最新刊期
卷 8 ， 期 10 ， 2003
- 摘要：The visual-characteristic-based compression technique breaks through the restriction of the traditional image compression technique that uses MSE as criterion, and combines the human visual characteristic with the coding arithmetic, in order to achieve the lowest bit rate and the best perceptual image quality. Now, much headway has been made in the research of this technique, but there still are many problems to be researched on. This paper systematically introduces the existing methods of this compression technique, and discusses some visual coding techniques that applied in JPEG-2000 and MPEG-4 standards. After researching on the techniques. Some results and conclusion are gaven as follow:perceptual measurement of the image quality, automatic ascertainment of the region-of-interest and recognition and segmentation of the arbitrary object are the main problems existing in the visual-characteristic-based coding technique, and they are also the research direction for the future.2381|184|0更新时间：2024-05-07
- 摘要：Recently, SAR (Synthetic Aperture Radar) is evolving to become an indispensable tool for earth observation and military reconnaissance. The collection capacity for SAR images is growing rapidly， and along with that growth is the expanding need for automated or semi-automated exploitation of SAR images accurately and efficiently. Automatic target recognition (ATR) is an important aspect of automatic or semi-automatic SAR imagery interpretation. An ATR algorithm finds target like regions (regions-of-interest (ROIs)) within a SAR image and computes a class for each ROI. The meaning of SAR ATR is explained, and the general architecture of SAR ATR system is introduced in this paper. SAR ATR systems are analyzed and classified according to their classification methods. The main difficulty of SAR ATR is analyzed, the present state and the future direction of SAR ATR research are also introduced.关键词：Image processing;Synthetic aperture radar (SAR) image;automatic target recognition (ATR)3066|231|0更新时间：2024-05-07
- 摘要：This work investigates the image registration from feature point sets. Image registration is a fundamental object recognition method in computer vision and it aims to find best matches between two or more point sets when there are geometric distortions, point measurement errors and contamination present. Up to now > closed form solution has been developed only when the geometric distortion is similarity transformation. This paper concentrates on image registration from feature point sets when the geometric distortion between the two images is affine transformation and gives closed form solution for the transformation parameters that minimize the root-mean-squared residual error of the image points by the linear least-squares techniques and the pseudo-inverse of matrix respectively. In order to give the simple closed form solution, the image points are represented by homogeneous coordinates and the theories of matrix are used. The algorithms are evaluated on both synthetic and real world images and the experiment results show that the methods given in this paper are accurate, stable and are only affected slightly by noise.关键词：Computer image processing;image registration;affine transformation;Least-squares;Pseudo3243|179|0更新时间：2024-05-07
-
De-noising in Color Image Based on The Noise-Adjusted Principal Components Analysis and Image Fusion
摘要：De-noising in color image using different filter will get the different image quality. In this paper, we analysis the filter feature of the classic filter,get the filtered illumination image with the median filter and the wavelet filter,and then merge the filtered illumination image with the color image after vector median filter to get the result image. Experiment result shows that the fused filter can suppress the pulse noise,Gussian noise and the mixed noise. Compared with images by vector filter and-TMF, the fused filtered image has the least normalized peak-value mean square error. It approximately reflect the real colors of nature object like the origin non-noise image.关键词：Computer image processing;image fusion;Color image filter;Noise-adjusted principal components analysis2349|192|0更新时间：2024-05-07 - 摘要：Geometrical distortion in optical lens imaging is inevitable, and the polynomial coordinate transform algorithm is useful in rectification of image geometrical distortion, but it takes too much computational time to be applied in real time image processing systems when high order polynomials are employed on condition that complex nonlinear distortions are encountered. To decreases the computational complexity, an approximation algorithm is proposed in this paper, in which the image is firstly divided into rectangles, and then an appropriate first-order polynomial is used to approximate the high order polynomial within each rectangle. Based on the analysis of the radial distortion, which is considered to be the main kind of image distortion, a novel image partition method is presented to ensure minimal memory expense for model parameters. With the approximation algorithm two-third of the computational time is saved and the memory expense is also limited within a reasonable range. Experimental results indicate that the algorithm proposed in this paper is effective and valuable in the engineering applications.3444|229|0更新时间：2024-05-07
- 摘要：Edge detection is the base of feature extraction, analysis and comprehension of image, the quality of which determines the results of subsequent processing. Thus it is an effort goal for people to find a kind of method that is insensitive to noise, precisely locates true edges and excludes false edges. This paper discusses and tests a novel algorithm of edge detection based on gray system theory. At first the characteristics of pixels are analyzed to decide non-edge referential sequence and the sequence to be compared, and then gray correlation degree of both sequences is employed to distinguish between non-edge pixels and edge pixels. Finally, the simulations prove that in binary image cases, the algorithm can obtain more precise edge pixels than some traditional edge detection algorithms, such as canny, log, sobel and zerocross, and tolerate some noise such as speckle, salt and gaussian. In grayscale image cases, it is unnecessary to change the referential sequence. Furthermore the quantity of edges can be controlled by adjusting the correlation degree threshold. So the technique is a new effective adjustable method of edge detection.关键词：Computer image processing;edge detection;Gray system theory;Gray correlation degree2953|292|0更新时间：2024-05-07
- 摘要：There are many applications of the robust eye states detection. In this paper, five methods are applied to detect the eye states in color eye images. They are gray model method, fisher method, projection method, detect-upper-eyelid method and detect-eyeball method. Each of the five methods has its advantage as well as its disadvantage. To increase the accuracy of detection, a combined method is proposed, which combines the above five methods so that it can take good advantage of them all. Experiments are done with 396 different human eyes, 264 of which are closed ones, and 377 of these experiments turned out satisfactory results. The accuracy reaches 95. 2 %. The inspiring outcome shows that the combined method is effective even if the illumination is nonlinear. At the same time, the combined method is practical because it is moderate in both time and space consuming. This paper introduces in details the idea of the five methods, the design of the combined algorithm, the experiments and their relevant results.关键词：Computer image processing;Eye states;Histogram equalization;Horizontal projection;Hough transform2778|213|0更新时间：2024-05-07
- 摘要：To process the nonlinear property of ultrasound inverse scattering image, one should alternately solves the well-posed forward scattering equations for an estimated total field and the ill-posed inverse scattering equations for the desired object property function. Forward scattering equations can be solved by common method while inverse scattering equations is ill-posed and should be regularized. For ill-posed inverse scattering equations, very little perturbation in data will cause great change in the solution. So the iterative procedure depends strongly on the precision of the solution of ill-posed inverse scattering equations. Previous work on the ill-posed inverse scattering equations commonly used Tikhonov regularization which by adding small filter factors to original least squares problem and can't filter noise efficiently. The method for choosing regular parameter is difficulty in Tikhonov regularization because the parameter is continuous. This paper adopts the truncated singular value decomposition (TSVD) method to solve the inverse scattering equations which can filter noise better than Tikhonov regularization. Since the regularization parameter is an integer in TSVD method, it can be revised by an appropriate method. Different 'images with different structure are simulated by truncated singular value decomposition method equipped with a revised parameter choosing strategy. Simulation results show that this method associated with a good approach for choosing regular parameter can efficiently filter noise, and hence the quality and reliability of the reconstruction image can be improved. At the same time, this method can decrease computations at the iterative procedure.关键词：Computer image procesing;ultrasound;Inverse scattering;Image reconstruction;singular value decomposition;regularization2739|216|0更新时间：2024-05-07
-
The Performance Evaluation Based on the Grey Multi-hierarchical Appraise Model of Image Segmentation
摘要：Image segmentation is the key component in the automatic target recognition system (ATRS). It's an important step of image analysis. Its results have critical implications for the information processing stages that follow: The evaluation of a segmentation algorithm makes an important part of the estimation of performance for are as an off-line ATRS; Based on the knowledge of data fusion and systems engineering, a new model of gray multi-hierarchical appraise by combining gray appraise method with the Analytic Hierarchy Process (AHP) and data fusion is proposed in this paper. It is used to evaluate the performance of image segmentation. The performance evaluation can be performed from both global and local aspects. The paper also discussed evaluation models and their experimental comparative results to prove its feasibility and to confirm the advantages of the new model. The application shows that the new model is more efficient and reasonable than other methods.关键词：Computer image processing;Grey relational grade;analytic hierarchy process;Data fusion;automatic target recognition;segmentation;Performance evaluation2400|172|0更新时间：2024-05-07 - 摘要：Independent Component Analysis (ICA) is a new development of signal processing. As an effective approach to the separation of blind signal, Independent Component Analysis has attracted broad attention and has been successfully used in many fields. The fundamental, discrimination condition and practical algorithm of Independent Component Analysis are discussed. Then, a fast Independent Component Analysis algorithm (FastICA) is introduced, and it is known that the time-consuming course is computing Jacobian Matrix. Reducing the time of Jacobian Matrix will improve the performance of FastICA algorithm. So a modified FastICA (M-FastICA) algorithm is developed. By modifying kernel iterate course, several iterations of FastICA are merged into one iteration of M-FastICA, then M-FastICA algorithm only need to compute Jacobian Matrix once time and achieves the correspondent effect of FastICA. So the convergence of ICA will be accelerated. Finally, M-FastICA is applied to image separation. The experiment images are mixed with a random matrix. Independent Component Analysis can separate the mixed images and obtain the approximate of source images. The experiment results show that the iterations of serial modified algorithm reduces 9 percent, and the iterations of parallel modified algorithm reduces 27 percent with the correspondent separation performance.关键词：Computer image proceesing;independent component analysis;Fixed-point;FastICA;M-FastICA;Image separation2807|184|0更新时间：2024-05-07
- 摘要：An improved approach of estimating fractal dimension by partition iterated function system (PIFS) is proposed, which is used to analyze the different texture surfaces of bone marrow granulocytic cell nuclei in the different maturation stages. Based on contractive affine transformations, a PIFS for cell nucleus gray-level image is constructed, scale coefficients, domain factors and a scale factor are obtained from PIFS and used to calculate the fractal dimension. In this paper, the evolution of computational methods of domain factor and scale coefficient is used to overcome fractal dimensions overestimated. In order to testify the validity of the proposed method, there are 60 images of different types of cell nuclei to be applied to the experiment. The results show that fractal parameters obtained by our method are able to differentiate with the different types of cell nucleus texture surfaces efficiently, they can be used to the recognition of cells as new features.关键词：Computer image processing;Partition iterated function system(PIFS);fractal dimension;texture analysis;Cell nucleus2553|169|0更新时间：2024-05-07
- 摘要：The security of digital images attracts much attention recently, and many image encryption methods have been proposed. Among them, the ID chaotic image encryption method was reported with appropriate performances by using the appropriate complexity and pseudo-randomness and extreme parameter sensitivity of chaotic sequences but with smaller key space as its serious drawback compared with others. So a new image encryption algorithm based on hyperchaotic sequences is studied. Several fast simple 2D hyperchaotic systems are given and studied. A linear approach of mapping from the model parameters to the key is proposed in order to make full use of the key space and keep the correspondency between the key and the model parameters used for encryption, The chaotic sequences generated by these models are mapped into the encrypt sequences, then the image is encrypted by the encrypt sequences by using two directional XOR horizontally and vertically. Experimental results show that this method is effecient, witch can provide faster encryption and larger key space and stronger anti-decryption-ability and rubustness to local breakage compared with the 1-D chaotic image encryption.关键词：Computer image processing;Image encryption;hyperchaos;two dimensional discrete systems;Lyapunov exponents2914|188|0更新时间：2024-05-07
- 摘要：Digital watermarks have been proposed in recent literature as a means for copyright protection of multimedia data. The robustness and the imperceptibility are the basic requirements of the digital watermark. In this paper, we propose a digital watermarking algorithm based on Shamir secret sharing scheme. We divide the field of the image into parts, and then we also make the information of digital watermarking into shares by the algorithm. We embed each share of the digital watermarking in the corresponding part of the image by contrasting DCT coefficients， and only need arbitrary shares in the recovery progress. In order to enhance the precision of the extracted watermark, we improve the extracting algorithm based on fuzzy set. Experimental results show that this algorithm has a good hiding effect and is robust for single attack and even collage attacks.关键词：Computer image processing;digital watermarking;Secret sharing;Fuzzy set2636|197|0更新时间：2024-05-07
- 摘要：The paper deseibes a new type of feature information card-images feature information card based on self-drawn figure by user(IIC).The Paper construets the mathematical models which are fit to the relation between geometries true ture of dynamic sampling IIC installing strueture of sensor;According to the method of vertical and horizontal inclining gradually，we take advantage of infrared double-diodeto arrange IIC pattern information and make use of software and hardware to collect it alternately and dynamically;The PaPe reonstruets electric eireuit models of sampling system of IIC;The paper proposes various effective algorithm storestore primitive double-value electronic image twisted for need of sampling to the original electronic image.At the same time，the sampling algorithm of IIC in combination with software and hardware reflectvivid features.IIC has some outstanding advantages，sue has eonvenient applieation，cheap hardware cost，ingenious bulk，distinet and reliable result integrated software with hardware.Because of the advantage of self-drawnfigurebyuser，IICwouldhave broad developing and applying prospect in lots of domains，sue has door-lock，authority management，pattern reeognition and its manipulation，etc.关键词：Computer image processing;Image feature information card (IIC);Self-drawn figure by user;sampling;Electric circuit models;Algorithm2550|183|0更新时间：2024-05-07
- 摘要：With the promotion and application of digital imaging technology in the medical domain, the amount of medical images grows rapidly. However, the commonly used compression methods cannot acquire satisfying results. Recently, some researchers proposed human vision system (HVS) in the research of image compression. It can visually remove the information in most degree that human vision cannot preserve. In this paper, according to the existed and stated experiments and conclusions, the physical and anatomic structure of human vision is combined and the contrast sensitivity function (CSF) is introduced as the main research issue in HVS, and then the main designing points of HVS model are presented. On the basis of multi-resolution analyses of wavelet transform, the paper applies HVS including the CSF characteristics to the inner correlation-removed transform and quantization in image and proposes a new HVS-based medical image compression model. The experiments are done on the medical images including CT and MRI. The results show that under common objective conditions, the method used in the paper can achieve better subjective visual quality. The compression ratio can reach 16 : 1 if the visually lossless effect is required, i.e. , almost all relevant medical information is reserved. In some occasions, the ratio even reaches 80 : 1.2704|201|0更新时间：2024-05-07
- 摘要：In order to accurately identify and locate the functional active information, an objective and effective threshold must be found to threshold the statistical parametric map of fMRI. This paper reports a method to threshold the statistical parametric map of fMRI, which combines controlling the false discovery rate with analyzing the spatial correlation of the cortical hemodynamic responses. Firstly, for identifying the active voxels, the proposed method determines threshold based on controlling false discovery rate. Secondly, it analyzes and tests the correlation of hemodynamic responses between the identified active voxels and their 26-neighboring voxels. Not only can it select the threshold adaptively, but can identify the false active voxels caused by stochastic factors, with the strong capability of identifying and locating the functional active information. The proposed method is validated by using both simulation study and real fMRI data analysis.关键词：Medical image;Functional magnetic resonance imaging (fMRI);Statistical parametric map;False discovery rate;Correlation analysis;Multiple hypothesis testing2310|173|0更新时间：2024-05-07
- 摘要：The paper brings forward a new image recognition method for pathological changes with morphology theory and SVM algorithm. Image character described by structure theory and frame sector is used as training data of SVM. Meanwhile, rapid classification of image character can be realized by the aid of favorable classification performance of SVM, by which image structure can be classified, which can improve recognition efficiency. The paper also gives the infrared image sample of galactophore to test the algorithm. The result shows that the algorithm can improve recognition ratios effectively and the algorithm has a broad foreground. As well as this method can be applied other similar pattern recognition.2486|233|0更新时间：2024-05-07
- 摘要：In this paper, a algorithm of region-based 3D motion tracking on multi-linkage is proposed. Firstly based on the fact that the length of fingers is same in motion, and each of them lies in the same virtual plane because of only one degree of freedom at their joints, which is defined as the rigidity constraint and complanar constraints respectively, we obtain the initial 3D structure of finger from the image by exploiting these constraints and special properties on fingers. Then by the motion properties of rigidity multi-linkage and corresponding gesture constraints, we give the general descriptions of 3D region tracking of the linkage. Finally we apply the techniques to 3D motion tracking of hand. Here genetic algorithm is employed to our optimal iteration. In order to verify the validity of our approach, we use 3D motion of fingers, which is obtained from the real image sequence, to drive our virtual model of fingers. The results demonstrate the validity of the method.关键词：Pattern recognition;3D motion tracking;Constraint fusion;Rigidity multi-linkage;Image sequence2863|188|0更新时间：2024-05-07
- 摘要：In order to improve the aceuraey and efficieney to identify the aspha1t pavement surface distress by the image information a method of asphalt pavement surface image feature representation 15 put forward，whiceh may reduce calculation of pavement surface distress image classification. A pavement surface image 15 divided into 64X64 pixels sub-images and the intensity variances are used to represent the sub-images feature.Meanwhile，the sub-image pattern classifier 15 designed based on BP artificial neural network，all of the sub-images pattern classifying results are arra”damatrix and the pavement surface distress image segmentation 15 represented by this matrix By the experiment to identify the typical surface distress of asphalt pavement，the conelusions are as follows:(a)the average value and a minimum value of intensity variances of the image segmentation are input to BP artificial neural network，50 that the models of distress sub-images can be identified accurately.(b)the counts of horizontal and vertical projection of the distress sub-images models，sum of the distress sub-images，and convinced coefficient of the distress sub-images models may be used to deseript the main features of different the pavement surface distress，(e)the distributing features of the pavement surface distress image segmentation are used to represent a integration feature of the pavement distress image，50 that a good result for identifying Pavement surface distress may be obtained.2794|224|0更新时间：2024-05-07
- 摘要：The car plate recognition system is an indispensable part of intelligent traffic system. In car plate recognition system, one of the most difficult problems is the feature and recognition of car plate Chinese character. The multiresolution property of wavelet is applied to extract car plate Chinese character feature extraction, and a method that directly extracts feature from gray scale image is proposed. The method firstly extracts two kinds of features: one is wavelet moments in different multiresolution, and the other is the zoning density of wavelet-decomposed image. Then, two kinds of features are combined and a feature selection algorithm based recognition rate is presented to choose better features. Finally, the selected features are as the input of BP neural network, which is adopted to recognize car plate Chinese character. As a result, the proposed method avoids the binary operation used in some traditional Chinese character feature extractions that will seriously destroy the Chinese character structure. Furthermore, the extracted features can describe the local and global property of the character. Compared with some other feature extraction methods, the proposed method can achieve better recognition performance.3085|237|0更新时间：2024-05-07
- 摘要：Edge contour is an important feature of surface, which plays a special role in shape matching, surface conjunction. Due to the irregular distribution of space mesh data, the contour of surface cannot be gained from coordinate of points directly. So a method to detect edges of triangular mesh surface is brought out in this passage. If the adjacent points of the certain point can be formed a closed curve form ridges of triangular mesh, this point can be determined to a edge point. Scan the adjacent edge points of every edge point to get the sequence of edge points. This method has the advantage of simple and easy to implement. Which has been employed in the computer aided cultural relic restoration system, and has been proved to process tri-dimension data correctly and efficiently. This method is universal, which can also be applied to other mesh data.关键词：Computer graphic;Edge detection;Triangular mash surface;Edge contour;Adjacent point;Edge point3689|222|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