最新刊期
卷 11 ， 期 3 ， 2006
- 摘要：Image segmentation is a classical and crucial problem in the fields of computer vision and image understanding. This paper gives a review on the variation based active contour model and level set method developed in recent years for image segmentation. The basic ideas of three types of active contour models, i. e. , edge based, region based and edge- region based models, are presented, their advantages and disadvantages are summarized, and a number of improvements are analyzed in detail. The level set method, which is numerically stable and capable of describing the topology change of the contour, is briefly introduced as an advanced numeric algorithm to solve these models. Finally, the potential application of active contour in image registration is discussed.关键词：computer vision;image segmentation;level set;curve evolution;active contour;snakes model;Mumford-Shah functional;image registration16562|3210|0更新时间：2024-05-08
- 摘要：In terms of the special gray distribution in retinal images, a novel blood vessel extraction method based on adaptive thresholding is proposed in this paper. The whole image is divided into many small sub-images with identical dimension, and u threshold is calculated respectively in each sub-image for segmenting local blood vessels. Because both vessels and background are locally uniform in retinal images, there must be a threshold which is able to segment vessels precisely in a certain sub-image. The method employed for determining the local threshold not only allows sub-images to be very small, but also ensures the threshold to be optimal in the sense of least square error. A new edge tracking algorithm based on zero-crossing edge detection technique is applied in the process of threshold computing. Further more, a feature synthesis method based on region growing is presented, which is used to clear fragments in results of adaptive thresholding. The experiments on many retinal images indicate that this blood vessel extraction method is computational efficient and can extract most blood vessels including very small blood vessels.关键词：blood vessel segmentation;adaptive thresholding;zero-crossing;edge tracking;region growing4800|536|0更新时间：2024-05-08
- 摘要：Because the representation of the spatial data is related with scale characteristics, element feature extraction is meaningful only when it is under certain scale rcpectively. Different sizes of structure element are suitable for image features on different spatial scales. It cannot obtain the optimal effects can not be obtained if the size of structure element. Therefore, the design of size of structure element plays an important role on extracting image information. Based on the analysis between the mathematical morphology characteristic and the size of structure element, the optimal quantitivc relationship is acquired, between the size of structure element, the size of feature factor and special scale of feature factor denotes size of structure element is unsuitable, A denotes size of feature factor and denotes spatial scale of feature factor. Then the adaptive algorithm for ascertaining the optimal size of structure element is designed by calculating and comparing the difference information between the neighborhood spatio-scale gradient images. And show experimental results the correctness and effectiveness of the algorithm. In the end, the difference between the extraction result by the classical edge operators and the operator presented in the paper are listed and analyzed,then the conclusion that the operator is superior to other edge operators on robust, position and contrast is reached.关键词：morphological gradient;size of structure element;adaptive algorithm;marine geographical information system(MGIS)4739|629|0更新时间：2024-05-08
- 摘要：Automatic text location in images plays an important role in image content understanding, and draws attentions of researchers in the domain of computer vision. Current text location algorithms are mostly adaptive to specific applications; they are sensitive to the variation of text or images and lack robustness. This paper presents a universal approach for text location based on histogram features and AdaBoost. The new algorithm extracts histogram features, which have strong discriminabilities for text and non-text. AdaBoost algorithm with cascade structure is introduced to design the classifier for text texture. The algorithm transfers the binary output of the texture classifier into probability form and generates corresponding text probability image. CAMSHIFT algorithm is used to search for the final location result in the text probability image. The experimental results demonstrate the robustness of the proposed algorithm, which is adaptive to the text of different languages, fonts or scales, and gets promising location results in variant types of images.关键词：text location;histogram features;AdaBoost;text probability image;continuously adaptive mean shift(CAMSHIFT)3939|542|0更新时间：2024-05-08
- 摘要：This paper presents a method of feature matching based on Laplacian spectral of graphs. Given feature points of two images, we define Laplacian matrices respectively, analysis the eigenvalues and eigcnvectors of the matrices, and construct a feature points matching matrix with information of magnitude and position of entries in the matching matrix, the feature points matching is done. Furthermore, we theoretically prove that our algorithm can acquire an exact matching under an equilong transformation or equiform transformation on images. Experimental results show that the approach attains accuracy 82% on real images.关键词：matching;feature point;graph;Laplacian spectrum3754|335|0更新时间：2024-05-08
- 摘要：Image registration is a basic problem during image analysis and image processing. Image registration has been widely applied in the fields of medical image, remote sensing and computer vision. This paper investigates two registration methods for ultrasound image relying on the intensity-based similarity measure. In tiw first method intensity information is provided by feature points which have been extracted using Harris corner detector. The registration similarity measure is then defined as a cost-function-error cost function. In the latter method the same cost-function is employed, but uniqueness control and region correspondence are different from the first method. Given this similarity measure, parametric ultrasound image registration is stated as a minimization issue. The experiment even exploits polynomial technique to transform the whole image dataset and estimates the sum of square error in the first method. Image quality is one of the most important issues in ultrasound imaging. Ultrasound images tend to be noisy due to the presence of speckle. Speckle originates from the constructive-destructive interference of the ultrasound pulses, and is produced to some extent by all ultrasound machines. This paper shows that it is possible to perform accurate registration on data from in-vivo examinations. The result indicates that the two methods are robust and meet our requirements, but the latter outperforms the first one. It can match B-scans and areas with certain angel, severe deformation and even with plus noise.3328|399|0更新时间：2024-05-08
- 摘要：The epipolar geometry is the fundamental and important constraint between stereo image pair. It is independent of scene structure and plays a very important role in feature correspondence, Homography characterizes the correspondences between the two views due to the same plane scene which is extensively applied in plane scenes. To avoid the degenerate configuration when directly computing the homography given the epipolar geometry in curve scene, the area-detection approach is presented in this paper. In addition ,the constraint relation between homography matrix and fundamental matrix is involved in the homography estimation. The two schemes can greatly improve the matching precision and efficiency of the algorithm, Experimental results with real image data have illustrate the performance of the schemes.关键词：image rectification;feature correspondence;weakly calibrated;homography;epipolar constraint2782|502|0更新时间：2024-05-08
- 摘要：In this paper,a fusion algorithm for enhancing the spatial resolution of MODIS images is proposed,based on the joint use of IHS representation and orthogonal wavelet decomposition(OWD).The basic idea consists of following steps represent the MODIS images in the IHS space,substitute the I with the high resolution intensity component obtained from the fusion with the approximate component of the SPOT image through an intensity correlation comment model based on the self-adaptively weighting(SCMM).Finally,three high-resolution MODIS images are achieved through an inverse OWD by using the three modulated MODIS images obtained by an inverse IHS transform as the approximate component and the details of the SPOT image as the high frequencies.Experiments confirm that the proposed algorithm can achieve better performance in terms of both preserving the spectral information and improving the spatial resolution of MODIS images visually and statistically.关键词：orthogonal wavelet decomposition(OWD);intensity correlation moment;image fusion;moderate resolution imageing spectroradiometer(MODIS);SPOT4505|491|0更新时间：2024-05-08
- 摘要：To fuse different multi-focus images effectively, an adaptive fusion method of multi-focus images based on regional features in wavelet domain is proposed. First, images involved in fusion are decomposed by wavelet transform. For the fusion of low frequency coefficient, it adds different feature to the result image based on reserving the common feature of fusing images. High frequency coefficient is fused according to the wavelet energy of its region. Finally, the fused image is reconstructed. This method can not only fuse multi-focus images adaptively, but also have versatility in multi-focus images. The results show that this algorithm has good fusion effects, so it is an effective method in multi-focus images fusion.关键词：image fusion;adaptive;wavelet transform;multi-focus3291|387|0更新时间：2024-05-08
- 摘要：It is a crucial problem how to produce two pictures for left-eye and right-eye respectively with proper horizontal position disparity in autotereoscopic 3D display. The theoretic interrelation between the left-viewpoint and right-viewpoint in the model space by Direct 3D and the horizontal position disparity in the display screen has been established in this paper. The method for getting proper horizontal position disparity has been put forward aiming to provide proper sensation of depth as in real world and the most comfortable vision, and the theoretic formula is presented as well. Furthermore, two parameters, the position of left-viewpoint and fight-viewpoint in model world and the movement of the perspective plane, have been given in this paper, which directly affects the view result, The conclusions from this paper have gone into application and marvelous auto-stereoscopic 3D view has been achieved,3384|351|0更新时间：2024-05-08
- 摘要：Ghost artifacts occur in EPI reconstruction because odd and even lines of k-space are acquired with opposite polarity,and experimental imperfections such as gradient eddy current,B_0 field inhomogeneity,susceptibility,and chemical shift result in different phase offset between the even and odd echos.The method for removing ghost artifacts is based on the phase retrieval.First,a ROS(region of support) is specified using morphological method and ghost artifacts outside ROS are set to zero.Then a model image can be obtained.According to the criterion of entropy in the imaging field,the algorithm for several iterations is applied to obtain an acceptable image.It is shown that applying this method will reduce the ghost artifact evidently.The method has been tested using computer simulation and successfully removed most of ghost artifacts.The experimental results show the effectiveness of the method.关键词：echo planar imaging;ghost artifact;magnetic resonance imaging(MRI);phase retrieval3189|336|0更新时间：2024-05-08
- 摘要：The automatic skew-correction of scanned color document image is a necessary step undergone before some processing such as OCR(optical character recognition).For the purpose of accurate and efficient automatic correction,two algorithms,namely skew-detection and skew-adjustment,are proposed.The skew-detection algorithm,which is based on the analysis of texture complexity of document image with homogeneous background,solves the problem of how the accurate skew angle can be found.Besides the rotation transform,the skew-adjustment algorithm includes a dual linear interpolation algorithm for color compensation based on linear dependence among the colors in the neighborhood to accomplish the skew-correction of a color image and the elimination of the "Saw tooth" phenomenon resulted from the integer-operation during the coordinates transformation.In the end,the authors illustrate their experimental results,which show that the presented algorithm is exact,efficient,and practical for color documents whose background are not complex.关键词：color document image;skew;detection;automatic correction;texture complexity;color compensation5067|610|0更新时间：2024-05-08
- 摘要：This paper proposes a robust algorithm of facial expression synthesis.Decompose a train set into person subspace,expression subspace and feature subspace by employing higher-order singular value decomposition(HOSVD),and then map them to expression subspace,which can be used to synthesize an expressive image of arbitrary strength,given any frontal facial image.By warping source image to generate the target image rather than by linear-combining the images in train set as usual,this approach not only lessens the train set data storage and the computation complexity,but also enables the system to cope with facial images of arbitrary size,background,illumination condition,coloration or pose.With square interpolation,expression facial images of any assigned strength can be obtained.Experiments prove that this algorithm has higher efficiency and can generate very excellent images.关键词：expression synthesis;higher-order singular value decomposition(HOSVD);expression decomposition;shape3301|347|0更新时间：2024-05-08
- 摘要：In paper the problem of removing inner knots from the knot sequence of a B-spline curve is discussed, where the theory of generalized inverse matrix is applied and then the necessary and sufficient condition for a inner knot of the Bspline curve being removable is obtained, based on which an algorithm of removing multi-knots is proposed. Experimental results show that the algorithm is reliable and efficient, hence can be applied to practice.关键词：Moore-Penrose pseudoinverse matrix;B-spline curve;knot removal3204|313|0更新时间：2024-05-08
- 摘要：The Logistic-Map chaotic sequence is often used to scramble and encrypt the digital images.In the former papers,the invalid-keys and the quasi invalid-keys existing in the Logistic-Map have not yet been discussed.If using the invalid-keys or the quasi invalid-keys as the initial value of the Logistic-Map,we can't get the chaotic sequence to scramble the digital images.Furthermore,there are infinite invalid-keys and quasi invalid-keys in the Logistic-Map.So we should take the problems seriously.In this paper,the invalid-keys and the quasi invalid-keys existing in the Logistic-Map chaotic sequence are deeply discussed.Then,an image-scrambling algorithm based on mixed chaotic sequences is proposed.Using this algorithm,the invalidity of the invalidkeys and the quasi invalid-keys is avoided thoroughly.关键词：chaos;mixed chaotic sequences;scrambling transformation;encryption;wavelet;invalid-key;quasi invalid-key4539|561|0更新时间：2024-05-08
- 摘要：A new universal blind steganalysis scheme is presented based on analysing the limitation of Farid's universal blind steganalysis on the dimension of the feature vector,the validity of the single feature,and the correlation of the features.Preprocessing is proposed using principle component analysis on the image statistics features and steganalysis classifier is constructed using RBF network.The scheme not only reduces the dimension of the feature vector enormously,but also improves the performance of the detection.Apply our scheme and Farid's scheme to detecting the stego images produced by JSteg,EZStego,and S-Tools respectively,and the comparison of these simulation results shows that after the preprocessing using principle component analysis the dimension of the feature vector in our scheme decreases 174(Jsteg),163(EzStego),180(S-Tools),and therefore simplifies the design of the steganalysis classifier.Furthermore,our scheme is quite more efficient because the stego image that the proportion of the embedding message to the maximal embedding capability is more than 60%(Jsteg),80%(EzStego),50%(S-Tools) can be detected efficiently by our scheme.关键词：steganalysis;universal blind steganalysis;principle component analysis2759|351|0更新时间：2024-05-08
- 摘要：Because the watermark lengths of many applied vector maps' polygonal lines are short,and the Gaussian distribution is not suited to describe magnitudes of discrete Fourier transform coefficients for vector maps' polygonal lines to difference polygonal lines in the vector maps,the uniform threshold of normalized-correlation-based watermarking detection is hardly confirmed.In this paper,to improve algorithm's efficiency and be invariant to simplification,the discrete Fourier transform is only used for the feature points of vector maps.The watermarking algorithm embeds the watermarks by changing both the magnitudes and the phases of discrete Fourier transform mid-frequency coefficients for vector maps' polygonal lines.The correlation coefficient of the detected watermarks and the embedded watermarks is regarded as the watermark detection value.Given applied map data,computing results show that watermarks generated by this technique are robust,and the algorithm is fairly good in efficiency and Feasibility.关键词：vector map watermark;discrete Fourier transform(DFT);normalized correlation detection;feasibility analysis;feasible robust watermarking algorithm2851|357|0更新时间：2024-05-08
- 摘要：The conflict between data provider and demander has given a direct impetus to the research on WebGIS.Considering the problem of implementing the participation and interoperation of geospatial data by using 3D visualization method,the authors construct a new sub-assembled WebGIS architecture,namely Geo-Spatial Data Sharing and Handling Integrated Platform(Geo-SD SHIP),which regards X3D/XML dataflow as the connection among heterogeneous geospatial data.The characteristics of Geo-SD SHIP are explained and its architecture is diagrammatized.In addition,the key technology to implement Geo-SD SHIP was demonstrated,which includes: 1) to create Web-3D virtual geographic scene based on X3D/XML;2) to construct X3D/XML components for geospatial data exchange and GIS spatial analysis;3) to implement parser components of X3D/XML data by utilizing VC++.Net and OpenGL API libraries.Geo-SD SHIP has had successful application in the project of National Science Foundation(Grant No.40471134) and the CAS pilot project of Knowledge Innovation Programs(Grant No.KZCX3-SW-326-03 and KZCX3-SW-327-01),and experimental results demonstrate that Geo-SD SHIP can achieve the participation and interaction of heterogeneous geo-spatial data on the Web.关键词：geo-spatial data;virtual reality;WebGIS;extensible threedimensions(X3D);extensible maxkup langnage(XML)4147|525|0更新时间：2024-05-08
- 摘要：The shortest path problem, in which label algorithms are of consequence, is a research topic in the field of geographic information science and computer science. Among label algorithms, label setting algorithms in which Dijkstra occupies the core position are always considered as the first choice in applications while label correcting algorithms are rarely used. In this paper, followed by the theory expatiation on label correcting algorithms, an optimization of the wellknown Pallottino algorithm is set forward on the basis of both storage and implementation structures. The author then analyzes its time complexity and spatial complexity and finally tests its actual efficiency in Beijing transportation network, It proves a better implemental efficiency and applicability comparing with the best label setting algorithm- Dijkstra algorithm implemented with approximate buckets, As a result, it provides a good choice for network related analysis works and should be particularly useful to researchers and practitioners in operations research, transportation, and GIS,3325|273|0更新时间：2024-05-08
- 摘要：JPEG2000 stands for the state-of the-art of still image coding based on discrete wavelet transform(DWT),and H.264/AVC stands for the state-of-the-art of video coding based on variable block sizes discrete cosine transform(DCT).In addition,the H.264/AVC I-frame encoder(H.264-IFE) adopted several advanced techniques,such as directional spatial prediction,variable and hierarchical block transform etc,which improve its coding efficiency significantly.Basically,H.264-IFE is also the state-of-the-art of still image compression algorithm based on block transform.In this paper,a performance evaluation between H.264-IFE and JPEG2000 in terms of PSNR is presented.The experimental results show that H.264-IFE achieves better reconstructed image quality than JPEG2000.These results demonstrated the H.264-IFE's advanced techniques such as directional spatial prediction and variable block-size DCT coding are potential methods for developing next generation still image coding standard.4189|396|0更新时间：2024-05-08
- 摘要：In this paper, a now color image compression algorithm in DWT domain is proposed, in which the visual masking and the correlation between color components are utilized. Firstly, the discrete wavelet transform is performed on three different color components in RGB color space, and the wavelet coefficients are weighed based on the human visual masking. Then weighed wavelet coefficients are converted to user’s defined Y′U′V′color space. Finally, the wavelet coefficients are progressively encoded in Y′U′V′color space by the extended zerotree structure and the improved EZW scheme. Experimental results show that the new color image compression scheme performs better than that of CEZW and CZW in the aspect of recovery image quality and coding/decoding time.关键词：color image compression;embedded zerotree wavelet(EZW);visual masking;correlation between color components2986|346|0更新时间：2024-05-08
- 摘要：As Wireless communication channels exist noise,single-bit error and burst error containing N successive error bits may occur during JPEG-coded transmitting image over wireless communication system.Therefore error detection and correction play a very important role in transmission joint photographic experts group(JPEG) images.JPEG image is sensitive to transmission error.A single transmitting error in codeword will affect subsequent codeword and may result in a significant degradation.Error concealment and self-synchronizing codes are traditional methods to cope with this problem.The basic theory of error conecalment method is interpolation.So it will degrade image quality in some measure.On the other hand self-synchronizing codes need a specific coder.It will reduce the compatibility of JPEG images.For the sake of better decoding performance this paper proposed a novel algorithm for transmission error detection and correction.Base on image correlation,the proposed algorithm correct transmissive error with full search method.The paper also analysed the practicability of the algorithm.Based on the simulation results obtained in this study,the proposed approach is able to fit all the single-bit transmission error and improve the quality of JPEG image greatly.关键词：JPEG image;error detection and correction;image correlation;bit inversion3029|336|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