最新刊期
卷 17 ， 期 4 ， 2012
- 摘要：The fuzzy c-means (FCM) clustering algorithm for image segmentation is one of the striking research fields in recent decades.Based on the analysis of the FCM algorithm,we combine the current application research in image segmentation,and we analyze and compare it in terms of measuring the expressions of the FCM algorithm.In this paper,through three aspects,such as single-resolution,multi-resolution,and the integration of other algorithms,the advantages and disadvantages of the improved FCM algorithms are expounded.In the end,some challenges and possible trends are discussed.关键词：fuzzy C-means clustering;image segmentation;objective function;the degree of membership5986|370|0更新时间：2024-05-08
- 摘要：The speed of the phase computation affects the optical three dimensional measurement speed in the Hilbert transform profilometry.Because the pixels which still need to be processed and the already processed pixels are not interdependent,the phase of each pixel can be calculated separately.Therefore,the same program can be executed on multiple threads in parallel with high arithmetic intensity.This allows the phase computation done on the GPU with the help of the powerful unified architecture graphics processor parallel computational capability.This paper analyzes the characteristics of the phase computation,and then implements the code for allowing the CPU with GPU working together to solve the problem of the low efficiency of the phase computation on CPU.Experiments show that by the GPU acceleration the computation speed has been greatly improved with the same quality of the phase computation.关键词：CUDA;GPU;Hilbert transform profilometry;phase computation3617|147|0更新时间：2024-05-08
- 摘要：The traditional iterative image inpainting algorithm which is based on partial differential equations can hardly be used in practice,for its complicated computation and unsatisfied time consumption.The fast marching method(FMM) which is based level set can inpaint the damaged area fast and effective,but it can not prevent the edge well.To solve this problem,an improved algorithm is presented.Continuity strength is introduced when weight function is designed and the factor which contains isophote is adopted to estimate the position between each pixels. When inpainting a single pixel,confidence factor is introduced to weight the interpolating point.Experiments show that the proposed algorithms can ensure the computation efficiency,and at the same time,it well improves the inpainting quality.关键词：image inpainting;fast marching method;continuous strength;pixel position estimatation;confidence factor3247|148|0更新时间：2024-05-08
- 摘要：Non-local Means Filtering (NLMF) has been a popular issue in the image filtering field.The existing NLMFs based pre-selections are analyzed,and it is pointed out that they all have deficiencies in terms of feature extraction from image patches.An adaptive and effeicient NLMF method is proposed using singular value decomposition (SVD) in the gradient domain.Our contributions to NLMF based pre-selection are:1)the robust pre-selection method based structure feature from image patch;2)the relation between size of the similar sets and filtering performance is analyzed;3)automatic selection of similar patches;4)local adaptive selection of the filtering parameter.In addition,the symmetry of the Euclidean distance is considered to accelerate the proposed method further.The experimental results show that the proposed method outperforms the original NLMF and other fast NLMFs on subjective and objective aspects,and has rapid running speed.The proposed method is an efficient filtering method.关键词：nonlocal means filtering;singular value decomposition;image feature;preselection;filtering parameter3586|162|0更新时间：2024-05-08
- 摘要：A PMD (photonic mixer device) camera is a three-dimensional imaging system based on the TOF (time-of-flight) technology.While obtaining the two-dimensional gray image,this camera can capture a range image and an amplitude image at the same time.However,the main drawbacks of the camera are the low resolution and random noise.According to this problem,we combine the amplitude information of the PMD camera and a bilateral filter.An improved discontinuity adaptive Markov random field (DAMRF) model is introduced by a combination of super-resolution reconstruction methods,which introduces the square of the modulating signal amplitude as confidence.Then this method will use it as the weight to carry on an adaptive weight for the distance items of the energy function of the traditional DAMRF model,so it increases the weights of the range image pixel in the process of smoothing.This method not only increases the spatial resolution of the range images,but also effectively filters and de-noises the range image.At the same time,it enhances the marginal information of the range image,and better maintains the continuity of the image edges.The experimental results demonstrate that the reconstruction result of this method is superior to that of the traditional DAMRF model.The method obtains a better improvement in the reconstructed image signal to noise ratio (SNR) and root mean square error (RMSE),and improves the visual effects of the reconstructed image.关键词：super-resolution reconstruction;PMD camera;TOF technology;discontinuity adaptive Markov random field model;confidence;bilateral filter3933|117|0更新时间：2024-05-08
- 摘要：In Micro-CT systems,further improvement of the spatial resolution of the reconstructed images is limited by the X-ray dose level,the pixel pitch,and the aperture of the detector element.In this paper,we study a total variation (TV) based optimization model for Micro-CT reconstruction based on an up-sampling of the reconstruction grid with original detector and X-ray dose.Using an extension of the gradient projection method,an alternating minimization algorithm is employed to solve the corresponding energy function.In the process of the minimization,the treatment is separated into the gradient step of the fit-to-data term,the total variation (TV) denoising,and the specific linear combination of the previous two points.Experiments on simulated data as well as real Micro-CT data are performed.Our results show that the proposed approach can dramatically improve the spatial resolution of the reconstructed images compared to the conventional Filter Back-Projection algorithm.关键词：super-resolution reconstruction;compressed sensing;micro-CT;total variation regularization.3898|155|0更新时间：2024-05-08
- 摘要：The theory of dual quaternion is introduced into the field of space resection.A new algorithm which uses dual quaternions to describe the position and attitude of space resection is presented.First,dual quaternions are used to describe the rotation and translation between different coordinate systems,then the strict collinear equation is linearized and calculated using an iterative method according to indirect adjustment with constraints.Our result show that the stability and reliability of the method are independent from the images initial position and orientation.Furthermore,the algorithm uses less time for the solution and can be applied to different inclinations,different heights,different sensing platforms,with a higher precision in location.关键词：dual quaternion;space resection;exterior orientation parameters3204|89|0更新时间：2024-05-08
- 摘要：In order to reduce the complexity of motion estimation and improve the efficiency of the video encoder,an adaptive motion estimation search algorithm based on motion intensity is proposed.Motion intensity reflects the intensity of motion between adjacent frames.By defining the concept of motion intensity,the movement of the next frame is predicted based on the measured intensity information of the current frame,and a different motion search algorithm is selected adaptively.If the motion intensity of the current frame is greater than the threshold,the UMHexagonS search algorithm is chosen,otherwise the improved hexagon search algorithm is chosen.Simulation results show that the coding efficiency is improved significantly,while the image quality and compression performance change little.In addition,the threshold of motion intensity in the algorithm is adjustable,thus the demands of different encoding can be met by changing the threshold.关键词：motion estimation;motion intensity;adaptive;video encoder3087|104|0更新时间：2024-05-08
- 摘要：A divergence thresholding method in kernel space is proposed.First,a new kind of Bregman divergence in parametric form is defined.Second,the new thresholding method based on Bregman divergence is presented.The new methed can unify cross entropy and Otsu’s thresholding method.Third,a new asymmetric kernel function in reproduced Hilbert space is constructed by means of Bregman divergence in parametric form.Image gray levels in Euclid space are transformed into the reproduced kernel space,and a divergence thresholding method in kernel space is obtained.Finally,the method for choosing parameters for the kernel function in the new thresholding method is analyzed.Experimental results show that the proposed divergence thresholding method in kernel space has a certain widespread adaptability.It can improve the segmentation performance of cross entropy and Otsu’s thresholding methods,and two kinds of classical thresholding methods based on cross entropy and Otsu are also regarded as the special cases of the kernel space divergence thresholding method proposed in this paper.关键词：thresholding method;Otsu’s thresholding method;cross-entropy thresholding method;divergence;Kernel function3163|116|0更新时间：2024-05-08
- 摘要：Considering the performance of traditional spectral clustering using Gaussian kernels,a new spectral clustering based on neighboring adaptive local scale is presented in this paper.Based on clustering consistency characteristics,the proposed method first emphasizes the flexibility of the local scale,which means each sample has a corresponding scale parameter.Furthermore,it overcomes the limitations of traditional methods in all samples with the same global scale parameter.Hence,it can depict the intrinsic structure of data sets better.Second,it stresses the convenience of parameter selection.It can determine the value of a local scale for one sample by computing the sum of weighted distances of neighbors.Therefore,it can determine the scale parameter automatically.This paper illustrates the proposed algorithm not only has inhibition for certain outliers but is able to cluster the data sets with different scales.Finally,experiments on both,artificial data and UCI data sets,show that the proposed method is effective.关键词：local scale;spectral clustering;neighboring adaptive;global scale2994|68|0更新时间：2024-05-08
- 摘要：In this paper,we propose an adaptive spectral clustering algorithm based on the Nyström method with multi-level structures in LUV color space.First,we introduce the LUV color space,which can effectively avoid the influence of barely noticeable differences on the segmentation results,achieving better result in texture and edge regions.Second,we combine the spectral clustering algorithm based on multi-level structure and the Nyström method.Our approach can reduce the operation time and solve the problem of memory overflow.Finally,in -means,through the analysis of the eigengap to adaptive select the value of ,this approach can automatically determine the number of clusters.The proposed method is applied to image segmentation,respectively,in LUV color space and RGB color space.The experimental results show that in LUV color space we can obtain even better results.The data computation and operation time as well as the segmentation result of the proposed algorithm are superior,compared to the spectral clustering algorithm based on the Nyström method (SC-N).关键词：LUV color space;Nyström method with multi-level structure;adaptive;-means;spectral clustering;color image segmentation3286|137|0更新时间：2024-05-08
- 摘要：An important challenge in the analysis of intravascular ultrasound images (IVUS) is the media-adventitia border detection. However, as a result of the inevitable artifact, plaque and imaging equipment effect, the target border always appears too fuzzy to be detected. In this paper, a new border detection method based on spatial-frequency domain image enhancement is proposed. The method uses enhancing processes in the spatial and xin the frequency domain during the detecting process. In the enhancing process, the directional filter band, neighborhood and histogram equalization are combined to overcome the defect of contrast reduction caused by directional filter and the defect of details vagueness caused by histogram equalization. Then in the detecting process, a heuristic graph-searching is applied to find the media-adventitia border by taking the enhanced image data matrix as the cost matrix. The experiment results show that the enhancing process not only strengthens the features of the media and adventitia greatly, but also improves the contrast and definition of the image. The graph-searching based on the enhanced results can detect the media-adventitia border accurately.The correct rate reached 92.76%.关键词：intravascular ultrasound;media-adventitia border;spatial-frequency domain enhancement;heuristic graph-searching3211|130|0更新时间：2024-05-08
- 摘要：A key step of constructing an active appearance model is acquiring a set of appropriate training shapes with well-defined correspondences. In this paper, we introduce a new point correspondence method (FB-CPD), which can improve the accuracy of the coherent point drift (CPD) by using the image feature information. The objective function of the proposed method is defined by both of the geometric spatial information and the image feature information. The original Gaussian mixture model in the CPD is modified according to the image feature of the points. FB-CPD is tested on the three-dimensioral prostate and liver point sets through the simulation experiments. The registration error can be reduced efficiently by FB-CPD. Moreover, the active appearance model constructed by FB-CPD can obtain fine segmentation, in three-dimensioral CT prostate images. Compared with the original CPD, the overlap ratio of voxels was improved from 88.7% to 90.2% by FB-CPD.关键词：active appearance model;point set registration;Gaussian mixture model;feature information3871|137|0更新时间：2024-05-08
- 摘要：Road detection of 2-dimension images is a key problem in vision navigation of mobile robots.Based on the energy minimization function of image segmentation,a new vision energy minimization model is derived which can be solved conveniently by the iterative optimization of swarm intelligence.And a multi-colony ant based dynamic cooperation optimization strategy is proposed to implement the optimized road detection.According to the divide-and-conquer principle,each colony optimizes a sub-problem independently.Then,a set of information exchange strategies are proposed for adaptive dynamic cooperation between neighboring colonies to realize the global optimization of road detection.Compared with the Graph cut based road detection method,the proposed dynamic multi-colony ant cooperative optimization method not only has better detection performance,but also can detect arbitrary number of clusters.It can be applied to detect complex road scenes that obtain multiple types of road.关键词：road detection;image segmentation;multi-colony ant algorithm;energy minimization3126|154|0更新时间：2024-05-08
- 摘要：Sports video analysis has received much attention in recent years and is a challenging research direction in the field of computer vision.A novel prediction method for the behavior of ball carriers in basketball matches is proposed in this paper.Aiming at the cluttered background,fast motion of the sportsmen,and the low resolution of the head images in basketball match videos,we propose the adoption of a covariance descriptor to fuse multiple visual features of the head region,which can be represented as Riemannian Manifolds.Then we map the covariance descriptor to the tangent space and complete the head pose classification through the trained multiclass Logitboosts directly in this space for determining the range of vision of the ball carrier.According to the distribution of all the sportsmen within the range of vision of the ball carrier,we predict the behavior of him—shooting,passing, and dribbling,through sportsmen information based on artificial potential field (APF).Finally,the tests on the basketball match videos verify the effectiveness of our method.关键词：covariance descriptor;tangent space;multiclass LogitBoost;artificial potential field6663|139|0更新时间：2024-05-08
- 摘要：An approach of 3D craniofacial geometry shape similarity estimation is proposed to solve the problem that current approaches mainly rely on subjective assessment for the lack of related quantities. In our approach the cranioface, which is to be compared, is transformed into another one in the database and its principal warps are computed to be the base for the representation of the shape change. First, the global landmarks are selected from the two craniofaces and the thin-plate spline function is used to establish a map between them. Then, the corresponding bending transformation matrix is computed. This matrix can be represented by the product of the principal warps and one coefficient matrix, which is used to measure the deformation degree of the craniofaces. Finally, the geometry similarity distance is defined on the basis of the coefficient matrix. Experimental results demonstrate the feasibility and effectiveness of the approach.关键词：craniofacial similarity;craniofacial reconstruction;thin-plate spline;landmark;craniofacial distance3423|104|0更新时间：2024-05-08
- 摘要：Interpolation is one computational complex and time-consuming operation in the fields of spatial analysis that can not meet the real time demand. With the rapid increase of GPU floating-point computing power, general-purpose computation on graphics processors (GPGPU) has became an evolving research field in spatial information processing, and it provides an opportunity to accelerate some traditional inefficient algorithms. In this paper, we map the inverse distance weighted (IDW) interpolation method to the compute unified device architecture (CUDA) parallel programming model. Taking the advantage of graphics processing unit (GPU) parallel computing, we build two-level indexes on GPU, then blocking schemes are used to assign computing task among different threads. After illustrating the parallel interpolation process, we conduct several experiments, The experiment result shows that the error of this new method can control under 10 compared with CPU-based method. With larger influence radius and massive data, the performance can obtain above 40 times speedups over a very similar single-threaded CPU implementation. It is demonstrated the correctness and high efficiency of our optimized implementation.关键词：geographic information system;parallel interpolation;graphics processing unit;compute unified device architecture3844|124|0更新时间：2024-05-08
- 摘要：To make the real-time rendering of large scale terrain more efficient,a batched LOD algorithm is proposed in which terrain blocks are handled as the processing unit.In the preprocessing stage,multi-resolution terrain data is partitioned into regular blocks with predefined sizes.These blocks are organized in the form of quadtree.Based on this form,an error metrics for LOD selection is designed in order to simplify the computation of level decision.Then by the means of adding skirts and geometry morphing,different levels are transited without a visual distortion or popping.When terrain is being rendered,frustum culling is used to reduce data transmitted to graphics hardware.To efficiently implement data loading and datasets management,terrain quadtree lists and a prediction strategy are introduced.From the result of final simulation,it can be seen that the algorithm can fully harness the power of current graphics hardware and reach higher rendering rates.关键词：terrain rendering;error metrics for LOD;quadtree;multi-resolution;view frustum culling3723|150|0更新时间：2024-05-08
- 摘要：A new method for ship recognition using optical remote sensing data based on rough set and hierarchical discriminant regression (HDR) is presented in this paper. First, a new shape feature called area ratio code (ARC) is proposed and extracted as a candidate feature. Based on the rough set theory, the common discernibility degree is used to compute the significance weight of each candidate feature and select valid recognition features automatically. Ultimately, a decision tree based on the HDR theory is built to recognize ships in data from optical remote sensing systems. Experimental results on real data show that the proposed method is generalizable and can get better classification rates at a higher speed than the KNN or SVM method.关键词：ship recognition;area ratio code;rough set;hierarchical discriminant regression;remote sensing3561|244|0更新时间：2024-05-08
- 摘要：In this paper,a de-speckling algorithm for SAR images using an adaptive over-complete learning dictionary is proposed.The algorithm is based on sparse representation of SAR images via an over-complete dictionary It has strong data sparseness and provides solid modeling assumptions for data sets.First,a practical optimization strategy based on statistical properties of the speckle noise is used to design a redundant dictionary via an iterative loop.Second,the SAR image is projected into a high dimensional space using the learning dictionary and a sparse representation of the SAR image is obtained.Third,a model for multi-objective optimization problem is built by a regulation method.Finally,the de-noising process is realized through a solution of the multi-objective optimization problem in which the mean backscatter power is reconstructed.The experimental results demonstrate that the proposed algorithm has good de-speckling capability while preserving image details.关键词：speckle noise;SAR image;dictionary learning;sparse representation;orthogonal matching pursuit algorithm3912|146|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