最新刊期
卷 14 ， 期 4 ， 2009
- 摘要：In remote sensing image，the event of mixed pixel can be presented by fuzzy extended probability measures.These measures are considered by fuzzy classification.However，due to the effect of nonlinear factor and the complexity of nature materials，spectral variability occurs to the endmembers，which leads to the primary error in the process of spectral unmixing.This paper focused on the fuzzy presentation on the endmember spectral variability by possibility distribution for remote sensing image: on the assumption of the multidimensional normal distribution，chi-square distribution is proposed for the fuzzy presentation.Based on the chi-square distribution，a possibility distribution fuzzy partition arithmetic is developed for fuzzy partition of endmembers in the whole image，which presents the endmember spectral variability.In order to evaluate the performance of the algorithm，an experiment using real hyperspectral image is demonstrated in this paper.The proposed algorithm clearly reveals the spectral variability of endmembers in the image，which provides valuable insight into the accuracy of mixed pixel analysis.关键词：fuzzy subset;possibility distribution;mixed pixel analyze;endmember;spectral variability5730|329|0更新时间：2024-05-07
- 摘要：With the development of remote sensing technology，using the remote sensing images for monitoring land and resources has become a hotspot.Image difference and post-classification analysis are two normal change detection methods.However，there are many disadvantages in these methods.A novel remote sensing image change detection algorithm based on Fuzzy C-Means is presented.The improved FCM algorithm is employed to classify remote sensing image.Then a multi-band integration change mask method is employed to execute change detection.Experiments show that the method is effective，and improves the veracity of change detection.The change detection results can offer an important reference for land and resources survey.关键词：change detection;fuzzy clustering;image processing;remote sensing5518|364|0更新时间：2024-05-07
- 摘要：The hyperbolic echo feature in Ground Penetrating Radar(GPR) image is one of the most important basis for pipelines recognition.According to the GPR image’s wavelet spectrum information，this paper presents a new ROI extraction method based on GPR gradient image.First for highlighting the hyperbolic echo regions and suppressing the background region in GPR image better，we revise the differential parameter of obtaining the GPR gradient image.Then we use the histogram to get the threshold value for the binarization process to the gradient image.After the binary gradient image is expanded，the connected region detecting algorithm is selected for region division.At last，small area regions are eliminated as clutter regions.Then the hyperbolic echo regions are extracted from GPR image.This new ROI extraction method has high calculate speed and accuracy，and is easy to be realized in actual detection analysis.关键词：gradient image;ground penetrating radar(GPR);ROI extraction;hyperbolic feature5467|287|0更新时间：2024-05-07
- 摘要：According to the relation of bridge with water，an object-oriented method to extract bridges from high resolution remote sensing image is put forward as follows.First，use the object-oriented image analysis method to classify the IKNOS image into two classes，water and the land.Second, export all the image objects of land class to a vector shape file.Then binarize the classified image as two parts，the land with white color and the water with black color.Fourth,process the binarized image based on mathematic morphology to get the connected water object.Fifth, extract the water object to a vector file.Finally，the destired bridge is extracted through the intersection of the land and water vector files.An experiment indicates that the method is of universal application with the features of simplicity，small amount of computation and high efficiency.关键词：object-oriented;high resolution;spatial analyses;bridge extraction5920|288|0更新时间：2024-05-07
- 摘要：In high-resolution remote sensing images the gray-scale and texture features，of the ships docking in harbor area，are similar to that of the coast，so it is difficult to automatically detect them.A automatic ship detection method，which is based on a neighborhood self-similarity local salient feature extractor and further space analysis，combined with the utility of prior information，is presented. Experiments show that this method can detect ships in the harbor area accurately and efficiently，and is of high practicability and robustness.关键词：high-resolution remote sensing images;neighborhood self-similarity local feature;feature space analysis;prior information5791|317|0更新时间：2024-05-07
- 摘要：In this paper，the Mumford-Shah model is used for the recognition and classification of satellite cloud imagry.Considering the characteristics of satellite cloud imagery，an improved Mumford-Shah model based on the kernel gray of object is presented.Using this model，high cloud and middle cloud and low cloud is classified from the one-channel cloud image.For the multi-channel image，the vector-valued image segmentation method using Mumford-Shah model based on the kernel gray of object is presented.The experimental results show that our method can get the precise position of middle cloud and low cloud in the infrared channel and visible-light channel.关键词：satellite cloud image;image segmentation;Mumford-Shah model;vector-valued image5488|313|0更新时间：2024-05-07
- 摘要：The generalized model characterizing most remotely sensed data pixel-level fusion techniques is very important for theory analysis and application.According to the imaging mechanism and the ideal pan-sharpening results of multi-spectral image，a generalized model for remote sensing data fusion is presented，which can clearly describe the mathematical relationship among original multi-spectral image，the spatial details extracted from high-resolution panchromatic image，and the adopted fusion strategy.Also three types of fusion algorithm are translated into the generalized model using mathematical expression.The implementation technology based on the generalized model is developed，which only calculates the variables affecting the fusion results instead of all variables.Then these calculation methods of the two key variables are listed for most common pixel-level fusion algorithms.Compared to other models, the generalized model is comprehensive and adaptive.Analyzing implementation steps and comparing with the regular fusion results，the implementation technology which can be applied to most fusion algorithms can reduce computation complexity and save calculation time，which is helpful to promote application of remote sensing data fusion.关键词：Remotely sensed data;pixel-level fusion;generalized model;implementation technology5485|317|0更新时间：2024-05-07
- 摘要：An affine Invariant Feature is proposed for remote sensing image registration.The key points are extracted from image as local extrema both in scale space differential and 2D Laplacian.For each key point a series of affine parameters templates are applied to transform the neighborhood pixels，which produce several slices so that projection distortion could be mitigated.Local invariant feature descriptors are extracted from these slices using scale-space invariant features transform (SIFT).An initial matching set is obtained by matching SIFT features using a special similarity criterion.The mismatched key points are removed with greedy algorithm iteratively，and then the matching set is modified.The image transformation parameters are estimated based on this modified set.The performance of this algorithm is tested under several interferences including affine transformation，partly barrier，tonal distortion，loss of contrast，and Gaussian noise.This method is confirmed with some filed satellite images.关键词：templates;scale invariant feature;affine invariant feature;similarity criterion;affine transformation;remote sensing image;random sample consensus(RANSAC)6156|317|0更新时间：2024-05-07
- 摘要：Texture is an important visual cue. It widely exists in images and is hard to describe. Texture classification and segmentation is a hot research field in image processing. Due to its importance in texture classification and segmentation, texture feature extraction has received considerable attention during the past decades and numerous approaches have been presented in the literature. Based on extensive investigation of existing articles, the development history and current challenges of texture feature extraction methods are reviewed in details. First, a comprehensive overview of texture feature extraction methods is presented in this paper. Then, classification and comparisons of texture feature extraction methods are given. At last, conclusions and future development of texture feature extraction are discussed.关键词：image texture;texture feature extraction;gray level co-occurrence matrix;Markov random field;fractal;wavelet;Gabor filter;autoregressive model21099|1042|0更新时间：2024-05-07
- 摘要：In the multi-view video system, random access performance of client is an important performance of interactivity. When users access one frame of multi-view video sequence randomly, it is necessary for the multi-view video system to know how to find and decode the corresponding frame quickly and efficiently. To improve random access performance of multi-view video system, this paper first analyzes previous decoding path calculation methods of random access, and discusses the problem of non-optimal decoding path. A concept of frame level matrix is presented for decoding path computation, and then a fast decoding path computation algorithm is proposed for the H.264-based multi-view video coding with the prediction structure of HBP（hierarchical B pictures). Experimental results show that the proposed fast decoding path computation algorithm can obtain the optimal decoding path, so it effectively reduces the decoder complexity and improves random access performance in the multi-view video system.关键词：Multi-view video system;random access;decoding path;frame level matrix4339|288|0更新时间：2024-05-07
- 摘要：Due to its effectiveness for removing impulse noise and preserving detail features, median filtering has long been a popular tool of filtering algorithm. But in practice, an important issue of applying median filtering is the filtering speed. In this paper, a fast median algorithm based on measure-integral is proposed. A step function is employed to expand the array for median, then the relationship between median and measure-integral is deduced and an algorithm is gained by it. To the question that the compute time of the algorithm increases rapidly when the values of the array or the function range become large, a method of compress the measure space is put forward, which is extended to multi-scale compress method at last. Experiments show that multi-scale space compressed measure-integral based median computation（MCMIM) has higher processing speed and can be combined with most of the earlier improved median filters.关键词：median computation;measure;measure-integral;step function;space compression4000|230|0更新时间：2024-05-07
- 摘要：In order to let the CCD recording and the reconstruction of the Fresnel hologram run well, a algorithm of numerical reconstruction is designed. Fresnel digital hologram is recorded directly with a CCD target instead of a conventional holographic plate on in-line configuration, and stored as a BitMap file in the computer. A real image of the original object can be reconstructed from the hologram by the numerical method instead of the optical diffraction. The paper deals with an experimental design of the in-line holographic recording with a CCD and the algorithm of numerical reconstruction and the un-destructibility of the digital hologram are verified with a computer program.关键词：charge-coupled device(CCD);numerical reconstruction;Fourier transformation;Fresnel diffraction field9098|302|0更新时间：2024-05-07
- 摘要：Comparing with the linear image restoration algorithms, the image restoration algorithm based on maximum entropy can obtain better performance.However, it has slow convergence rate.To improve the convergence speed of the maximum entropy based image restoration algorithm, we firstly present the aperiodic model of deconvolution, and then a fuzzy inference system is introduced to determine the iterative step size online.Since we adopt a variable step size, the convergence speed is significantly improved.The computer simulation results show that the proposed algorithm for image restoration has a faster convergence speed and yields improved restoration performance.关键词：maximum entropy;image restoration;aperiodic model of deconvolution;fuzzy inference system4926|310|0更新时间：2024-05-07
- 摘要：It is very urgent to build an evaluation criterion in image compression, enhancement, restoration and analysis fields.In this paper, fractal dimension is utilized to be an image quality assessment index and describe image luminance and texture characteristics from non-linear aspects.Lacunarity changes can also describe image quality.Therefore fractal dimension difference and lacunarity difference are integrated into one single mathematics model by linear regression analysis.Experiment results show that against traditional model such as PSNR, SSIM and so on, the proposed approach can not only assess different distortion types, same distortion type with different distortion levers accurately, but also has stronger correlation with MOS, more agreement with the perceptual of human beings, and can assess image quality accurately and effectively.关键词：quality assessment;fractal dimension;lacunarity;SSIM;PSNR3690|283|0更新时间：2024-05-07
- 摘要：In image-aided terminal inertial guidance system geometric distortion is induced in missile-borne image with varying missile attitude.The affine projection maintains the parallel structure of the objects，on which a method of geometric distortion correction of missile-borne image is proposed，and the affine projection is applied to the terminal corrected images. The function of the perspective target and its affine projection with the missile pitch is deduced based on the perspective geometry character of the missile-borne camera. That is used to achieve the terminal corrected image using the integrated information of the inertial navigation system (INS) and image by setting appropriate reference frames of the two systems. The results of simulation demonstrate that the method proposed is feasible to correct the geometric distortion induced by the missile attitude and increase the object position accuracy. Furthermore，the method is not sensitive to the INS error and consumes little time.关键词：affine projection;geometric distortion;missile-borne image;image-aided terminal navigation3652|272|0更新时间：2024-05-07
- 摘要：This paper makes some improvements on Non Local Means（NL-Means) image denoising algorithm. A quantitative method is given to estimate the optimal filter parameter h. Noise variance is estimated from noise image, and h parameter is estimated from this variance and noise image standard deviation. Based on the symmetry of weighted Euclidean distance, the most complex distance computation was halved for every pair of pixels, so the computation complexity was reduced to about half of original NL-Means without performance decline. Experimental results on several images show that our adaptive non local means algorithm（ANL-Means) gives nearly best performance with only about half computation time if using original NL-Means.关键词：image denoising;gauss distribution;noise;variance5241|382|0更新时间：2024-05-07
- 摘要：Noise reduction is an important research topic of the infrared image processing，but the commonly used noise reduction methods will cause the loss of the details.In order to effectively reduce the noise and protect the edge at the same time，a new anisotropic diffusion algorithm based on the information measure and the edge membership is introduced. The core content of this algorithm is to divide the image into two areas，the edge area and the non-edge area. While the conventional P-M diffusion equation is used into the non-edge area to filter the noise，and the nonlinear diffusion equation based on the information measure is used into the edge area to filter the noise and enhance the edge. The final results show that this algorithms PSNR，MAE and radiometric resolution are better than the traditional algorithms. So this algorithm has practicality and potential application value.关键词：image noise inhibition;anisotropic diffusion;information measure;edge membership3912|249|0更新时间：2024-05-07
- 摘要：The CT images segmentation is one of key technologies for the 3D reconstruction and quantitative analysis of plant root system in situ. In order to improve the precision and efficiency of images segmentation,in accordance with the inherent indistinction of CT images, a fuzzy thresholding algorithm was implemented with the criterion of maximum fuzzy entropy and genetic algorithm. The initial thresholds were obtained with histogram analysis. The CT images were divided into several different regions fuzzily through designing a simple fuzzy neighborhood function. And according to the criterion of maximum fuzzy entropy, a genetic algorithm was used to find out the best thresholds of CT images segmentation. The result of programming test shows that the algorithm is effective to improve the precision and efficiency of root CT images segmentation.关键词：genetic algorithm;fuzzy segmentation;CT images;root in situ3840|246|0更新时间：2024-05-07
- 摘要：Because the texture and color the tongue and the lip is very close, it is difficult to extract the accurate contour of the tongue body only with the active contour model： Snake.This paper introduces a method that combines radial edge detection with Snakes model to solve this problem.Firstly,we use radial edge detection to get rough contour of the tongue.Then, we use pair-color-remove to remove the lip.Finally, we use snake to get the exact contour of the tongue. The experiment results are satisfying.关键词：tongue image segmentation;radial edge detection;Snake model3688|270|0更新时间：2024-05-07
- 摘要：An algorithm for detecting forest-fire was proposed, which was based on multiple features fusion of smoke and spatial accuracy compensation.Firstly the detection results derived separately from motion analysis and wavelet analysis were fused, and then the fusion results were cumulated using smoke color and motion cues to realize the smoke detection.The smoke profiles were extracted with spatial accuracy compensation based on the scene horizon.We have tested our algorithm on a number of image sequences, and the results show that our algorithm can provide significant improvements over accuracy of forest-fire detection, decrease false alarm rate and enhance the robustness.This approach has important application value in forest-fire surveillance system.关键词：smoke detection;wavelet analysis;multiple features fusion;special accuracy compensation3733|271|0更新时间：2024-05-07
- 摘要：Face recognition is one of the hottest research areas in pattern recognition.Many face recognition methods have been proposed.Recently，a lot of learning algorithms have been proposed and applied it in face recognition tasks successfully.Among them，locality preserving projections (LPP) is one of the most effective methods.In this paper，we propose a new face recognition method——orthogonal discriminant locality preserving projections with Schur decomposition (ODLPPS).In comparison with LPP，the objective function of the proposed method incorporates scatter difference information of between-class and within-class and makes the basic vectors orthogonal.Experimental results on ORL and Yale demonstrate the proposed algorithm achieves better face recognition performance than some existing methods such as eigenface,Fisherface,LPP and orthogonal LPP(OLPP).3832|238|0更新时间：2024-05-07
- 摘要：A novel approach for detecting and recognising various traffic sign shapes in outdoor environments is proposed.In order to reduce the influence of digital noise, separate and extract the shape of each individual traffic sign, the external boundaries of traffic signs segmented based on color information are simplified and decomposed through discrete curve evolution.The evolution level is determined by the arc similarity measure in tangent space.The recognition of a candidate shape is achieved through matching against templates.The minimum geometric difference between a candidate shape and templates is utilized to determine the classification of the candidate shape.The experimental results justify that the proposed algorithm is translation, rotation, and scaling invariant, and gives reliable shape recognition in complex traffic scenes.关键词：color-based segmentation;template matching;traffic sign shape recognition3898|321|0更新时间：2024-05-07
- 摘要：Non-rigid image registration algorithm based on viscous fluid model is an appropriate method for registering objects with large difference.The critical part of the model is the viscous fluid kinematical partial differential equations（PDEs), which are very time-consuming to be solved by direct discretization combined with successive over-relaxation method(SOR).In order to reduce the time cost, a fast method based on viscous fluid B-spline model is developed.Firstly, B-spline is used to model the vector fields in the PDEs, changing the unknowns to the coefficients of B-spline.Secondly, by using of fast Fourier Transform(FFT) and some special attributes of B-spline, formulas of directly calculating the coefficients are deduced.Experimental results show that new method is a fast non-rigid registration method without precision loss.关键词：non-rigid registration;viscous fluid model;B-spline;FFT large deformation3478|270|0更新时间：2024-05-07
- 摘要：Stereo matching is studied in the paper, which has been a very hot research topic at present.An energy based algorithm inspired by PDE and machine vision theory is proposed to estimate a dense disparity map between two images.Firstly, the effects of matching pairs at various relative positions to the attachment item are analyzed.Secondly, anisotropic heat diffusion equation adapts to disparity map is presented, which inherited from the ability of the Alvarez defining regularization item that keeping the discontinuities across the boundaries of the image and smoothing disparity inside the boundary.In addition image noise shielded function and second order directional derivative are introduced to separately control disparity diffusion velocity of different area and diffusion direction of edge position.At last, new energy function according to our approach is defined, adopting the output of the area stereo matching method as the initial value and steepest descent is exploited to solve the energy functional.Experiment results demonstrate the effectiveness of our approach, both in the visual effect and 3D depth retrieval.关键词：stereo matching;dense disparity map;partial derivative equation (PDE);regularization item;attachment item4723|273|0更新时间：2024-05-07
- 摘要：An approach to addressing the stereo correspondence problem is presented using particle swarm optimization algorithm with adaptive hierarchy to obtain a dense disparity map.Firstly, the image features are precisely extracted by using SIFT feature detection, and accurately matched by using SIFT matching algorithm, so the disparity range is rightly and easily calculated from matching features.Secondly, according to restriction of the image size and the disparity range，the coarse to fine adaptive hierarchical image pyramid is built to search fast and reduce wrong matching.Thirdly, a regulation parameter varying with matching window is used to give different power for grayness and smoothness data in optimization function while the matching window is different in dissimilar supporting areas, and improved particle swarm optimization algorithm with variation operation for integer is used to find the fittest solution from a set of potential disparity maps avoiding Genetic algorithm’s blind searching and easy getting in local best solutions.Finally, experimental results on synthetic and real images show that the proposed approach performs dense disparity estimation accurately and quickly.关键词：adaptive hierarchy;particle swarm optimization;disparity map;stereo correspondence4010|255|0更新时间：2024-05-07
- 摘要：A variety of models for plant leaf morphogenesis have been reported, but none of them can be generated with great accuracy or level of detail. In this paper B-spline curve is used to describe the silhouette and midvein of plant leaf, and the leaf blade surface is meshed with Delaunay triangulation. Then an adaptive refinement scheme is introduced to smooth the triangular mesh. This may provide a common method for describing the morphology of plant leaf blade. B-spline curves which pass through the control points gives an intuitive mechanism to model the silhouette of a wide variety of plant leaf, and the usage of Delaunay triangulation solves the difficulty of meshing complex leaf blade such as lobbed leaf. In addition, our adaptive subdivision scheme can generate relative smooth leaf surface, and the curled effects can be obtained basing on the subdivided surface. The experimental results verify the validity of the proposed method.关键词：morphological modeling;3D plant leaf;B-splines;subdivision surface3775|295|0更新时间：2024-05-07
- 摘要：Aesthetic virtual environment cannot be satisfied if using traditional image based rendering. By introducing non-photorealistic rendering into the Virtual Reality, this paper presents a novel method for aesthetic virtual environment generation. First, structure tensor is calculated to characterize local structure of image.Then, total variation partial difference equation is imposed for smoothing vector field. Based on this we use line integral convolution for guiding the stroke to generate image with fluctuant sense. Last, merging with 3D scene, roam-able aesthetic virtual environment can be generated. Experiment Results show that we can simulate fluctuant sense image, and thus special style of aesthetic virtual environment can be satisfied using our algorithm.关键词：image-based rendering;non-photorealistic rendering;structure tensor field;line integral convolution3982|328|0更新时间：2024-05-07
- 摘要：Traditional plane based modeling and rendering of ocean wave lacks in consider ations on the effect of globe curvature in detail, thus fais to realistically simulate the change of shape, color of the ocean waves on sphere. In this paper, considering sphere shapes influence and free viewpoint, a new method of modeling and rendering of ocean wave is presented. Its kernel is the modeling of ocean waves on sphere and the screen-subdivision algorithm. The ocean wave model based on the knowledge of ocean, analytic geometry and the characters of digital globe is set up first. It reduces the computation obviously and resolves the “compress phenomenon” at high latitude. Then, relying on a procedural ocean wave model, the method restricts computations to the visible part of the ocean surface using the screen-subdivision algorithm and adapts the geometric resolution depending on the viewing distance. The “border slit” on screen can be removed by the “Clamp Sphere”. Finally, the realistic scenes are real-time rendered based on GPU. It allows the user to interactively fly over an unbounded animated ocean.关键词：ocean wave modeling and rendering;sphere based modeling of ocean waves;view-dependent levels of detail;screen-subdivision adaptive algorithm3848|276|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