最新刊期
卷 15 ， 期 11 ， 2010
- 摘要：The research of Freehand 3D ultrasound has been playing an increasingly important role in the 3D reconstruction techniques of ultrasound images in recent decades. 3D ultrasound is classified firstly according to different taking manners. Then the characteristic of Freehand 3D ultrasound system and the principle of 3D reconstruction for Freehand ultrasound images are illuminated. Finally, the overview and comparison analysis of Freehand 3D ultrasound reconstruction algorithms are provided. Synchronously some challenges and research orientations are also indicated.4048|258|0更新时间：2024-05-07
- 摘要：Discrete cosine transform (DCT) has been applied extensively to the area of image compressing; in order to improve image encoding, this paper introduces a class of orthogonal complete piecewise k-degree polynomials in L［0,1］ (so-called U-system). Firstly, a class of new U-orthogonal transform is constructed using U-orthogonal basis, and an algorithm of image coding based on U-orthogonal transform is presented by investigating 3-degree U-orthogonal transform (so-called U3). Secondly, two methods of calculating discrete U-orthogonal transform matrices are established, and the fast U-transform is derived from symmetrical characteristic of U-transform matrices. Thirdly, coding gain and de-correlation efficiency of U3 are studied, and then JPEG algorithm is realized using U3 instead of DCT. The experiments show coding gain and de-correlation efficiency of U3 are close to that of DCT, and the computational complexity of U-transform is approximate to that of DCT which computed using fast Fourier transform algorithm. Moreover the effect of reconstructed image from our scheme is comparable to that of decoded image from baseline JEPG. So it is effective to apply U-transform, which may be used extensively in the application of video coding, to the field of image compression.关键词：U-orthogonal transform;discrete cosine transform;image encoding;de-correlation rate;coding gain3783|301|0更新时间：2024-05-07
- 摘要：Because of vast amounts of information, compression of video signals before saving and transmission is necessary. Communication channels are not error free and, consequently, the encoded bit streams are vulnerable to transmission errors-usually causing loss of data and/or loss of synchronization. Therefore, a temporal error concealment algorithm based on interframe matching mean and variance is presented. The algorithm searches the best match block by matching the pixels adjacent to the error block, and then calculates the matching mean and variance of difference between different neighborhoods. From the result, the motion characteristics are estimated and different candidate blocks are chosen for different motion characteristics. Experimental results show that both the visual quality and PSNR of the image processed using the proposed error concealment method are better than other temporal algorithms.关键词：interframe matching;error concealment;mean;variance3271|241|0更新时间：2024-05-07
- 摘要：To reduce the influence on video decoding quality induced by transmition error, This paper analyzes several classical error concealment methods and evaluates them on AVS-M. Based on the test result, an adaptive spatial-temporal error concealment (ASTEC) algorithm is proposed. According to the boundary matching degree of switch between spatial/temporal concealment methods adaptively. Simulation results show that, compared with the original spatial and temporal method, the proposed algorithum achieves better performance, especially on the quality of reconstructed image.关键词：video communication;AVS-M;error concealment;TMBMA2848|204|0更新时间：2024-05-07
- 摘要：Distributed video coding(DVC) based on ideal ARQ is unrealistic. And little attention has been paid to the correlation between side information and distributed frames which has large influence on the compression performance. In this paper，through statistical analysis of the similarity between side information and distributed frames，a PSNR-based semi-feedback DVC is proposed. In order to fulfill the applications without using feedback channels, a PSNR-based unidirectional DVC is proposed too. Simulation results demonstrate that，the performances of semi-feedback and unidirectional DVC based on PSNR can significantly reduce decoding complexity，and are practical as well as flexible to control rates，while almost maintaining the same encoding complexity. But the decoded video quality is a little inferior to the ideal ARQ DVC.关键词：distributed video coding;Turbo codes;semi-feedback;unidirectional transmission;side information3162|223|0更新时间：2024-05-07
- 摘要：Based on the analysis of the reason that the TSS algorithm is unfit for searching low drastic video sequence，an extensible three step search algorithm（TSS）is proposed based on the prediction of initial point and enhanced searching in small range .The proposed algorithm has a significant improvement over the TSS both on matching precision and the reduced number of searching points，which enhances point searching in small range effectively，and the hardware structure mapping to the algorithm is easy to be implemented at the same time. The algorithm is also very fit for Variable Block-Size Integer Motion Estimation.关键词：median prediction;enhanced searching in small range;rectangle template;VBSME2980|284|0更新时间：2024-05-07
- 摘要：One advantage of C-V model among the variational level set methods is that it can detect image boundaries which were not defined by gradient. However, when detecting these type boundaries, the C-V model only consider the mean value of each region without local information, so though the C-V model can get non-gradient defined image boundary, its segmentation result contains errors. The above problem is solved by importing the motion factor to the C-V model in this paper. Where, the motion factor is defined as a function of local convexities of image. By adjusting parameters of the motion factor, the novel model can adjust the height of its 0-level set, i.e., can make the 0-level set get close to the plane which the target belongs to, so can eliminate the partition errors. We present the partial differential model, and experiments validate the quality of the segmentations obtained.关键词：C-V model;level set method;image segmentation;motion factor3326|215|0更新时间：2024-05-07
- 摘要：For the problem that the remaining fingerprint cannot be segmented easily,the authors propose a segmentation algorithm for fingerprint images with remaining ridges based on continuously distributed directional image. The algorithm firstly calculates the point directional image,and uses the histogram filtering technology to filter directional image,and then the continuously distributed directional image can be obtained. Secondly,according to the direction inconsistency on the location where the remaining ridges border on the fingerprints in the foreground,we mark out the boundary between the remaining fingerprint and normal ridge based on the sharp directional information change. After dealing with false points in the post-processing,these true boundary points are used for curve fitting. Finally we remove the remaining fingerprint. Experiments have shown that such a continuously distributed directional image transits smoothly and naturally. It not only has good continuity,gradual change and noise immunity,but also has very high accuracy. And it can effectively segment the remaining fingerprint from the foreground.关键词：fingerprint segmentation;point directional;remaining fingerprint;continuously directional image3135|561|0更新时间：2024-05-07
- 摘要：Recently, video vehicle tracking as a key technology of intelligent transportation system(ITS) has got more attention. This paper introduces a video vehicle tracking algorithm based on Kalman and particle filter. The algorithm improves the traditional particle filter, whose non-linear and non-Gaussian may result in non-robustness of tracking process, the algorithm uses the targets color histogram statistical model based on the key regional to model video vehicle, and applies it to update Kalman filter. Then through the use of Mean Shift algorithm, the Kalman filter is added to the particle filter to calibrated the vehicle running tracking so that the experiment achieves a partial linear filtering, maintaining tracking system as a whole on the non-linear and non-Gaussian, and at the same time it takes into account the local characteristics of a linear Gaussian. Experimental results show that the proposed method in comparison with the traditional particle filtering can be more accurate on tracking of vehicles and ensure the robustness of performance in a complex environment.4399|337|0更新时间：2024-05-07
- 摘要：License plate(LP) location is key to vehicle license plate auto recognition system. In this paper, we propose a novel LP location method on RGB color space, which includes four steps: color feature extraction, feature images binaryzation, area filling and de-noising with morphology processing, plate candidate regions verification. Traditional color feature extraction is sensitive to iluminition based on which, we propose a new color feature extraction method that doesnt include the intensity component. For plate candidate regions verification, we discards some geometry features such as the area, length/width ratio, rectangle degree etc, those of which are widely used but sensitive to image size, and we utilizese the special characteristic of LP on character number and character arrangement uniformity to verify the candidate regions. 605 images with various size and illumination condition were tested, success rate of LP location was over 96%, which shows the color feature extraction method is robust to illumination change, and the plate verification is suitable for various image size.关键词：license plate location;RGB color space;character uniformity of license palte;morphology3865|337|0更新时间：2024-05-07
- 摘要：It is an increasingly important topic to reconstruct 3D shape from images by structured light vision. However, their practical applications are mainly limited to low resolution, texture and sensitivity of environment illumination. This paper proposes a new color coded structured light technique for robustly reconstructing shapes from a single image. This technique works by projecting a pattern based on De Bruijn stripes with white gaps. The correspondence problem is solved using a color classification algorithm. The statistical cluster parameters of the captured image are adopted so that it is adaptive to different scenes and contexts. In this paper, the stripe boundary is located accurately by local a searching method. Additionally, a scheme is presented to achieve dense shape reconstruction by shifting the same pattern. Practical experimental results are provided to demonstrate the performance of the proposed method.关键词：shape reconstruction;color classification;shifted one-shot pattern;color structured light2949|295|0更新时间：2024-05-07
- 摘要：In this paper, we propose a novel image clustering algorithm for effective image retrieval in Web2.0 tag-space. Different users may use different tags to describe the same object, causing inconsistency in tagging. Our algorithm capture the semantically similar tags to perform query expansion, and retrieve the candidate images which are possibly relevant to the query. The candidate tags can be shortlisted according to their tag relevances to the query tags. The shortlisted tags are then clustered on-the-fly using a graph partitioning algorithm. The candidate images are clustered based on the tag cluster results. The proposed algorithm is implemented in a prototype system called PivotBrowser. Experiment results performed on a large scale images that random downloaded from Flickr reveal that our proposal effectively address the inconsistency and ambiguity problems in tag-space image retrieval, and provide improved user satisfactory.关键词：tag;inconsistency;ambiguity;relevancy3143|236|0更新时间：2024-05-07
- 摘要：In this paper, we propose a novel 3D model retrieval algorithm based on Relative angle-distribution and clustering (RAC). A geometric feature vector based on relative-angle distribution (RAD) of surface points is defined. The experimental results demonstrate that RAD is a good global feature for shape classification. To reduce the feature dimensions and improve the computational efficiency, clustering is employed. The model classification results show that compared with other methods, our algorithm RAC (relative angle clustering) achieves better retrieval accuracy and efficiency.关键词：model classification;relative-angle;histogram of angle;clustering3405|258|0更新时间：2024-05-07
- 摘要：Visual salient objects detection is an important fundamental application research of visual attention mechanism. It plays an important role in image retrieval, scene analysis, image annotation and object recognition. This paper proposes a novel approach for visual salient objects detection in natural scenes based on Treisman’s feature integration theory and Koch’s framework. In this approach, a visual saliency model for colored natural scenes is proposed and different feature saliencies are considered and computed. Then an effective method is given to extract salient objects from an integrated saliency map which is combined by different feature saliency maps. Comparing with Itti’s model, the experimental results indicate that not only the detection speed of this approach is faster, but also this approach can separate visual salient objects from their backgrounds more accurately and more efficiently. From this aspect, the approach in this paper is more similar to human’s real visual attention process than Itti’s model.关键词：salient object detection;visual saliency;visual attention;natural scene3751|284|0更新时间：2024-05-07
- 摘要：Recently, SVMs(support vector machines) have been widely used in image retrieval as a method to improve the retrieval performance. However, conventional SVMs encounter four problems: small size of positive samples, asymmetry problem of training samples, over-fitting and weakly real-time. To solve these problems, an asymmetric bagging based fuzzy support vector machine (AB-FSVM) is proposed. An asymmetric bagging is made to negative samples, and then based on fuzzy theory and SVM, the retrieval images are gotten. Experimental results based on a set of Corel images show that the proposed system performs much better than the previous methods, especially when the size of positive samples is small.关键词：content-based image retrieval (CBIR);asymmetric bagging (AB);fuzzy support vector machine (FSVM)3329|203|0更新时间：2024-05-07
- 摘要：The existing projector relies on a high quality screen for exhibiting a remarkable image, which would be affected by the reflecting and geometric characteristics of the screen. Aiming to relax this confinement, the projector-camera system is then developed to compensate and correct, in radiometry, the shortcomings of the display quality on the screen, in which the geometric registration between the pixels of projection and camera images becomes a foundational and significant issue. Based on the methods of lump sampling, morphologic processing, identifying and locating the sampling patches, as well as binary polynomial fitting method, a registration algorithm was achieved. The established algorithm was tested under various scenes of different display surfaces, including both flat and curved screens, together with clean and colour-patterned screens. The experimental results indicated that the mean deviation of image registration was between 0.2 to 1.0 pixel units, regardless of the change in sampling quantity. Meanwhile, the executive time of the registration was less than 10 seconds. This sound efficiency and high precision implied this algorithm is applicable to practical demands.关键词：projector-camera system;image registration;geometric mapping;binarization3162|182|0更新时间：2024-05-07
- 摘要：The obscuring effect of clouds is one of the major factors which restrict the observation capabilities of optical remote sensing satellites. In this paper, aimed at image restoration for optical remote sensing images covered by thin clouds, three contributions have been made. Firstly, with characteristics of clouds, an enhanced linear mixing model has been proposed, in which the influence of clouds on the measured spectra has been presented explicitly. Secondly, two spectral unmixing based image restoration methods have been given, namely the direct elimination method (DEM) and the abundance adjusting method (AAM). At last, with different combinations of the two spectral unmixing algorisms VCA/MDC-NMF and two image restoration methods DEM/AAM, the capabilities and results of relevant methods are analyzed qualitatively and quantitatively, using both simulated and real datasets. Experimental results show that, the combination of MDC-NMF and AAM can achieve the best image restoration result.关键词：spectral unmixing;image restoration;thin cloud;optical remote sensing3546|254|0更新时间：2024-05-07
- 摘要：Mixed pixels are always the case in remote sensed images, and how to analysis and explain mixed pixels is of importance in remote sensing applications. Sub-pixel mapping is a technique designed to obtain the spatial distribution of the classes inside the pixels with information of different endmembers to improve the accuracy of the classification. In this paper, a new BPMAP model is introduced by combination of the neural network and super-resolution reconstructed technology. The spatial distribution of the sub-pixel can be determined by establishing of observation model between the high-resolution and the low-resolution images after the neural network mapping; with restricted by Maximum A Posteriori (MAP) algorithm. The proposed model was tested on both simple synthetic image and ETM image in the three Gorges area. Results indicate that this method can mapping sub-pixel efficiently, and better performance was observed compared to that of the original ANN model.关键词：mixed pixels;super-resolution;BPNN model;MAP;observation model3669|253|0更新时间：2024-05-07
- 摘要：City road traffic system is characterized as a system of nonlinearity, uncertainty and spatial-temporal correlation, which makes traffic system parameters description and knowledge extraction difficult, and results in current short-term traffic forecast methods can not obtain satisfactory accuracy. This paper presents a hybrid multiple-kernel support vector machine model (MSVM) for conducting short-term traffic forecast. With statistical analysis of large amounts of traffic condition data samples, the proposed model not only has a capacity of recognizing and dealing with different types of input data separately, but also takes advantages of global optimization, generalization and adaptability of support vector machine. Moreover, the parameters of the hybrid model is optimized with an improved particle swarm algorithm (PSO). Aiming at the linear correlation between real time and historical traffic condition, the nonlinear correlation between real time and previous time period, and also up and downstream traffic condition, the proposed model uses a linear kernel to extract the linear pattern of traffic flow and a nonlinear kernel to map the nonlinear pattern of residuals from the linear kernel. Both the historical regularity and time-variation characteristics of city road traffic are considered in the MSVM model so as to obtain the knowledge from the influential factors of real time traffic condition in order to improve forecast accuracy. The experimental results show that the proposed model behaves satisfactory performance and robustness, and has a good potential for applications.关键词：short-term traffic forecast;support vector machine;multiple kernels;particle swarm algorithm3591|211|0更新时间：2024-05-07
- 摘要：In this paper, we introduce a series of algorithms for vector map overlay: We proposed an algorithm for large amounts of points inclusion test, which first preprocesses the polygon and then adopts the ray-crossing idea for each point inclusion test. We proposed an algorithm for large amounts of lines clipping, which builds indexes on the whole line and the segment, and avoids unnecessary intersection-computing. For map overlay on polygons, we use the improved plane sweep algorithm to get all the intersection points, and the complete categories on distribution of segments passing the same point cover all the special cases of intersection points. We record the ID in the process of constructing the result rings, which simplifies the two processes-finding outer ring for holes and attribute propagation. All the geometric algorithms and the corresponding overlay functions are implemented, and the comparisons with the state-of-the-art algorithms and ArcGIS’s function demonstrate its correctness, robustness, efficiency and usability. The above functions have been used in GIS platform and work well.关键词：GIS;spatial analysis;overlay analysis;vector map overlay;point inclusion test;line clipping3620|203|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