最新刊期
卷 14 ， 期 1 ， 2009
- 摘要：The bandwidth restricts airborne communication,so compression airborne SAR image with loss is a feasible way to enhance the real-time performance and image quantity. This paper proposed a SAR image compression strategy which extracts and protects target regions based on image decomposing technology. Firstly, the SAR image is decomposed into structure component and texture component. Then the target information is detected and classified from the texture component to construct ROI mask. Finally, the ROI mask is used to protect the important target information during compression by allocating more bits to it while reducing bit allocation to the residual.The experimental results indicate that the reconstructed image using the proposed approach has better visual effect than those processed by JPEG2000 arithmetic under the same bit rate.关键词：image decomposing;automatic target recognition;image compression;total-variation6685|515|0更新时间：2024-05-07
- 摘要：A new algorithm for SAR image denoising based on adaptive shrinkage in Nonsubsampled Contourlet Domain, is presented . The nonsubsampled contourlet coefficients of SAR images are reduced by the corresponding Pizurica adaptive shrinkage factor. The Pizurica shrinkage factor takes into account not only the local noise measure ,but also prior directional spatial consistency, and combines the shift-invariance and direction selectivity of the nonsubsampled contourlet transform. The shrinkage factor is assembled at each high frequency subband to modify the nonsubsampled contourlet coefficients. Inverse nonsubsampled contourlet transform is performed on the updated coefficients to get the denoised image. Compared with the denoising methods based on wavelet soft-threshold and contourlet soft-threshold ,the proposed algorithm can reduce speckle noise more effectively while preserving the edges of the SAR images.5515|473|0更新时间：2024-05-07
- 摘要：As one of the classic threshold methods for image segmentation, Otsu algorithm has been widely applied in image processing. Based on the analysis of the original 2D Otsu which has the disadvantages such as histogram area partition and time consuming, this paper introduces a new method of area partition in terms of speckle noise in SAR images, and brings forward a new rule of threshold selection which fits the image model better. The experiments show that the segmentation effect of the improved algorithm is better and its computational speed has been greatly improved.关键词：threshold segmentation;Otsu algorithm;2D histogram;maximum between-class variance;SAR5535|385|0更新时间：2024-05-07
- 摘要：Clutters in SAR (synthetic aperture radar) image is various and complex. So different kinds of distribution models should be used for CFAR (constant false alarm rate) based target detection of airborne SAR images, which increases the difficulties and complexities of automatic target detection. Gaussian mixture model (GMM) is an extension to Gaussian probability density function. Theoretically, it can approximate any distribution smoothly. In this paper, clutter distributions of SAR image are estimated using the GMM. And a novel CFAR threshold expression was proposed. Describing different clutter distributions with one uniform model, the proposed CFAR detector is more universal since it is less dependent on clutter distributions. The experimental results show that, even though the clutter distributions are unknown, the proposed technique still has a nice performance.关键词：GMM (Gaussian mixture model);CFAR (constant false alarm rate);target detection;SAR6345|407|0更新时间：2024-05-07
- 摘要：In SAR images, the goodness-of-fit of distribution model to SAR clutter data has great effect on the performance of statistical based CFAR (constant false alarm rate) target detection algorithms. In extreme nonhomogeneous region, the common distribution models can not describe the clutter data accurately due to strong impulsive interference. Whereas, α stable distribution which is based on Generalized Central Limit theorem can model the strong impulsive interference accurately and is fit for both homogeneous and nonhomogeneous clutter regions. So a CFAR detection algorithm using α stable distribution is studied in this paper and some key steps, such as parameter estimating, model standardization and threshold calculating is also included. The results on real data show that the SαS CFAR has good performance and can achieve high detection rate and low false alarm rate.关键词：SAR images;α stable distribution;CFAR(constant false alarm rate) target detection5526|355|0更新时间：2024-05-07
- 摘要：In this paper, a robust slowly-moving target detection and location algorithm is investigated. In practice, due to channel imbalance such as gain and phase error, the clutter suppression performance of the interferometric SAR/ GMTI is significantly degraded, and makes it difficult to detect and locate the moving target. A channel error calibration algorithm based on sub-patch SAR image is proposed, and the spatial adaptive processing is used for the moving target detection and location.The simulation results are presented to demonstrate that, the proposed algorithm is robust to system errors, and is suitable for engineer implementation.关键词：synthetic aperture radar (SAR);ground slowly moving target detection;channel error5801|384|0更新时间：2024-05-07
- 摘要：The attributed scattering center model is based on the solutions to both physical optics and the geometric theory of infraction. The model provides the characteristic feature of attributed scattering centers which can be well used in the target recognition. In order to get the feature on high resolution SAR Imagery, the attributed scattering center model is discussed and the method of extracting the feature of attributed scattering centers from a high resolution SAR image is researched in this paper. It is a parameter estimation method based on a SAR imagery. The method includes four steps: image segmentation, the selection of the model type, the estimate of initial value and the parameter optimization. The seven parameters of each scattering center can be obtained through above four steps. After the recurrence procession, the parameters of all the scattering centers in the imagery can be attained. At last, the validity and accuracy of this method can be demonstrated by the results of simulation.关键词：SAR imagery;attributed scattering centers;feature extraction7074|456|0更新时间：2024-05-07
- 摘要：A new solution to internal wave detection and parameter estimation in SAR ocean image is presented in this paper. It improves the traditional Radon transform through normalizing the Radon coefficients in an ellipse region and thus can detect the inconspicuous lines with ocean background. Making use of the idea of glide windows we use the proposed transform to detect the internal wave region, and then we can estimate the wave direction and the wavelength simply through dealing with the Radon coefficients. The experiment on real SAR images show that the proposed method can process the SAR images automatically, it can detect the internal wave region correctly and the estimated parameter is consistent to the real observation.关键词：SAR;RADON transform;internal wave detection;internal wave character estimation5918|354|0更新时间：2024-05-07
- 摘要：As for hyperspectral image compression, there is no standard or mature paradigm and distributed source coding (DSC) is a new scheme. In this paper, first,the concept and significance of distributed source coding in hyperspectral image compression is introduced. Second,the theoretic foundation and implementation of DSC is reviewed. Third the predominant methods of applying DSC to hyperspectral image compression are systematically discussed, and the application outlook, the current development and the technical difficulties for this technique are pointed out.关键词：distributed source coding;hyperspectral image;lossless and lossy compression4454|337|0更新时间：2024-05-07
- 摘要：To solve the limitation of the traditional evaluating models of video quality in accessing the quality of video over a packet network, such as unsuitable for on-line usage, not considering human visual perception, this paper introduces a video quality evaluating model based on packet loss. This model, which is inferred based on the statistics theory, can access the video quality over a packet network with a quantitative index according to the characteristics of video coding, packetiing and translation. It consumes little computing resources in quality evaluation and is suitable for real-time applications. In order to demonstrate the feasibility and accuracy of this model the simulations and experiments utilize this model and MPQM model, which objectively assesses video quality based on human visual properties, to evaluate a series of video sequences distorted with different packet loss rates and contrast their evaluation values. The experimental results have shown that the evaluation values of this model are approximately equal to MPQM values and reflect human visual perception.关键词：packet loss;IPTV;video quality4605|412|0更新时间：2024-05-07
- 摘要：According to the distribution of the best inter coding modes, we give three early-terminated strategies for deciding whether one of the SKIP, P16×16, P16×8 or P8×16 can be selected as the best inter coding mode. If the conditions can be satisfied, the computation of RD cost of the rest mode can be saved. Experimental results show that the proposed algorithm can save about 50% to 70% time compared with a full-searching algorithm in H.264/AVC, meanwhile the decreasing of PSNR is about 0.17dB and the increas of bit-rate is about 2.42%.关键词：video coding;H.264/AVC;inter prediction;fast mode decision;encoder optimization3616|319|0更新时间：2024-05-07
- 摘要：SAR speckle filtering is an important research subject. An ideal algorithm should smooth the Speckle without blurring texture and fine detail. A new speckle filtering method, using selective threshold in directionalet domain, is proposed. Image was decomposed and transformed in 12 directions by using multi-directional frames of directionalet. According to the characteristics of different length pixel sequences, the pixel sequences are divided into three groups and dealt with different threshold method respectively by Gaussian noise statistical distribution interval. To retain many fine details in the image, the median of high frequency coefficients variance is used as noise variance estimate value in every scale. A real SAR image is used in the experimentation. The result shows that the proposed method outperforms several classical speckle reduction algorithms in visual quality. ENL and ESI are enhanced by 97 and 0.15.关键词：SAR image;speckle filtering;directionalet;multi-directional frames;selective threshold3916|382|0更新时间：2024-05-07
- 摘要：Image inpainting is an important research field in the area of image restoration, which can be used to restore damaged images and videos, remove text and conceal errors in videos. Now many inpainting algorithms perform well in the gray image inpainting, but they are time consuming and not quite applicable to color images. This paper introduces coherence-enhancing diffusion into image inpainting, and adapts it to improve the efficiency and performance of inpainting. Moreover our method can extend for color image inpainting naturally. Many experimental results prove our method is effective.关键词：coherence-enhancing diffusion;image inpainting;structure tensor;isotropic diffusion;anisotropic diffusion4259|310|0更新时间：2024-05-07
- 摘要：As to typical noisy image restoration methods, noise is regarded as a disturbed source.As long as noise becomes strong and image signals become weak, it is difficulty to restore noisy image because of the limitation of these methods.Based on theory of suprathreshold aperiodic stochastic resonance in Hodgkin-Huxley model, a new self-adaptive adjusting adding optimum noise image restoration method is proposed.Results indicate that, as to the restoration effect of gray-scale noisy image, especially strong noisy image, this new method is better than other typical methods used.Thus, this study suggests a new approach in engineering application where under strong noise background, such as image restoration and target identification, given that this new method is of good robustness.关键词：image restoration;Hodgkin-Huxley model;suprathreshold aperiodic stochastic resonance;peak signal-to-noise ratio;self-adaptive adjusting4103|314|0更新时间：2024-05-07
- 摘要：Partial differential equation(PDE) has become an important tool in image processing and analysis. A basic content of digital image processing is image zooming, namely an image processing technology obtaining a high resolution image from a low resolution image. This article proposes an image zooming algorithm based on the TV-norm (Total Variation norm) interpolation model and fourth order partial differential equation interpolation model, which adopts the idea of total variation model and two kinds of functional models. The validity and feasibility of the algorithm is tested through a simulation experiment and two appraisal criterions—mean square error(MSE) and peak value signal to noise ratio(PSNR).关键词：partial differential equation(PDE);image zooming;variation;interpolation4979|406|0更新时间：2024-05-07
- 摘要：A new algorithm called fuzzy inverse fisher disciminant analysis is proposed. The inverse fisher discriminant analysis (IFDA) is effective in extracting discriminant features, but is assumed the same level of relevance of each sample to the corresponding class. Distribution information of samples is represented with fuzzy membership degree in the fuzzy inverse fisher discriminant analysis (FIFDA). Furthermore the information is utilized to redefine the corresponding scatter matrices, which are different from the IFDA. The experimental results on ORL and FERET face database indicate that the performance of FIFDA is superior to that of Fisherfaces and IFDA.关键词：Fisher discriminant analysis;inverse Fisher disciminant analysis;fuzzy inverse Fisher discriminant analysis;feature extraction;face recognition4031|375|0更新时间：2024-05-07
- 摘要：Modular 2DPCA is an extension of 2DPCA algorithm. The recognition performance of modular 2DPCA is more robust than that of 2DPCA.In this paper, the defects of modular 2DPCA about computing the total scatter matrix of training samples and selecting eigenvectors are analyzed. An improved modular 2DPCA algorithm is presented. Experiments show that the improved modular 2DPCA algorithm can select better eigenvectors and extract facial features more effectively.关键词：modular two-dimensional principal component analysis;eigenvector;feature extraction;face recognition4034|307|0更新时间：2024-05-07
- 摘要：This paper presents a real-time color image tracking method. In order to implement an effective and robust tracking task, a novel approach of weighted color target model based on color histogram is discussed, It takes into account the targets shape and position of pixels as necessary factors in target model. This makes the description of targets feature more reasonable. Furthermore, the starting position of target is found with Kalman filter in every frame, and then mean shift is utilized to track the target position.Experiment result shows, in Contrast with mean shift, the proposed method can locate object more accurately and have lower complication of calculation.The proposed method is more robust，especially for some complex tracking conditions such as occultation and object which is moving fast.关键词：colored object real-time tracking;weighted color probability contribution;mean shift;Kalman filter3631|343|0更新时间：2024-05-07
- 摘要：Particle filtering has drawn more attention recently due to its superior performance in nonlinear and non-Gaussian problems, which uses an observation model to describe the interested target and its performance depends strongly on the observation model. A novel particle filter algorithm has been proposed for head tracking, which focuses on finding a solution to the problem that the feature of the tracked object does not always match the particle observation model with changing time. In brief, it unifies the difference of the tracked object features with the particle observation models, and thus it forms a group of particles with different models to track the target. Based on the changing displays in the feature cues of the head with the rotation, particles with different observation models convert alternately in the tracking process. To evaluate the novel head tracker performance, some real sequences is tested and some results are shown that the new tracker is robust to the rotation of head in a cluttered background and has better tracking precision than the standard particle filter.关键词：particle filter;head tracking;observation model4285|363|0更新时间：2024-05-07
- 摘要：An object tracking algorithm based on adaptive particle filter is proposed in this paper. Boosting algorithm is introduced into particle filter algorithm, and adaptive particle filter is constructed. Features classifiers are constructed utilizing object information and background information, and the outputs of these classifiers taken as important information of observations of particle filter are used to calculate particles’ coefficient. Also, these classifiers are updated during tracking in order to update particles’ coefficient adaptively. The experiment result shows that the tracking algorithm we proposed can adaptively select features for tracking utilizing different background information, in applications such as existence of covering, appearance changed, clutter in the background and illumination changing. The objects can be tracked stably.关键词：particle filter;adaptive features selecting;tracking;boosting algorithm16511|496|0更新时间：2024-05-07
- 摘要：A vision-based posture recognition system is proposed utilizing Optimal DAGSVM (Directed Acyclic Graph Support Vector Machine) classifier to achieve natural and reliable human-robot interactions. Coarse-to-fine feature detection scheme extracts skin-colored candidate regions, followed by face and hand verifications with Gabor filtered eye features and wavelet-moments of hand edge respectively. Statistical invariant moments and relative coordinates of face and hand regions are calculated as pattern feature vectors.A set of binary SVM classifiers are combined using Decision Directed Acyclic Graph with optimal structure to construct a more accurate multi-class DAGSVM classifier. Experimental result validates the reliable performance of the approach, where a natural and friendly interaction is achieved with a service robot.关键词：posture recognition;wavelet-moments;directed acyclic graph support vector machine(DAGSVM);human-robot interaction4211|338|0更新时间：2024-05-07
- 摘要：A video object segmentation method based on MAP (maximum a posteriori) of pixel blocks is proposed. Firstly, a reference frame is built from three continuous frames and a pixel block-based difference image is calculated. Secondly, a pixel block-based difference image is described by Gibbs random field model, the moving object blocks are detected by MAP. Last, the contour of the moving object is extracted by the snake method. The experiments results show that the algorithm can avoid the losing and mistakes with a good speed.关键词：pixel block;video object segmentation;MAP;Gibbs random field;Snake model4551|352|0更新时间：2024-05-07
- 摘要：A defect detection approach based on Gabor wavelets is proposed. Acquired image is multi-channel filtered with a bank of 2-D real Gabor filters. Next, feature images are obtained by subjecting each filtered image to a nonlinear transformation and computing a measure of energy through smoothing filter. In the learning phase, statistical parameters are computed using learning sample which is free of defect and are used to supervise the thresholding of inspected image. Finally, threshold-fietered feature images are fused inter-scale and inter-orientation and binarized in order to decrease false alarms. Experiments demonstrated that this approach has good detection ability performance and needs less learning samples, which makes it suitable for many types of defect and textured material.关键词：machine vision;defect detection;Gabor filtering;multichannel5272|396|0更新时间：2024-05-07
-
The Multi-scale Lacunarity Feature of Mobile Targets in SAR Imagery and the Discriminating Algorithm
摘要：A new algorithm is developed to discriminate mobile target from natural clutter in high-resolution SAR imagery with multi-scale lacunarity feature in this paper. The concept of lacunarity is introduced, and the different methods are proposed to calculate the lacunarity in the binary image and gray image. And then, the multi-scale lacunarity are used discriminate the vehicle target. Finally, the MSTAR data with vehicle targets and natural terrains are used to validate the above algorithm, and the performance of this algorithm is good.关键词：SAR imagery;target discrimination;mobile target;lacunarity4102|274|0更新时间：2024-05-07 - 摘要：Nowadays large volumes of data with high dimensionality are being generated in many fields. Many approaches have been proposed to index high-dimensional datasets for efficient querying. ClusterTree is a new indexing approach representing clusters generated by any existing clustering approach. Lots of clustering algorithms have been developed, and in most of them some parameters should be determined manually. The authors propose a new subspace-cluster indexing algorithm, which based on the improved CLIQUE and avoids bias on any parameters caused by user. Using multi-resolution property of wavelet transforms to reprocess the distribution curve of samples, the proposed approach can cluster at different resolution and remain the relation between these clusters to construct hierarchical index. The results of the experiment confirm that the subspace-cluster algorithm is very applicable and efficient, and show that this hierarchical indexing structure does well in the content-based image retrieval.关键词：content-based image retrieval;high-dimensional data index;subspace cluster;cluster tree3994|296|0更新时间：2024-05-07
- 摘要：There is noise and defective data on the 3D scanning point cloud. A robust mesh reconstruction algorithm is proposed. Surface normals are estimated by tensor matrix with enhanced features. By computing 3D fast Fourier transform (FFT), discrete iso-surface is extracted. Points are moved onto the iso-surface by an iterative clustering along gradient field, where the noise and outliers are removed and defective data are repaired. Point cloud is decimated adaptively, and then a new triangle is generated using sphere-intersected method. The experimental results have shown that the algorithm is fast, robust and use low memory.关键词：reverse engineering;Fourier transform;meshing;denoising4079|309|0更新时间：2024-05-07
- 摘要：In this paper we bring forward an image edge similitude extent algorithm based on the steadiness character theorem of plane similitude graphics—similitude included angle theorem, and the algorithm is realized by constructing an AFMSDF (mean-square deviation function of angle fluctuation). The AFMSDF is used as a method of scaling the similitude extent of two different edge contours. This algorithm provides a reasonable method for graphic match, graphic search, efficient graphic similitude transform,and clustering, etc. Theoretical analysis and experiments indicate that the method is feasible.关键词：graphic edge similitude;similitude extent;graphic match;graphic search4439|291|0更新时间：2024-05-07
- 摘要：With the rapid development of recent 3D scanning and graphic modeling technique, the size of 3D model data increases drastically, which brings many challenges in its storage, display and transfers. Therefore a simplistic representation of 3D model data is in great demand. First current algorithms of dynamic mesh simplification were analyzed, and then an algorithm for mesh simplification was designed to construct topology alterable progressive mesh representation of 3D model. Based on this progressive mesh, a view dependent, compact and dynamic multiresolution structure was constructed by reorganizing the simplified model data. Finally a view dependent dynamic simplifying algorithm was presented. Theory analysis and experiment result showed that this novelty method can generate proper detailed simplified model based on view parameters, which achieved good simplification effect preserving both its geometry shape, and its attribute characters such as color etc. The presented algorithm can be widely used in many applications because of its small storage requirement, low cost and high adaptability.关键词：view dependent;topological simplification;multiresolution;mesh simplification3831|317|0更新时间：2024-05-07
- 摘要：Most of spatial data sharing platforms at present are still based on traditional geographic metadata mode，mostly in the format of simple text or brief figure; On the other hand, with the popularization of internet, integrating and visualization sharing of multi-source spatial data unified on internet are urgently required. On the basis of problems fore-mentioned, firstly, in this paper we analyzed the requirement of integrating and visualization sharing of multi-source spatial data. Then，aiming at the characteristics of coastal zone multi-source spatial data,we discussed the methods of integrating coastal zone multi-source spatial data unified on internet, the interactive querying of vector and raster graphics, and downloading multi-source spatial data visually. Finally,we took “Taiwan Strait area multi-source spatial data integration sharing system” as a case study, introduced the detailed realization process of integration and information services of multi-source spatial data in coastal zone based on ArcIMS. It shows that the technologies for the integrating and visualization sharing of multi-source spatial data in coastal zone provided in this paper, can satisfy users sharing requirement on internet in a way of “What you see, what you get”.关键词：visualization;multi-source spatial data;interactive querying of vector and raster;ArcIMS4101|267|0更新时间：2024-05-07
- 摘要：In this paper, a hardware algorithm for graphics rasterizer on embedded platform is presented. By dividing the pixels in the triangles bounding box into a number of regular tiles, the algorithm do the scan conversion, pixel interpolation and projective correction on the basis of tile architecture. After a lot of optimization, the algorithm is implemented and tested on FPGA. As compared with the traditional algorithm, the algorithm presented in this paper has increased the pixel hit rate, and reduced the computational complexity, as well as effectively reduced the hardware cost. Testing results show that, the quality of the algorithms rendered images has reached the level of OpenGLES 1.1.In general scene, the rendering speed reached 30 fps, meeting the requirements of real-time rendering. In terms of the synthesized hardware resources, it is small within 5 545 slices on Xilinx FPGA Vertex2P xc2vp30-7ff89.关键词：3D graphics;graphics rasterizer;hardware implementation;tile3827|285|0更新时间：2024-05-07
- 摘要：In recently years, with the increment of urbanization, the computer aided urban planning has become more and more important. Generally, the computed aided process is accomplished through the help of multiple software systems. There are several issues with the above solution such as weak applicability, lack of efficient auto-evaluation method, etc. The working principle and main modules of CAUPS are briefly introduced. The key technique for auto-evaluation, i.e., the design and implementation of evaluation models definition system in CAUPS, is emphasized. A series of simple operations and resolution for the entity relations are firstly provided to describe all the possible evaluation models, and then users can customize their evaluation models by combining the proper entity relations and entities based on the practical planning demand.关键词：CAUPS;evaluation models definition system;keywords match;syntax resolution3689|267|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