最新刊期
卷 13 ， 期 3 ， 2008
- 摘要：Speckle reduction is always one of the important tasks in the application of SAR(Synthetic Aperture Radar) images . With the development and application of SAR, many speckle filtering algorithms have been proposed aimed at speckle reduction in lately twenties of years. These algorithms can be divided into two categories: one is based on single channel data, and the other is based on multi channel data. Comprehensively and systemically, this paper investigates the principles of speckle reduction in polarimetric SAR images. All of the algorithms proposed in recent years are summarized and their performances are analyzed and compared.关键词：SAR image;polarimetric;speckle reduction;review3298|211|0更新时间：2024-05-08
- 摘要：Distributed video coding is a new paradigm for video compression, with low encoding complexity and robustness to channel losses. In this paper, principle and characteristics of distributed video coding is firstly introduced, followed by two key techniques including highly effective compression and interpolation of side information. Then, recent development of the research on two main application areas of distributed video coding is reviewed, along with the future trends of distributed video coding.关键词：distributed video coding(DVC);side information;low complexity encoding;robust transmission2931|190|0更新时间：2024-05-08
- 摘要：A method for image denoising is proposed. After performing multi resolution wavelet decomposition on corrupted image, the wavelet coefficients of noise are Gaussian distribution, and have different variances in different levels. Based on this, wiener filter is applied to the wavelet coefficients on different subbands and orientations, and leave wavelet coefficients in the low frequency domain without change, to estimate the wavelet coefficients of the clean image. Then the inverse wavelet transform is applied to the modified wavelet coefficients, resulting in the denoised image. At the end, experimental results show the validity of the proposed method.[WTHZ]关键词：wavelet transformation;Wiener filtering;image denoising3013|248|0更新时间：2024-05-08
- 摘要：Comparing with the traditional image denoising methods, nonlinear diffusion has a very good performance in preserving the important characteristics of the image. Since the day the nonlinear diffusion method was proposed, a number of work has been done 〖JP2〗continuously to research and improve it. In this paper, a new nonlinear diffusion method is proposed in which the orientation information measure is used to control the diffusion coefficient. The orientation information measure can efficiently reflect the area character of the image. It also decreases the influence of noises without Gaussian convolution. The experimental results demonstrate that the proposed method works well for image denoising while edge and detail information is substantially preserved.关键词：image denoising;anisotropic diffusion;orientation information measure2519|216|0更新时间：2024-05-08
- 摘要：This paper aimed at solving the problem of using complexity histogram steganalysis to BPCS steganography(Bit Plane Complexity Segmentation Steganography) and presented an improved method. For compensating the BPCS security deficiency, we used a half pixels of “noise like” block for embedding secret information, and the other half pixels were used for adjusting complexity. Experiment results showed this method can properly counteract the attack of complexity histogram.关键词：information hiding;steganography;BPCS(Bit plane complexity segmentation);histogram3213|178|0更新时间：2024-05-08
- 摘要：In this paper, a new watermarking technique based on chaos and video contents is proposed and implemented. A chaotic sequence which is generated by a secret key as seed is automatically transformed to two value image, and eventually converted to gray image. Selected parts of the gray image are embedded in the corresponding parts of a video frame. The selection of parts is based on the histogram of the video frame so as to embed watermark into smooth parts of the video frame. This algorithm can resist some attacks such as averaging filtering, loss compression.关键词：digital video;chaos;linear collusive attack;histogram;digital watermarking2347|199|0更新时间：2024-05-08
- 摘要：To change the situation that many methods about watermarking were based on embedding one single watermark, we presented a new compound image watermarking method, which was embedding both robust watermark and fragile watermark to one image, so that we can not only convey the ownership, but also justify whether the source image is tampered. Therefore, a multipurpose watermarking system has been designed in this paper. In addition, both the watermarks need not the source image information during the process of extracting or detecting. That is to say, our algorithm is a blind one. Simulation results show that the algorithm we present is very effective, and can be widely used in practice.关键词：singular value decomposition(SVD);robust watermarking;chaotic sequence;fragile watermarking2505|184|0更新时间：2024-05-08
- 摘要：In this paper，in order to obtain more robust digital watermark scheme，firstly the theory of iterated function system（IFS）and the stochastic algorithm by which IFS attractor is constructed are proposed Secondly，by analyzing the robust character against geometric distortion of IFS attractor theoretically，a new scheme is provided，which is a asymmetric digital watermark one Thirdly，using tri point method，the paper translates the information of digital watermarking into IFS code，and the indices set can be attained by reconstructing the image of fractal watermarking and manipulating the original image relatively Finally the index set is signed and time stamped by the owner and trusted third partys private keys Using the owner and trusted third partys private keys，the watermark can be detected without the original image The experimental results show that the watermark scheme is robust against image manipulations，such as noise adding，filtering，compression and rotation关键词：digital watermark;IFS attractor;digital signing;timestamping;image manipulation2428|197|0更新时间：2024-05-08
- 摘要：Multi hypothesis motion compensated prediction has been used as bi directional prediction mode in B picture of the H264/AVC video compression standardThe optimal motion compensated prediction signal，named hypotheses，is composed of blocks in both forward and backward 〖JP2〗reference pictures，which are simply added and averaged to synthesize the final prediction signalIn the original algorithm，more than one motion vector and reference information has to be encoded and the accuracy of motion compensated prediction is still insufficientTo overcome these disadvantages，a new rate distortion optimization 〖JP〗based MHMCP algorithm is proposedAfter the optimal MBs encoding mode has been obtained by motion estimation and mode decision procedure，bi directional motion search with iterative local motion vector refinements will be startedAt the same time，two extra mv tracking modes which need not to encode additional side information will be examined，accompany with adaptive hypothesis coefficients adjustmentExperimental results show that not only the energy of the residual signal can be decreased but also the quality of the motion compensated prediction signal can be improvedTherefore better R D performance can be achieved by the proposed algorithm关键词：multi hypothesis motion compensated prediction（MHMCP）;rate distortion optimization;H.264/AVC;motion estimation2653|183|0更新时间：2024-05-08
- 摘要：Fractal image coding is a promising lossy compression technique in terms of achievable compression ratios and decoded image quality; However, it has the primary disadvantage of high computational demands resulting in unacceptably long encoding times. Most of the encoding times are spent on searching for the best matched block to each of range blocks in a usually large domain pool. This paper thus proposed an accelerating scheme by the determinants of normalized range and domain blocks, which can find out the best matched block to an input range block in a relatively small search neighborhood. Experimental results on three popular 512×512 test images showed that, depending on the search neighborhood size, the proposed algorithm not only can achieve the speed up of about 30 times with the same PSNR(peak signal to noise ratio) as the baseline fractal algorithm with the full search, but also can obtain the speed up of 1 [KG-*7]000 times or more at the cost of tolerable degradation of the decoded image quality.关键词：image compression;fractal;fractal image coding;determinant2741|219|0更新时间：2024-05-08
- 摘要：An effective approach for generating continual chaotic system by using the Digital Signal Processor(DSP) is presented in this paper. Taking the Jerk system as an example, we introduced how to implement using the DSP device, and designed analog hardware circuit. Then, the experimental results of analog and digital are given respectively at the same time. The two results are in good agreement. This confirms the feasibility of our method. A number of experiments show that this method can also be applied to other continual systems. It is convenient, flexible and easy to use in digital secure communication.关键词：Jerk system;hardware experiment;multi scroll;continual chaotic system2576|250|0更新时间：2024-05-08
- 摘要：CAMSHIFT is a color based tracking algorithm. However, it is always unstable in the case of small objects because of their limited color information. In this paper, an improved algorithm based on Monte Carlo Method is proposed, which enhances the robustness of color feature and the tracking stability by producing reasonable Gaussian random samples. Finally, experiments demonstrate the effectiveness of our method.关键词：Camshift;Monte Carlo method;visual tracking2648|331|0更新时间：2024-05-08
- 摘要：Robust and reliable vehicle detection is the first step in automotive driver assistance systems. In this paper, a new approach of model matching based on curve projection is presented. The similarity measurement is based on the weighted sum of integrity of projection, offset expectation and variance of matched point. A scheme for vehicle detection and tracking is presented based on model matching. Results for natural traffic scenes demonstrate high reliability of the proposed method.关键词：model matching;vehicle detection;tracking3565|240|0更新时间：2024-05-08
- 摘要：In the parking space detecting system, it must recognize all kinds of cars which may park on the parking plot to decide is a parking space is free or not. But vaviable shapes of cars make it difficult for computer to make decision. After drawing a black solid triangle on a parking space and detecting whether there is a solid triangle in image, it shows that whether the parking space is free. If there is no triangle in image,it shows that the parking space is taken by a car. The problem of detecting free parking space is transformed into detecting whether there is a triangle in image. It is easier to detect whether there is a triangle in image than to detect a car. Traditional Hough transform cannot effective detect whether there are triangles in image. The algorithm presented in this paper proposes an effective algorithm to detect whether there are triangles in image. It first detects the edges of image by sobel algorithm, then extracts connected edges and fills the area enclosed by the edge extracted just now. Finnaly a judgement can be made that the area filled just now is a triangle or not by the relationship of its area and its edges. When the edge extracted just now has been analyzed, it continues to extract next edges in the image until all edges in the image are extracted. If there are triangles in the image, it outputs the number of triangles in the image otherwise,it outputs zere.Exper imental results, it show that the algorithm presented in the paper has little mathematical calculation and it is fast and with small memory cost .关键词：edges extracting;connected edge;dilate;A filling;triangles detection3472|203|0更新时间：2024-05-08
- 摘要：The texts embedded in video images have obstructed video intercommunication and processing among different languages. This paper proposes an approach for automatic text detection and removal in video images based on CEMA and texture restoration. First, text regions in the video images are detected by a CEMA based algorithm. Second, we remove the detected texts from the original video images by texture based restoration technology; meanwhile, restore the background occluded by texts in video images. Experiment results show that the proposed method performs fairly well.关键词：color edge detection;mathematical morphology;text detection;text removal;image inpainting3425|242|0更新时间：2024-05-08
- 摘要：Analyzing the characteristics of the shadows in traffic video sequence, we propose a small improvement to an existing background modal based on several current shadow detection algorithms, and incorporate a novel technique for shadow detection in grayscale video sequence. We gain the background establishes a robustness and auto adapted background modal. First we filtrate the shadow fuzzily, and then re establish the images of car to remove the shadow exactly. This algorithm works well for traffic video sequence, and has a widespread application prospect.关键词：shadow detection;image process;intelligent transportation system2824|185|0更新时间：2024-05-08
- 摘要：The eye positions are very important landmarks in face recognitionIn this paper a fast hierarchical based algorithm is proposed to locate eyes from face imagesThis new algorithm involves three main stepsFirst，MER（minimum extremal region）is introduced here to locate eye roughlyThen，a three layer filter is applied to rule out false detections step by stepFinally，an accurate eye locating method is used to adjust the eye locations preciselyThis algorithm has good results in CAS PEAL，JAFFE and our own face databaseTherefore，it is considered to be an effective method关键词：eye locating;face recognition;minimum extremal region;hierarchical structure2882|176|0更新时间：2024-05-08
- 摘要：Neural network ensemble based on rough sets reduction is proposed to decrease the computing complexity of conventional feature ensemble selection algorithms. Firstly, a dynamic reduction technology, which integrates genetic algorithm and resample method, is used to get reduct sets that have stable and good generalization ability. Secondly, Multiple BP neural networks based on different reducts are built as base classifiers. According to the idea of selective ensemble, the best generalization ability neural network ensemble can be found by some search strategies. Finally, classification based on neural network ensemble can be completed by combination with vote rule. The method has been verified in the experiment of classifying Landsat 7 bands remote sensing image of chosen area. A number of feature sets of poor performance were discarded by reduction based on rough sets. Compared to conventional feature selection algorithms, the method takes less time, has lower computing complexity, and the performance is satisfying.关键词：rough sets;reduction;neural network ensemble;remote sensing image classification2280|234|0更新时间：2024-05-08
- 摘要：Gauss Markov random field model takes advantage of both image intensity and spatial information imposed by Gibbs smoothness prior to the pixel labels and thus can be used to effectively in segmenting the noisy imagesHowever it is always difficult to confirm the Gibbs penalty factor βAs usual，to get good segmentation result for every segmenting to be image，various values of β will be tested by handSo to solve this problem，this article defines a new and simple class adaptive penalty factor βIt is automatically calculated from the posterior probability and is anisotropic for each classFurthermore the model iteratively obtains their parameters estimation in the EM MAP algorithmFinally，by application of this algorithm to brain MR Image segmentation，the proposed segmentation scheme is proved effective for noisy image and at the same time it distinguishes itself by higher correct classification ratio and correct classification ratio for each class关键词：Gauss Markov random field;class adaptive;penalty factor;EM MAP;image segmentation2837|236|0更新时间：2024-05-08
- 摘要：A novel approach for video object extraction fusing temporal and gradient information is proposed. Experiments demonstrate that our method can extract the video object effectively from complex background, and can resolve the vacancy phenomenon when the foreground and background are of the same color using background eliminate method. Firstly, in the temporal domain we introduced background eliminate and adjacent frame difference method to generate elementary video object, and then we used erosion and dilation morphologic method to the video object. Secondly, in the gradient domain we detected video object’s edge using Sobel operator. Associating with the video object generated in the temporal domain, we obtained the exact video object contour’s edge. Finally, we applied heuristic search method to link video object contour’s edge points, sequentially extracted video object.关键词：video object extraction;background eliminate;edge detection;edge connection2646|216|0更新时间：2024-05-08
- 摘要：Natural image matting is an important algorithm on image processing to extract the foreground objects from the background image. This paper proposes a Markov random field(MRF) model based approach to natural image matting with complex scenes. The image is manually, divided into three regions:fore region, back region and unknown region, which is segmented into several sub regions. In each sub region, we partition the colors of neighboring background or foreground pixels into several clusters in RGB color space and assign matting label to each unknown pixel. Each label is modeled as an MRF and the matting problem is then formulated as a maximum a posteriori(MAP) estimation problem. Simulated annealing is used to find the optimal MAP estimation. The better results can be obtained under the same user interactions when the image is complex. Results of natural image matting experiments performed on complex images using this approach are shown and compared in this paper.关键词：blue screen matting;natural image matting;Markov random field;simulated annealing3343|241|0更新时间：2024-05-08
- 摘要：In order to improve the efficiency of stereo matching and resolve the problem caused by discontinuity of the disparities in one region, new pixel\|set based stereo matching algorithm using belief propagation is proposed in this paper. Firstly, the initial disparity estimate is evaluated by hierarchical belief propagation in pixel domain. Secondly, the reference image is segmented in turn according to color and disparity information. Thirdly, the segmented pixel\|sets are fitted to a set of planes based on the strategy of splitting and merging, which can eliminate the mistakes due to color segmentation. Finally, the ultimate disparity estimate is achieved in the pixel\|set domain by employing standard belief propagation. Experiments on the international benchmark demonstrate that the performance of our algorithm is comparable to the state\|of\|the\|art stereo algorithms on various data sets.关键词：stereo matching;belief propagation;image segmentation;plane fitting2808|189|0更新时间：2024-05-08
- 摘要：The problem of Pose estimation based on single camera arises in many areas of computer vision，including object recognition，site inspection and updating，object tracking and autonomous navigationMany methods are used to solve this problem nowadaysThese methods can be categorized into two groups depending on whether correspondence is known：one is to estimate the Pose parameters if the correspondence has been solved; the other is to solve the Pose and correspondence problem simultaneously〖JP〗A new method based on single camera for estimating the Pose parameters of the object is presented in this paperDifferent from the existing methods，the correspondence problem which is coupled with Pose problem is completely avoided in the proposed approachAn energy function including the Pose parameters is defined to the degree of superposition between the model projection and the object area in the imageThe optimal Pose parameters are found by optimizing the energy functionHooke Jeeves Algorithm is used to minimize the energy functionThe performance of the method is evaluated in a variety of simulations on synthetic dataThese tests show that the method is accurate and effective关键词：single camera;pose estimation;energy function;Hooke Jeeves algorithm3438|206|0更新时间：2024-05-08
- 摘要：Conventional geodetic monitoring methods have some disadvantages such as intense workload, difficulties of realizing real time monitoring and acquiring instant 3D moving deformation information in subsidence areas of mine surface. In order to cope with the practical problem, we proposed a method using digital close range photogrammetric technique to monitor subsidence areas. First, digital stereo pairs of mine surface were captured using high resolution digital camera. And then digital elevation model (DEM) was extracted by a series of photogrammetric reduction processes such as relative orientation, image matching etc. By comparing the constructed DEM with the original, the subsidence parameters such as subsidence range, depth and volume etc can be calculated. Practical research results indicated that the precision acquired by digital close range photogrammetric technique can meet the accuracy requirement under this test condition. A feasible approach is provided for subsidence monitoring of mine surface.关键词：subsidence areas;camera calibration;digital close range photogrammetry;wavelet transformation;direct linear transformation;image matching;DEM2681|217|0更新时间：2024-05-08
- 摘要：The fractal theory developed by the French scientist Mandelbort in 1970s is beneficial in many areas. It greatly expands and deepens our knowledge on irregular geometric bodies. Fractal theory quantifies these phenomena mainly by ascertaining their fractal dimensions. Box counting algorithm is the one most practical and also most frequently adopted method. The traditional box counting algorithm is based on the grid document and has some serious shortcomings, such as the distortion of the image being enlarged, the trivialness of the process and the finite of the iterative degree, etc. The vector box counting algorithm developed in this paper takes vector document as the carrier and has three advantages. First, the image will not be distorted after being enlarged. Second, the process is completely handled by computer, simple and reliable. Third, to some degree, the iterative degree can be infinite. Therefore, it can ascertain precisely the scaling space of the graph and acquire accurate fractional dimension value. This paper expounds the data structures, the process of disposing and the main functions in detail. Whats more, it proves the precision and advantages of the vector box counting algorithm by making use of Koch curve, osteoma boundary and river system. The result shows that the vector box counting algorithm is a convenient, useful and precise way of dimensional calculating method.关键词：fractal;box counting algorithm;grid;vector;Koch curve;osteoma boundary;channel systems2629|202|0更新时间：2024-05-08
- 摘要：It is a tough challenge for the geologists to model 3D shape of geological factors because of the complexity of geologic architecture and the insufficiency of exploration data. This paper presents research on modeling 3D shape of geo bodies and geo fields through 3D interpolation and iso surface constructing from limited and scatted exploration data. The Kriging algorithm，as its favorable for revealing asymmetrical and anisotropic distribution of data in the geo field, is used as the 3D interpolation method to convert the insufficient and random arrayed data set into sufficient grid data set. The iso surfaces are auto created from the scatter data through Marching cube algorithm. The above processes are achieved through VC++.net and OpenGL programming. By application of the above methods, the 3D iso surfaces of ore grade from drill data and inversely deduced resistance from magnetotellurics surveying data are simulated. Compared with the traditional methods, this produces for us good results and can be used as more convenient and effective tools for 3D visualization in geo exploration.3226|167|0更新时间：2024-05-08
- 摘要：For researching on periodicity of M J of Chaos fractal images, Firstly, a series positive integer ordered M J Chaos fractal images of the complex mapping are drawn with rotating escape time algorithm. Secondly, the distribution of general Mandelbrot set(M set) period buds are analyzed. The topological invariance on the periodic buds fibonacci sequences in the general M set are validated. With lots of computer mathematical experiments, In the end ,the correspondence is discovered between the M set parameter plane and Julia set images on the kinetic plane. And the formula on periodic orbits of general M J set is presented.关键词：Fibonacci sequences;general M J set;rotation escape time algorithm;topological invariance;periodic buds2402|176|0更新时间：2024-05-08
-
The Algorithm of Automatic Organization of Complex View Outline Information Based on Midpoint Method
摘要：The information of view outline is one of the important issues in computer Graphics.It plays an important role in automatic partitioning of engineering drawings, assembly diagram processing and dimension arrangement. Lots of experts have proposed many methods to these issues. To improve the efficiency of acquisition of view outline information and solve the problem of existing methods which require angle discriminance and much calculation , this paper puts forward automatically acquires the information of view outline based on midpoint method. At first, this method sorts the candidate path according to the quadrant of coordinate system, and then fleetly eliminates some candidate path by directly comparing their coordinate values. Thirdly, according to the midpoint method, the candidate path which was distinguished arduously can be determined finally. 〖JP2〗The midpoint method only includes shift and subtraction calculation, avoids complex angle calculation and nodical acquisition calculation, so the efficiency of this method can be improved. This method is applicable in automatic organization of complex view outline information. The result of test indicates that this method is a simple, highly efficient and reliable algorithm.关键词：information of view outline;midpoint method;candidate path;trend of path2875|200|0更新时间：2024-05-08 - 摘要：With the improvement of graphics cards performance and network bandwidth, people become more interested in the 3D simulation applications on the internet. Based on the latest development of virtual reality technology, combined with the plan of Beijing Olympic Park, the basic frame of constructing virtual Olympic Park is provided. The design of system and developing flow based on components are discussed in detail from several aspects such as function module, performance and configuration. At last an instant of web release for 3D virtual scene of Beijing Olympic Park is given and the wide foreground is shown关键词：virtual Olympic park;3D simulation;web release;components2504|188|0更新时间：2024-05-08
- 摘要：Interaction techniques based on virtual hand play an important role in applications such as human computer interaction and ergonomics test, etc. For the purpose of a naturally real time interaction between virtual hand and virtual objects, also for the purpose of the feedback force calculation, a new non linear spring model is proposed. A hand model with facility movement control and visually realistic appearance is established, the established hand model is the same as real man hand in fingers degree of freedoms. We also presented the virtual hand movement control method which is based on vector rotation. The simple spring model is improved and the grasping force is calculated with a non linear spring model. The non linear spring model is expressed in a three piecewise function. The results feedback to the users are in the form of visual rendering a quantitative analysis is done on simulation rate which is fast enough for screen refresh rate and force rendering rate. Experimental results show that virtual grasping can be executed naturally in real time.2730|226|0更新时间：2024-05-08
- 摘要：Human activity analysis is receiving increasing attention from computer vision researchers. One challenge is the segmentation of human body into meaningful body parts. A semi supervised clustering based body parts segmentation algorithm of 3D reconstructed human is presented in this paper. Firstly, we segment human body parts with the help of posture parameters of the previous frame. Then the structure information of human body is adopted to classify some points and initialize the centers of the semi supervised clustering. Finally, based on the shape of body parts, semi supervised clustering method is used to segment the body parts. In addition, body posture parameters are estimated with the segmentation result of body parts. The system is validated with IXMAS database, which includes 6 actors and 6 kinds of activity. The experimental results show that the presented algorithm can adapt the variety of views, actors and activitiesy.关键词：body parts segmentation;semi supervised clustering;posture estimation;activity analysis4512|234|0更新时间：2024-05-08
- 摘要：Finding human in images is critical for several applications in computer visionWe combine the cascade of rejection approach with the Histograms of Oriented Gradient （HOG） to form a fast and precise human detectorThe detector consists of coarse to fine linear SVM classifiersOur experiments show that our method can process average 12 frames with 320×240 image per second，while maintaining the comparable accuracy to Dalals method关键词：human detection;cost sensitive SVM;coarse to fine detector4947|224|0更新时间：2024-05-08
- 摘要：Moving object detection is an important process in intelligent video surveillance systemsOn the basis of analyzing existing detection approaches，this paper proposes a detection algorithm based on a gradient directionsFor a video image sequence，edge gradient images of each frame are achieved firstly by means of directionSecondly，the traditional frame difference is improved and it car extract the rough edges of moving objects by processing two temporal correlative frames using uint8 formatFinally，after connective areas of moving objects are obtained，the complete moving object contours are effectively detected using the gradient directionExperimental results show that，the proposed algorithm overcomes the shortcoming of not correctly 〖JP〗detecting moving objects with traditional frame difference and can effectively and accurately extract moving object contour among indoor and outdoor environments with complex backgrounds关键词：intelligent video surveillance system;moving object detection;frame difference;gradient4511|255|0更新时间：2024-05-08
- 摘要：There are two problems when tracking multiple targets in sequence images. Firstly, the targets kinematic state cannot be estimated accurately when targets abrupt maneuver happens. Secondly, the image matching method cannot discriminate the targets which belong to the same category. To resolve above problems, 〖JP3〗a novel multi target tracking method based on fusion of targets feature matching is proposed. On base of target detection in local standard deviation map(STDM), the region of interests（ROI） image matching result is used to reduce the error of state estimation matching, and the state estimation matching is used to reduce the ambiguity of image matching. Under the global optimal association cost frame, fusion of above two matches is realized to improve the accuracy of tracking, which can resolve the complex multi target tracking problem availably.关键词：multi target tracking;target detection;association cost;matching5590|234|0更新时间：2024-05-08
- 摘要：Traditional mean shift tracking algorithm based on color histogram or spatial color histogram sometimes fails to get accurate results under intricate conditions, such as scale modifications occur to the objects. That is because the histograms based on color cannot distinguish objects and background have the same color. This paper presents a new mean shift algorithm based on spatial edge orientation histograms, using space distribution and texture information as matching information. Experiments proved the new algorithm can deal with intricate conditions, such as occlusion, different luminance and scale modifications, and track objects accurately, effectively and real time. The new algorithm also overcomes the limitations of the traditional one.4604|242|0更新时间：2024-05-08
- 摘要：Detecting moving objects from video sequences is a fundamental task in Intelligent Transportation Systems（ITS）. A common approach is used to perform background subtraction，so it is important to obtain the background accuratelyThis paper presented a new background subtraction algorithm：mode method. We compared the performance of our algorithm with the common approach according to the determinant criteria we defined关键词：background subtraction;moving target;detection background;difference gray image5162|228|0更新时间：2024-05-08
- Postal code：100190
- Tel：010-58887035/58887030/58887418 Email：jig@aircas.ac.cn
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备05080539号-4
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0