' is adopted in the genetic operations. Finally, experiments are made and the simplicity and efficiency of the method are discussed.2798|233|0更新时间：2024-05-08
- 摘要：Fractal code is approved an effective method to compress digital image. It proposes a new method to analyze digital image retrieval by fractal iterative function system. By image fractal code and obtained its iterative function, then the image and its iterative function are saved database becoming index file of the image. When database is searched and image index file is compared so that user retrieval images similar to query image. The index files of based on fractal code have three characteristics, the first, similar images have similar iterative functions so can produce similar index files; the second, similar index files can retrieval similar images; the third, no similar images have no similar iterative functions, vice versa. Compare to other methods are used to create database, their index files have not these characteristics. The fractal code produces large of data and needs an effective method to retrieval. So it combines to fractal function information, based on discriminant analysis estimating images similarity, so that determine correlation sequence of all images in database similar to query image. Experiment shows this retrieval method specialties, efficiency and robustness.2831|194|0更新时间：2024-05-08
- 摘要：The technique of image super-resolution restoration makes it possible that high resolution images could be restored from low resolution images recorded by low resolution sensors. super-resolution restoration algorithms may be divided into two classes, particularly frequency domain and spatial domain. All frequency domain approaches made use of the aliasing effect; spatial domain algorithms there are mainly three approaches, i.e. Iterative Backward Projecting(IBP), Projection Onto Convex Sets(POCS) and Bayesian methods. In this paper, a parallel genetic framework algorithm for image (sequence) super-resolution restoration is presented . The parallelism of the real-valued genetic algorithm based on the island model enables better integration of the information of the multiple frame images. Especially with the iterative method of other super-resolution algorithms being the mutation operator, the convergence of the genetic searching in the solution space is fast. The experiments demonstrate that the proposed algorithm is efficient and applicable.2596|82|0更新时间：2024-05-08
- 摘要：A method of wavelet-based multiresolution volume rendering is presented for accelerating 3D reconstruction and interaction of volume data set in network. The scheme is adopted that the volume rendering takes place on client's workstation, in which volume data set on net server is first decomposed into the discrete approximation and detail coefficients by wavelet multi-resolution analyzing, and then these coefficients are orderly transmitted to client's workstation on which the low resolution image is first rendered by using approximation coefficients and successively refined by using detail coefficients as they are arriving. In this process a group of 3D Mallat filters is employed to speed up 3D wavelet decomposition and reconstruction of volume data set, and a discrete and predigested optical model for wavelet domain rendering is brought forward to satisfy real time request of volume rendering. The experiment results show that the method in this paper is highly propitious to the network for frequently selecting and interacting images, because high-quality images and/or outlines can be produced in 12.5 percent and/or much less of volume data set.2654|222|0更新时间：2024-05-08
- 摘要：Although line drawings consist of only line segments on a plane, they convey much information about the three-dimensional object structures. For a computer interpreting line drawings, some intelligent mechanism is required to extract three-dimensional information from the two-dimensional line drawings. The labels of the line drawing as an efficient aidded method is take part in recovering the 3-D scene the line drawing. Many impossible configuration can be exclude by consistent labeling.In this paper, according to current the theory of labeling ,a new labeling method for the line drawing of an object with tangent curve surface is proposed. The labeling rules are established. The natural line drawings can be labeled correctly by using of our labeling method. There are 38 kinds of possible junctions for a hidden-part-draw natural line drawing that is the projection of a three-dimensional object with tangent curve surface, in which there are 6 possible Y junctions, 18 possible W junctions, 14 possible V junctions. The labeling method is efficient to the line drawing of a complex object consisting of polyhedrons and objects with curved surfaces.2678|75|0更新时间：2024-05-08
- 摘要：Sequential simulation of 3D geometry of complex plant that is faithful to botany may occupy a lot of computer resources. In this paper, a method of building complex structures begun by the simplest stems is presented; the simpler substructures are pasted recursively to form more complicated branches. In doing that, the repetitive structure in plant is computed only once, and then can be reused according to its index, thus this method can increase computation speed of complex tree structure, and decrease storage size for the geometrical information as well. These information includes not only geometrical data of the simulated plant at each growth cycle, such as orientation and position of each botanical organ, but also that of each substructure. Although substructure method is a top-down algorithm that is inverse to growth sequence of plant growth, it is shown in this paper that bending branches can still be simulated in this way, as well as reiteration and inflorescence structures, which shows the generality of this method.2886|194|0更新时间：2024-05-08
- 摘要：Coordinating transformation between image of GIS data and geodetic coordinates is problem frequently met in processing of remote sensing images, which involves a contradiction between processing speed and conversion accuracy. To solve this contradiction, a new algorithm which introduces a new conception of Index Array is developed base on analysis existing methods of registration, which can be used to retrieve pixels of corresponding coordinates directly and avoid heavy computation need by traditional searching. This algorithm is characterized of quick accurate transformation between the two coordinates, and allows real-time superposing geographical info, such as geographical coordinates and boundaries, on real time received remote sensing images which proves to be correct by experiments in MODIS real time data receiving.2787|233|0更新时间：2024-05-08
- 摘要：More and more attention has been paid to extraction of roads and other linear features from high-resolution synthetic aperture radar(SAR) images. Due to the complicated background of objectives and the speckle noise in the high-resolution SAR images, it is almost impossible to extract roads directly from original remote sensing images. In order to extract roads precisely from high-resolution SAR images with complicated background and speckle noise, a method of using genetic algorithm(GA) is developed to extract main roads in the paper. After the original SAR images are filtered to suppress the speckle noise, Fuzzy C means is used to classify the images unsupervisedly into vegetation, built areas, roads and other class, and the pixels belonging to the main roads are isolated from the images to simplify the original problem. Then, according to the membership of pixels to main roads and the uniformity of gray, a road model is constructed, and the genetic algorithm is used to search globally optimized roads. The experimental results show that the presented approach can effectively extract main roads from high-resolution SAR images.2678|187|0更新时间：2024-05-08
- 摘要：The ability with distributed computing is a main characteristic for new WebGIS platform. An organizing strategy of distributed WebGIS (DWebGIS) based on environment of server group and its implementation are presented. Then the distributed architecture of DWebGIS and engine of map services for COM+ were analyzed in detail. A distributed solution and process methods about DWebGIS are discussed, in which network load balancing (NLB) of server group, hidden marks of no-status objects, just-in-time (JIT) activation and object pooling are the key technique. Also, process of image based map-entities-objects in Client/Server, "three factors" (The alone file- name are created by SessionID、server systematic time and random number), cycle of system request/response are expatiated on. Lastly, the correctness and feasibility for design of WebGIS system platform based on COM+ technique are proved by an example in Zhejiang Provincial Digital Drainage Area Planning System. Our further R&D work is in progress.2533|210|0更新时间：2024-05-08
- 摘要：Spatial Information Search Engine (SISE), aiming to search the World Wide Web for online geographic information to meet the end users needs, is totally a new research area in the field of spatial information sharing and geographical information system interoperability. In this paper we propose a prototype system of SISE based on the OpenGIS WMS specification. The system architecture, working principles, detailed function of each component, implementation strategies and system performance test results of this SISE prototype are introduced. This SISE prototype is capable of discovering available WMS Servers dynamically, saving and updating capabilities information of them into relational databases and maintaining necessary R tree indexes. For each spatial information request from clients, it can choose qualified WMS servers according to the request autonomously and fetching the matched geographical information for them from remote WMS servers transparently.2668|207|0更新时间：2024-05-08
- 摘要：In order to improve the computation efficiency of image parallel thinning algorithms implemented on serial computers, this paper proposes a coding scheme for each 44 binary pixels, representing the 44 pixels with a double-byte integer, based on which an effective parallel thinning algorithm for serial computers is presented that transforms the 16-pixel based thinning processing into reading, writing and comparison operations of a double-byte integer. By use of this algorithm, the 16 pixels can be processed simultaneously in the thinning operations with serial computers. Computation complexity analysis of the new algorithm is made in the paper that shows about 70% saving in computations compared with the existing parallel template thinning algorithms (e.g., the OPTA algorithm). The algorithm is suitable for all kinds of parallel template thinning algorithms operated on commonly used personal computers, and can significantly improve the computation efficiency of the algorithm with the same thinning results. The potential advantage of the algorithm is also discussed in the paper for using the MMX (or SSE) technology integrated in PCs to further increase the parallel granularity and speed up the thinning processing.关键词：image processing;thinning;parallel algorithm;Template matching3599|204|0更新时间：2024-05-08
- 摘要：ion is a short summary of the content of a longer video document, which helps to enable a quick browsing of a large collection of video data and to achieve efficient content access and representation. In this paper, we propose a novel approach of generating video abstraction at two levels, namely, the shot level and the scene level. The hierarchical video abstraction can facilitate video browsing and retrieval at different granularities. Firstly, the video stream is segmented into shots. A new key frame extraction algorithm, which does not rely on threshold, is put up to extract key frames from shots based on the content variation. An updated time-adaptive algorithm is used to group the shots into scene. Based on the defined shot similarity formula, the video scene structure is constructed after shot clustering and adjustment. Representative frames are extracted from the key frames of each scene using the method of generating Minimum Spanning Tree. Key frames and representative frames can represent the content of shot and scene, respectively. The sequence of key frames and representative frames is the two-level video abstraction. Experiments based on real-world movies show that the method above can provide users with better video summary at different levels.2911|209|0更新时间：2024-05-08
- 摘要：According to the effect of fog from images, a new method for enhancing images taken in fogs is proposed in this paper. It dose not need atmospheric model and just performs on the image, directly enhancing details of the scene in the image. A moving mask is adopted to segment scene at different depth, then each pixel in the mask is processed using block-overlapped histogram equalization. At the same time the influence of mask's size also is discussed in this paper. Though the enhance operation will improve visual result, it also will bring noise amplification in sky region. To restrain the adverse effect, the sky region should firstly be segmented before enhance operate. Using optimal normal distribution to approach the characteristic of image gray distribution, then by the estimated distribution, the range of pixel value in the sky region can be got. By the information, the sky region can be segmented from the image. Experiments show that this algorithm can efficiently improve the degradation of image and enhance the clearness of image.关键词：Fog;Image clearness;Gray distribution characteristic;Normal distribution;Block-overlapped histogram equalization3724|243|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