最新刊期
卷 10 ， 期 1 ， 2005
- 摘要：Due to color image providing more information than monochrome image, color image processing is being paid more and more attention. Image segmentation is critical to image processing and pattern recognition, so all the typical approaches are presented and discussed in this paper. Basically, color image segmentation techniques are based on monochrome ones operating in different color spaces. This paper first reviewed some major color representation methods, then summarized the major color image segmentation approaches including histogram thresholding, characteristic feature clustering, region based approaches, edge detection, fuzzy techniques, neural networks, physics based method. The merits and drawbacks of the methods were discussed too. Fuzzy set theory provides a mechanism to present and manipulate uncertainty and ambiguity, which is desirable for image processing. So, the fuzzy approaches will have a promising application in the color image segmentation area.5815|343|0更新时间：2024-05-07
- 摘要：The visual analysis of non rigid motion is becoming a major area in computer vision. A classification of non rigid motion performed by objects on various degrees of non rigidity is elucidated in this paper according to the curvature changes of the objects, indicating that an 8 class scheme can cover all the objects from rigidity to high non rigidity. A large number of existing approaches in this area are surveyed and divided into two major categories: feature based methods and model based methods, and the advantages and disadvantages of the two classes are discussed respectively. The last section of this paper expatiates on the difficulties confronted by non rigid motion analysis and expects its possible future directions. Non rigid motion analysis is a burgeoning research area but still in its early development. The available body of research has begun to make inroads into only a few of the many difficult problems. Most of the extant models are far from perfect and people are far from a general solution to this problem. Some related research fields, such as speech recognition and computer graphics, may benefit this field.4614|216|0更新时间：2024-05-07
- 摘要：In this paper, some initial investigations are conducted to employ Clonal selection for classification of multi spectral remote sensing image. The clonal selection is used to explain the basic features of an adaptive immune response to an antigenic stimulus. The general algorithm, named clonal selection algorithm(CLONALG), is derived from clonal selection to perform machine learning and pattern recognition tasks and it has been adopted to solve optimization problems. In this paper, image classification task by CLONALG is attempted and the preliminary results are provided. The experiment is consisted of two steps: Firstly, the classification task employs the property of clonal selection of immune system. The clonal selection proposes a description of the way that the immune systems copes with the pathogens to mount an adaptive immune response. Secondly, classification results are evaluated by applying three known algorithm: parallelepiped, minimum distance and maximum likelihood. It is demonstrated that our method is superior to the three traditional algorithms, and its overall accuracy and Kappa coefficient reach 93 63% and 0 915 respectively.2613|221|0更新时间：2024-05-07
- 摘要：In this paper a new image segmentation model based on techniques of curve evolution, piecewise smooth Mumford Shah functional for segmentation and level sets was proposed as improvement of C V method. New method shows global optimization and less insensibility of initialization and can detect objects whose boundaries are not necessarily defined by gradient and solves the problem of locating the edges on images with non uniform brightness, for which the previous methods based on piecewise constant Mumford Shah model, including the C V method, are not applicable. Besides, the model was improved for the location of subtle and complicated edges of target objects by the modification of PDE. In order to further stabilize and fasten the level set evolution procedures, the paper addresses an improved approach to construction of the signed distance function using new Voronoi source scanning method, which needs simple comparison and few multiplication operations, faster than the traditional approaches. Finally, various experimental results for synthesized and real images will be presented to prove the proposed model efficiency and stabilized.2336|206|0更新时间：2024-05-07
- 摘要：In cell images the clustering phenomenon frequently appears. In order to separate the clustering cells, an automatic segmentation algorithm based on a novelB2spline active contourmodel (B2Snakemodel) is proposed. The novel B Snake’s energy terms depend notonly on the cell image itself butalso on amodified nonlineardistance image. Firstly, the algorithm generates the distance image. The distance image is called a nonlinear distance image because of the nonlinear relationship between the gray levels ateach pixel and the distance from thatpixel to its nearest background pixel. To produce themodified nonlinear distance image, an iterative erosion method with a dynamic structure rather than a fixed one is used in the algorithm. Secondly, the B Ssnake model is initiated via morphological operation on the modified nonlinear distance image. Two initializationmethods, one is faster but the other ismore precise, can be chosen according to the cell types in the cell images. Finally, the novelB2Snakemodel obtains the cell boundaries under the effect of both the original image force and the modified nonlinear distance image force. Experimental results show that our algorithm is effective for automatic cell image segmentation.2474|217|0更新时间：2024-05-07
- 摘要：The original purpose of Snakesis is for image segmentation. The method suffers from a strong sensitivity to its initial position and can not deal with topological changes. Its sensitivity to initialization can be overcame by the genetic algorithms (GAs). The GAs is a global optimal searching algorithm and has better numerical stability. But its disadvantages are the computational complexity and the rapid increasing of computation by the augmentation of the search space. They both affect the convergence rate of the GAs. This paper presents an image segmentation method of Dual T Snakes model based on the GAs. By making use of the Dual T Snakes model, it inherits the capability of changing the topology of the T Snake, reduces the valid search space for the GAs to remedy its limitations. The solution of the Dual T Snake consists of two curves enclosing each object boundary, and it is composed the valid search space of the GAs. The optimal object boundary can be obtained through the operation of selection, crossover, and mutation. The new model can accelerate the convergence rate while inheriting the capability of changing the topology of the T Snake, avoid local minima from Snakes model, and maintain the global optimal ability of the GAs, then obtain more precise segmentation. Better results are achieved in application of this method on segmentation of cardiac magnetic resonance images.关键词：Snake model;Dual T-Snake model;image segmentation;genetic algorithms2590|194|0更新时间：2024-05-07
- 摘要：Edge detection is one of the basic pre processing methods in digital image processing and computer vision. It is one of the most important algorithms making the processing going further. Edge detection is essentially a problem of color difference computing. Although edge detection has been explored deeply in gray image, it is still a difficult problem in color image for its trouble to define the color difference including not only value, but also direction. In this paper, a new color edge detection algorithm is proposed. This algorithm is based on Human Vision System (HVS) separating the chroma and intensity information of RGB color to emphasize the more important one. Inspired by this important property of HVS, the intensity difference and chroma difference are computed respectively, and then the weighted mean over these two differences is taken as the final color difference. This algorithm solves some problems produced by the existing algorithms and is proved to be a practical and effective color edge detection algorithm.2888|218|0更新时间：2024-05-07
- 摘要：This paper presents an algorithm for extracting line segments from character images and thinning character images simultaneously. At first, the line segment region, the subsegment region and the junction region of an image are defined, and an image to be thinned is considered as a set of line segment regions. Secondly, themethod of extracting all junction regions and sub2segment regions is described based on connected pixel segments in rows and columns of an image, and themethod ofmerging all subsegment regions and junction regions that belong to the same line segment region is presented. The Bernstein Bezier polynomialmethod to fit every line segment region’s skeleton is also discussed. Finally stability analysis and experimental results of the proposed algorithm are reported.2925|201|0更新时间：2024-05-07
- 摘要：As a key image enhancement technology to remove the noise in images, image smoothing is a necessary function in various image processing software. Excellent smoothing algorithm should be able to deal with images corrupted by various noises and has capability of preserving detailed information. Based on analyzing a few of keeping details smoothing algorithm (gradient weighting filter, self adapted median filter, robust smoothing, edge keeping filter) and according to each adaptation to various noise, a robust smoothing method is presented combining non linear and linear filter. Tests on corrupted image of various noise show that the algorithm can deal with corrupted image robustly and keep well image detail at the same time.2679|203|0更新时间：2024-05-07
- 摘要：Image mosaicing has many applications, such as image based rendering, video retrieval ,scene matching, etc. Aiming at images with rotation and scale, a method of image mosaicing based on log polar coordinate mapping is put forward to obtain scene representation of wide view field .In this method, firstly each image is transformed from Cartesian space to log polar coordinate space, which causes scale and rotation in Cartesian space to be manifested translation in log polar coordinate space, and the translation between the pair of input images is calculated ,then phase correlation method is used to estimate rotation angle and scale factor between the pair of input images, and then these motion parameters are refined by nonlinear least square iterative optimization algorithm, lastly image blending is imposed to implement image mosaicing. The experiments show that it is an available method of image mosaicing.2930|193|0更新时间：2024-05-07
- 摘要：To the problem of shape from shading, the existence of self shadowed regions in the image is one of the factors which could cause the deformation to the final 3D reconstructed surface and it should be coped with in its special way. In the paper, a set of method is presented, which is composed of shadow region segmentation, coordinate system transmission, and the parabolic data interpolation. The method is based on the principle that the distortion of self shadow should be enclosed within its region, in order not to destroy the global reconstructed shape. The light source projected system is introduced, which is defined on the image plane and the direction of light source. This can greatly simplify the problem, and makes it possible to analyze the self shadowed data sections in the image vow by vow. And the primary goal could be achieved by recovering the depth gaps across the self shadowed data sections. Meanwhile, the height incremental values at each shadowed pixel could be estimated by parabolic interpolation, and makes the recovered shadowed shapes connect with other regions smoothly. It has demonstrated that, this proposed method could improve the reconstructed 3D shapes from shading efficiently.2453|189|0更新时间：2024-05-07
- 摘要：In order to realize fast,high quality enlargement of image, this paper introduces a new approach to achieve the high order interpolation and realize the enlargement of images by means of hardware. In order to ensure high quality of the enlarged image, the paper adopts the cubic B spline interpolation to calculate the gray of pixels of the enlarged image. What is more, the paper develops an IIR(infinite impulse response) based and FIR(finite impulse response) based digital filter to realize fast cubic B spline interpolation. In order to realize different multiple enlargement of image in the system, the dynamic reconfigurable design based on FPGA(field programmable gate array) has been applied in the system, which can real time reconfigurate FPGA so as to achieve the different arithmetic. And it designs a fast and high quality image enlargement and displays system based on dynamic reconfigurable FPGA for the images in 256 gray levels.2471|227|0更新时间：2024-05-07
-
- 摘要：According to the position similarity of corresponding points in multiimages of correlation based on computer vision, a corresponding pointmatching rule based on“the regulation of the minimum summation of euclid distance”is presented when studying the problem of corresponding pointsmatching from X2ray images. This rule is different from the conventional corresponding points matching methods based on gray level and based on regional geometric feature. It employs the model of sequence matching algorithm. According to the condition that the relative position of corresponding points in two images are almost unchanged in adjacent area, this rule minimizes the summation of corresponding points distance to match the corresponding points by adjusting the sequence of feature points array with evolutionary programming searching algorithm. By the experimenton PC, the result demonstrates that this algorithm canmatch the most of corresponding points correctly in a low time cost just based on the position similarity of corresponding points. The idea of this rule is summarized and its application is generalized in the end.2337|194|0更新时间：2024-05-07
- 摘要：Based on log polar mapping, five filters have been designed and applied in this paper. In order to check the digital watermark embedded in digital images, we compared the classical matched filter, amplitude only filter, phase only filter, binary phase only filter and a new filtering method. Experimental results demonstrate that the new filter, phase only filter and binary phase only filter can be used to identify the digital watermark in the digital images when there is no rotation, scaling and translation for the images embedded digital watermark; Suppose the image embedded digital watermark is rotated ten degree and scaled down ten percent, it can be found the classical matched filter, amplitude only filter, phase only filter and binary phase only filter can't be used to identify the digital watermark in the digital images ,but this new filtering method can still be used to identify the digital watermark in digital images.2212|248|0更新时间：2024-05-07
- 摘要：In order to realize secret transmission of great color video information. This paper presents a new scheme capable ofhiding low levelMPEG2Ⅱcolor video streams intomain levelMPEG2Ⅱcolor video streams. Itmakes fulluse of vision character of our eyes and selects information of intermediate/high frequency as revisable host vector in the DCT domain ofhostvideo streams, which are notsensitive to our eyes. A novel embeddingmethod is proposed in this paper. As many as Lbits of information data can be embedded by changing, atmost,Lbits in the(2L+1)bits cover data, which makes higher dataembedding efficiency and better secrecy. The scheme can be also used for secret transmission of other stream media and great information. The scheme has been simulated in Local areaNetworks and better experimental results have been given.2379|198|0更新时间：2024-05-07
- 摘要：In order to eliminate the effect caused by noise, complex motion, and uncovered background in the process of real time segmenting video objects, a new method for real time segmenting video objects based on statistical change detection is proposed. During the process of statistical change detection, because t distribution is used to eliminate the effect of noise, the noise variance is not needed. Two frames at a distance of k are used for statistical change detection instead of two successive frames, thus articulation motion and slow motion can be better coped with. The intersection between two successive statistical change detection results can eliminate the effect of uncovered background, and most of the residual noise are eliminated at the same time without adding any computation complexity. The experimental results show that the new method solves the problems existing in the process of traditional statistical change detection caused by noise, complex motion, and uncovered background, and it can automatically segment video objects by real time. It can meet the requirements of many object based video applications such as MPEG-4.2623|258|0更新时间：2024-05-07
- 摘要：After analyzing the principle and the shortcoming of detection and parameters estimation for the multi component Chirps signal based on the bilinear time frequency distribution Hough transform, an improved method has been proposed in this paper, which the bilinear time frequency distribution of the multi component Chirps signal is regarded as an image, and ridges extraction operation is applied to the image and then the principle of detecting the line of Hough transform is used. The simulation results have shown that the proposed method not only can eliminate false detection resulted from pretense peak value of the cross term and shorten the time of executing Hough transform, but can also increase the accuracy of detection and parameters estimation and reduce noisy interference.2681|198|0更新时间：2024-05-07
- 摘要：Multiwavelets are developed in recent years. They are a new addition to the body of wavelet theory. They preserve many properties of single wavelets while overcoming limitations of single wavelets. In practice, they perfectly combine smoothness, short support, orthogonality, symmetry.ect. The existing algorithms of multiwavelets denoising only provide theories of denoising but there is no concrete performing method. Hence, multiwavelets are combined with HeurShrink denoising method in this paper. A feasible denoising method ,called MDH, is proposed and the denoising performances of D4 scalar wavelet are compared with that of MDH. The experimental results show that MDH is superior to D4 scalar wavelet in denoising via HeurShrink.2463|190|0更新时间：2024-05-07
- 摘要：Spatial database has become one of the dominating techniques tomanage spatialdata in geographical information systems, and geographic structured query language (GSQL) is the key issue ofspatialdatabase design and implementation. Atpresent, GSQL for vector data has been widely studied. However, when considering the raster data, little researchwas made for their specifications on visiting and operationa.l Based on OpenGIS○R, a GSQL extension named GSQL2R supporting raster data is discussed in this paper. At first, some abstractdata types (ADTs), includingPixe,l RasterRegion and RasterCoverage, are defined. The relationships between Pixel/RasterRegion and RasterRegion/RasterCoverage are part/whole. In these definitions, data objects and operations for each ADT are described in a formalized way. Then, GSQL2R syntax and some instances are presented. The GSQL2R description includes three parts, .i e. data definition language (DDL), data table schema and data manipulation language (DML). The example indicates that GSQL2R provides an efficient and integrated way for raster data access and operation.2444|190|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