最新刊期
卷 18 ， 期 9 ， 2013
- 摘要：The high resolution remote sensing images can provide detailed ground information, and they have complex spatial structural features. Effectively describing and modeling them will play an important role in better realizing the tasks such as image interpretation, object recognition and extraction, scene understanding, and so on. This paper firstly introduces the concept and properties of spatial structural features, and divides the spatial structural feature into three levels: pixel structure, object structure and scene structure. Then, a systematic review on recent development of spatial structural feature description and modeling methods is presented. The review focuses on the basic principles and research status of these methods, and some advantages and drawbacks are pointed out. Finally, conclusions are drawn and future research directions are discussed.关键词：high resolution remote sensing image;spatial structural feature;feature description;feature modeling4059|642|0更新时间：2024-05-07
- 摘要：The full-polarimetric (FP) imaging mode has enriched the information of Synthetic Aperture Radar (SAR) data, and has been widely used in remote sensing applications, e.g. for land cover classification, environmental monitoring,or target detection. However, the application of FP SAR has been constrained by the design and maintenance complexity, power consumption, image swath, and data-downloading rate. Compact Polarimetric (CP) SAR systems not only overcome the drawbacks of FP SAR system, but also keep the FP information to a certain extent. Preliminary results are available in several fields, such as forest parameters estimation, feature classification and target detection. This paper provides an overview of compact polarimetric SAR system fundamentals, describes the methods of FP information reconstruction and decomposition, and summarizes its main research achievements of the last decade. At the end, we give its development trends.关键词：fully polarimetry;synthetic aperture radar(SAR);compact polarimetry;full polarimetry reconstruction;decomposition4224|322|0更新时间：2024-05-07
- 摘要：It is difficult to recognize objects when they are distorted and partially occluded or broken. In order to solve this problem, a new registration algorithm is proposed in this paper. First, we define the CCCTI(cyclic code of corner and tangent and inflexion points) of a planar curve. This code can easily determine the corresponding relation of the key feature points between the model and target contour. Second, contours are segmented by using the key feature points, and the transformation matrix is estimated based on two endpoints of the sub-curve and its centroid. Then, all estimate matrices are clustered using hierarchical clustering method. This process can reduce the final error of the estimation matrix and improve the robustness of the algorithm, making the algorithm suitable for partial occlusions or broken objects. Finally, we calculate the total centroid of corresponding sub-curve segments, and estimate the transformation matrix again using the total centroid and two endpoints of the corresponding sub-curve segments, to improve the accuracy of the registration. The theoretical analysis and experimental results show that our algorithm can be effectively used for affine registration, and that it can deal with partial occlusions and broken objects.关键词：affine registration;hierarchical clustering;cyclic code of corner and tangent and inflexion points (CCCTI);partial occlusion2702|219|0更新时间：2024-05-07
- 摘要：In order to deal with the problem of lower render quality and unsatisfactory subjective effect of virtual view in the virtual views synthesis system, and to improve the compression performance of the depth map,the filtering for decoded low resolution depth maps and a depth upsampling method are investigated in the depth map encoding method based on the down/up sampling in this paper. First, H.264/MVC is used to compress the low-resolution depth map according to the character of the depth map. Then, the À-Trous wavelet filter is used to filter the decoded low-resolution depth map to decrease the artificial effect, and a depth upsampling method based on multi-similarity adaptive support weight is used to reconstruct the original resolution depth map. Finally, up-sampling depth map and an uncompressed color sequence is used to render the virtual view and to evaluate the render quality by PSNR of the virtual view. The experimental result shows that the PSNR of two proposed methods are 0.04 dB and 0.22 dB higher than ASW(adaptive support-weight) respectively, and subjective quality of the virtual view is improved a lot, especially the later one.关键词：depth map compression;upsampling;À-Trous wavelet filter;virtual view rendering;rendering quality3113|323|0更新时间：2024-05-07
- 摘要：In order to effectively solve the multiple-instance image classification problem, we put forward a new classification method, which transforms the multiple-instance image into a single instance image in the new space-bag space. First, the whole image is regarded as a bag and each region as an instance of that bag. According to the same visual regions of image samples are put into one cluster and k-means clustering algorithm is used to determine the visual words for each class of images. At this step, we use the information that labels of negative samples are all known has been used to select the typical visual words. Then, we construct a new bag space with these visual words and use a nonlinear function based on these visual words to transform each multiple-instance image into a point in the bag space. Finally, standard SVMs are trained in the bag feature space to classify the images. Experimental results and comparisons on the Corel image set are given to illustrate the performance of the new method.关键词：bag space;multiple-instance learning;image classification;visual words3075|439|0更新时间：2024-05-07
- 摘要：Multimodal emotion recognition has been a very important research topic in affect computing. This paper mainly focuses on the methods of bimodal emotion recognition based on body gesture and facial expression and presents a new bimodal emotion recognition method based on bilateral sparse partial least squares(BSPLS). First, the spatio-temporal feature is extracted as the emotion feature vector for video-based body gesture and facial expression respectively. Then we propose a new bilateral sparse partial least squares(BSPLS) method to extract emotion feature and fuse facial expression and body gestures as new emotion feature. Finally, we use two classifiers in emotional classification. We compared the BSPLS method with some subspace methods including PCA, CCA and PLSR based on the data from the FABO database. The experimental results show that the fusion feature methods are all better than the monomodal emotion recognition and our BSPLS feature fusion provides the best recognition performance.关键词：body gesture;facial expression;bimodal emotion recognition;spatio-temporal feature;bilateral sparse partial least squares(BSPLS)4362|320|0更新时间：2024-05-07
- 摘要：To cope with the drift problem of tracking caused by environmental changes (such as illumination variations and occlusions), we propose a multi-target tracking algorithm with a hierarchical associative structure, which first coarsely matches the targets and then accurately locates them using Particle Swarm Optimization (PSO). Compared with the state-of-the-art tracking algorithms, context information is integrated into the generation of the particles during the coarse matching stage in this paper, thus enhancing the accuracy of target matching as well as reducing the number of false-tracked targets. To ensure the tracking accuracy, the targets' locations with prominent deviations in the phase of accurate tracking are rectified via Metropolis-Hastings algorithm; meanwhile, the targets' templates are updated. Experimental results show that the proposed algorithm can track the occluded targets more accurately under the occlusions.关键词：visual tracking;template matching;object detection;multi-target tracking;particle swarm optimization2906|254|0更新时间：2024-05-07
- 摘要：As an effective visual target tracking algorithm, the tracking-learning-detection(TLD) has drawn wide attention around the world. In this paper, we propose an improved TLD visual target tracking algorithm, which is obtained by making several improvements based on the original TLD algorithm. The improvements include modifying local tracker placement as well as local tracker failure predicting method for the tracker of TLD to improve the precision and robustness of the tracker; employing Kalman filter in the detector of TLD for estimating the location of the target to reduce the scanning region of the detector and improve the speed of the detector; adding Markov model based target moving direction predictor in the detector of TLD to increase the discretion for targets with similar appearance. Experiments have been conducted to compare the performance of the original TLD and improved TLD. The experimental results show that, compared with the original TLD, the improved TLD has more accurate tracking precision, faster tracking speed and stronger distinguishing ability for targets with similar appearance in the scene.关键词：target tracking;tracking-learning-detection(TLD);tracking precision;processing speed;target moving prediction3113|334|0更新时间：2024-05-07
- 摘要：In this paper, we propose an algorithm named Multi-feature Robust Principal Component Analysis(MFRPCA), to integrate multiple visual features for Video Moving Object Segmentation. The aim of Video Moving Object Segmentation is to separate the moving objects from the static information. Its main process is to decompose the multiple feature video matrices into low rank and sparse matrices. The decomposition is to solve a minimization problem formulated as a constrained combination of nuclear norm and Lnorm, which can be solved efficiently by Augmented Lagrange Multiplier(ALM) method. Compared to other methods developed recently, the proposed method integrates color, edge and texture features. The quantitative results for Recall and F-measure obtained from experiments on the change detection benchmark dataset are 0.486 0 and 0.559 7 respectively. The results which outperform other methods well show that the proposed MFRPCA can achieve more robust and reliable performance.关键词：video segmentation;sparse and low rank;multi-feature robust principle component analysis;augmented Largrange multipliers;nuclear norm minimization3390|303|0更新时间：2024-05-07
- 摘要：To overcome the shortcomings of the traditional FCM algorithm, such as the difficulty to determine cluster numbers slow iteration, and the tendency to plunge into local optimization,as well as sensitivity to, the initial values,an improved fuzzy c-means clustering combining with the global best harmony search algorithm(GBHS-FCM) is proposed. First, the initial cluster numbers and cluster centers of the FCM algorithm are obtained by the GBHS algorithm, while taking the advantages of global superiority and robustness of the GBHS algorithm. Then a new fuzzy clustering function is presented by combining the pixel intensity information and the spatial dependence to the neighboring pixels together, which enhances the spatial continuity of the segmentation results. Finally, a new distance formula is proposed to replace the traditional Euclidean distance formula, which enhances the robustness of the new algorithm to noise. The simulation results show that the GBHS-FCM algorithm performs better than FCM algorithm in accuracy, speed and robustness.关键词：image segmentation;fuzzy C-means(FCM) clustering;global best harmony search algorithm (GBHS);clustering accuracy2951|269|0更新时间：2024-05-07
- 摘要：The factorization method is an important method for recovering the geometric structure of a rigid object from image sequences. First, the conventional factorization method is introduced, as well as the analysis of its shortcomings. In order to avoid the invalidation, an improved factorization method is then proposed. Meanwhile, the complex process of solving the corrective matrix in a conventional way is avoided. The rotation matrix of each frame is directly corrected according to the property of a rotation matrix, which has been decomposed by the conventional method using singular value decomposition(SVD). Then, we calculate the structure matrix using linear least squares method, which directly combines the watching matrix with the solved rotation matrix. The experiments using synthetic and real images illustrate that the proposed method can recover the geometric structure from image streams very efficiently, and it also improves the stability, compared with the conventional method.关键词：factorization methods;structure from motion;singular value decomposition;correction of rotation matrices2703|651|0更新时间：2024-05-07
- 摘要：Scan-convension is the essential part of the graphic pipeline. It takes full advantage of data parallel processing. High performance computing mainly relies on large throughput, but the conventional state management method in the graphics pipeline reduces the throughput because of processing unit stalling. An efficient state management method is proposed in this paper. In order to maintain command ordering, the proposed method divides the commands into two classes: compute command and control command. The method reduces the complexity of maintaining ordering and state management compared with conventional method, and the state-mask generated by state processing unit decreases the count of the broadcast communication and reduces the bandwidth needed for state management. Experimental results show that for applications such as computer games, in which the backend states change frequently, the rate of raster engine usage is improved by 5%~9%.关键词：state management and ordering;scan-convention;GPU(graphic processing unit);parallel processing2145|212|0更新时间：2024-05-07
- 摘要：For holes in 3D human skull are various, which can not be repaired with specific algorithm of mesh editing in computer graphics, a novel hole-repairing algorithm for big and specific area holes in mesh surface models of the 3D Human skull is proposed. In order to fill the big and complicated holes of skull, a new inward recursive solution (IRS) method based on the standard BHRA method is proposed to improve the limitations of the traditional methods which make the patch mesh too flat, and splices the patch mesh and source mesh of model smoothly; in addition, a template matching algorithm (TMA) method is introduced for repairing the particular holes, which uses a template model to restrict the deformation. The experimental results show that our methods can produce desirable results for repairing different kinds of holes in mesh surfaces of human skull, especially for the big and particular holes.关键词：hole repairing;hole types;implicit surface;inward recursive solution;template matching algorithm2350|250|0更新时间：2024-05-07
- 摘要：Ultrasound images have been widely adopted to investigate the human muscle fascicles as a non-invasive method that imparts no radiation. However, manual measurement of fascicle orientation from ultrasound images can be subjective and time-consuming in clinical enviroments. In this paper, a new approach is proposed to detect and track the fascicle orientation automatically by using the texture feature of the fascicle region. The ultrasound images are projected using a normalized Radon transform. For each Radon orientation,the variability of the results of the normalized Radon transform is statistically measured by Segmented-Variance. Compared to traditional methods that directly detect the line feature of the fascicles can be directly detected in this way and the proposed method makes full use of the texture feature of the parallel distribution of the fascicles. The result shows that the robustness of the orientation detection has been enhanced. Four ultrasound sequences are tracked consecutively to evaluate this method. The correlation coefficient of the detection results between the proposed method and manual method is 0.90±0.01, and the average of the absolute difference between these two methods is 0.41°±0.31°. We believed that the proposed method has a great potential in ultrasonic image analyzing and the biological mechanics field.关键词：fascicle orientation;ultrasound image;normalized radon transform;segmented-variance method2736|258|0更新时间：2024-05-07
- 摘要：In low light or night conditions, the signal-to-noise ratio is low and the details are not clear in color images. Although near infrared cameras can get image with rich texture, the edge details lacks color under this condidtion. In order to get a clear color image, we propose an improved bilateral-filter image-fusion algorithm in low light conditions in this paper. In the algorithm, the kernel function is re-designed using the power function to replace the exponential function, and canceling the design parameter. In order to avoid image texture and edge smoothing, the large difference of pixels is used as a pixel similarity in the color image and near IR image for image fusion. By using of a color image and an infrared image, this method and several other typical fusion methods are tested under low light condition. The experimental results show that the color images have higher definition compared with other fusion algorithm. The color is more close to the source image. The computation speed is more than 6 times compared faster to the Eric.P algorithm.关键词：bilateral filter;video fusion;Kernel function;near IR2566|399|0更新时间：2024-05-07
- 摘要：This paper presents a new ship-wake detection algorithm in synthetic aperture radar (SAR) images based on the length normalized scan method. The range component of a ship's movement leads to azimuth shift in SAR images. Therefore, the start point of wake must be located in the azimuth direction. According to this physical facts, length normalized scan is performed on-line on image parts which may be wakes. By means of linear integral and length normalization technology, linear feature detection is transformed to point detection. Then the probability model in the scan domain is constructed to realize constant false alarm rate (CFAR) detection. COSMO-SkyMed data is used to validate this method. The experimental results show that this algorithm is effective for ship wake detection and can extract ship's velocity with high accuracy.关键词：synthetic aperture radar;ship wake detection;length normalized scan;constant false alarm rate2769|289|0更新时间：2024-05-07
- 摘要：Using three-dimensional(3D) visualization technology to realize the management of transmission lines is a current developing direction on electric power information. However, the Earth is an irregular ellipsoid, which makes it hard to work out the coordinates in world space on the tower hanging point. Furthermore, it cannot truly describe the connecting condition between lines and tower hanging point. For these reasons, on the basis of analysis and research, we provide a hanging point calculating method that use the tower space of Google Earth as a starting point. The method is applied to the developed transmission line management system. With the characteristic of simple calculation, high precision, as well as having a real connection line and hanging point, etc. This method can meet the requirements of existing applications.关键词：three-dimensional(3D) visualization;electric power informatization;transmission line management;hanging point calculation2693|211|0更新时间：2024-05-07
- 摘要：We want to apply map symbols in the web map service(WMS), and allow a flexible customization of the WMS visualization. To achieve this, we studied the Styled Layer Descriptor implementation specification(SLD), which support WMS, work on the publishing method for map symbols for the web, and worked on the analysis of WMS and SLD. We designed the structure of the web map symbol service, and put forward the coupling method between web map symbol service and WMS based on SLD, in order to solve the problem of not being able to change map style by WMS on WebGIS. We validated our metkodby a prototype system. The experiments show that this method can effectively associated map symbol service with WMS, allowing a flexible and customizable WMS visualization.关键词：web map symbol service;styled layer descriptor;web map service;services coupling;visualization2436|234|0更新时间：2024-05-07
- 摘要：Robustness analysis of city road networks requires the evaluation of structural change in networks under attacks. The difficulty of robustness analysis is that when modeled at different granularities city road network structures have different characteristics and show different sensitivities to various attacks. In this article, a city road network is first modeled as dual graph at three different granularities. Then, a series of successive simulated attacks are designed based on network structure characterization methods from complex network theory. The robustness analysis is carried out by characterizing the structural change of a city road network under these successive simulated attacks. Six world cities with different urban morphology forms are tested. The results show that at the same granularity, the performances of different city road network under attacks are similar. However, at different granularities, a city road network has quite different performance. Thus, it is important to choose appropriate representation granularity when conducting robustness analysis of city road network for a specific application. We argue that road network represented by segment is suitable for the evaluation on how traffic interruption at certain point affects the network and can provide suggestion on related emergency response. Road network represented by stroke is a good option when we want to evaluate how filiform traffic interruption affects the network and can help to decide traffic control strategies. Road network represented by community is more suitable for robustness analysis under the condition of big events that affect the traffic in a larger area. Moreover, community is an appropriate representation for robustness analysis in dynamic traffic environment since the zonal road traffic interactions could be taken into account.关键词：city road network;robustness analysis;representation granularity;complex network;community detection2591|244|0更新时间：2024-05-07
- 摘要：Reconstructing contours through a DEM is one of the methods of the DEM accuracy assessment. It can be used to detect terrain levels of fidelity. However, these methods show all a relative fidelity accuracy.Therefore, we present a membership function of the matching between drainage and terrain in this paper, we use the matching difference between drainage and terrain to quantitatively assess the DEM precision and to detect terrain levels of fidelity. First, we establish the matching offset membership function in accordance with the linear element displacement criterion, the scale criterion, the average slope criterion, the maximum matching offset criterion, and the visual effect criterion. Second, we use a 1:50 000 topographic map data of the four regions as the source data. Then the experiment of quantitative descriptive model of the matching precision explained how to use it in scaling transform of DEM. Finally, the experimental results show that the quantitative descriptive model of the matching precision not only can be used for the quantitative assessment of the extent of DEM terrain fidelity, but also can be used to determine the conversion of DEM scale appropriate range.关键词：digital elevation model;reconstructing contours;fidelity accuracy;membership function;scaling transform2630|282|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