最新刊期
卷 16 ， 期 12 ， 2011
- 摘要：As an interactive geometry editing technique,mesh deformation plays an important role in geometry modeling and computer animation.Detail-preserving mesh deformation,especially differential domain deformation techniques gains high interests in the computer graphics community in recent years.Combined with recent advancements on mesh deformation methods classified into skeletal deformation,surface deformation,and space deformation,this paper provides a systematic survey on this area,trying to give a comprehensive introduction and comparison to those typical mesh deformation methods from their algorithmic principles,advantages and limitations.Furthermore,the opportunities of future research in mesh deformation are discussed.关键词：mesh deformation;space deformation;differential domain deformation;survey5399|272|0更新时间：2024-05-07
- 摘要：During the imaging process of remote sensing images,it will cause the image noise,including Gaussian noise and salt-and-pepper noise,which reduce the definition of the images and impact their practical application to a large extent.The goal of noise elimination of remote sensing images is to properly protect texture information of the image while efficiently eliminating noise.Based on the disadvantage of the Lin Shi algorithm and the anisotropic diffusion equation of nonlinear wavelet threshold value,the thesis puts forward the anisotropic diffused noise-eliminating model of average gradient of identical direction.This model effectively eliminates noise,at the same time,also properly protects edge features and texture information of remote sensing images,without the disadvantage the Gaussian noise and the salt-and-pepper noise cannot be eliminated by the Lin Shi algorithm and anisotropic diffusion equation of nonlinear wavelet threshold value.Our experiment show the stability and feasibility of our model,and the PSNR of image after using this model respectively increases 2～4 dB compared to the Lin&Shi algorithm or the anisotropic diffusion equation of nonlinear wavelet threshold value.关键词：identical-directional gradient;Lin&Shi algorithm;anisotropic diffusion equation;remote sensing image3796|248|0更新时间：2024-05-07
- 摘要：In this paper,a novel approach is proposed for removing and restoring random-valued impulse noise and preserving fine details at the same time.The impulse noise detection technique is presented,which is based on the so-called peer group concept.A pixel is noise-free only when its peer neighbor number is local maximum or large.Extensive simulations show that the proposed filter provides better performance than many of the existing filters and the computational complexity is lower than many of the median-based filters.In particular,the threshold is adaptive to different image types.关键词：image denoising;random-valued impulse noise;noise ratio;noise detection3016|227|0更新时间：2024-05-07
- 摘要：Conventional improved interpolation introduces blurs,jagged edges,or artifacts.The Joint Bilateral Interpolation (JBI) is proposed in this paper.The low-resolution (LR) input image is first being interpolated with a fast and simple algorithm and then the interpolated result is used as a guidance image to apply the joint bilateral filter to upsample the LR image.And the parameters of Gaussian kernel function are given in the paper.Experiment results demonstrate that our new interpolation algorithm substantially improves the quality of the interpolated images and costs less calculation time than most existing nonlinear improved interpolation algorithms.关键词：image interpolation;bilateral filter;joint bilateral upsampling;edge preserving3813|213|0更新时间：2024-05-07
- 摘要：A new JPEG image enhancement method based on Retinex theory is proposed to improve the existing enhancement methods. The exiting methods cannot enhance images effectively because after JPEG compression images will lose some information. The proposed algorithm enhances images before quantization in the JPEG compression process to use full primal information of images. The discrete cosine transform (DCT) coefficients are divided into the incident component (DC coefficients) and the reflection component (AC coefficients) based on the Retinex theory; the DC coefficients are adjusted in the image dynamic range, and the AC coefficients are adjusted to enhance the details with the enhancement factors of the DC coefficients. The block artifacts will be suppressed by smoothing the DCT coefficients. Compared to other enhancement algorithms, results show the new approach’s efficiency in details enhancement and color preserving, while keeping the same compression ratio as the normal JPEG compression.关键词：JPEG compression;image enhancement;Retinex;discrete cosine transform3140|339|0更新时间：2024-05-07
- 摘要：Structural similarity (SSIM) is an image quality assessment algorithm with the advantage of simplicity, high efficiency and better consistency with human subjectivity. However, it often fails when measuring badly distorted or cross distortion images. In this paper, an improved algorithm called structural similarity highlighting edge regions (HESSIM) is proposed based on the idea that edges are the most important information in an image.The humans eye is very sensible for distorted edge information. In the proposed HESSIM, the edge regions are first divided from an image by Otsu’s method, then those with obviously perceptual distortion are chosen by the JND model, and their distortion measures are highlighted. Experimental results show that HESSIM is more consistent with HVS than SSIM,especially for distorted images which are blurred or comtaminated with white noise.关键词：HESSIM (structural similarity highlighting edge regions);edge regions;Otsu;JND(just-noticed distortion)3735|245|0更新时间：2024-05-07
- 摘要：Steering kernel regression is an adaptive and effective image restoration algorithm,which has been widely used for image denoising,enlargement and deblurring.However this model is based on the Gaussian kernel function,which often blurs the image edges,especially micro-edges in its restoration.A new anisotropic image restoration model based on robust statistics is proposed,which improved the Steering kernel regression model.The new method is to incorporate anisotropic distance and introduce robust estimation kernel function instead of Gaussian function.Extensive experiment results demonstrate that the new method can yield superior performance to that of the steering kernel regression,especially in preserving the details of the image edges.关键词：image restoration;kernel regression;edge preserving3875|249|0更新时间：2024-05-07
- 摘要：During the vector morphology processing of color image, the criterion of vector ordering is vague and one-sided. A new method of vector ordering is proposed in this paper. The gap between the visual colors is proportional to the Euclidean distance in the lab uniform color space coordinate system, and this characteristic is used to calculate the color difference, which is used as the criterion of vector ordering. The color image quaternion model and morphology structure element quaternion model are established to delimit the new morphology operators. At last, this method is used in some applications and compared with the HSV method. The experimental results show that it is better to use the proposed method to extract the image edges and reduce the salt & pepper noise.关键词：uniform color space;quaternion;color image morphology;edge detection;salt & pepper noise3118|221|0更新时间：2024-05-07
- 摘要：Multidimensional scaling (MDS) has been applied in many applications such as dimensionality reduction and data mining. However, one of the drawbacks of MDS is that it is only defined on "training" data with no clear extension to out-of-sample points. Furthermore, since MDS is based on Euclidean distance, it is not suitable for detecting the nonlinear manifold structure embedded in the similarities between data points. In this paper, we extend MDS to the correlation measure space (CMDS). In contrast with MDS where the mapping between the input and reduced spaces is implicit, CMDS employs an explicit nonlinear mapping between the both. As a result, CMDS can directly provide predictions for new samples.Correlation is a similarity measure, so the CMDS method can effectively capture the nonlinear manifold structure of data embedded in the similarities between the data points. Theoretical analysis also shows that CMDS has some properties similar to kernel methods and can be extended to the feature space. The effectiveness of our approach is demonstrated by extensive experiments on various data sets, in comparison with existing dimensionality reduction algorithms.关键词：unsupervised subspace learning;correlation measure;multidimensional scaling2719|193|0更新时间：2024-05-07
- 摘要：In this paper,a computation technology of variational optical flow based on Image-and Flow-driven is proposed.The idea to obtain the data term is to use the combination of gray and gradient constancy assumption,the combination of local and global constraint,as well as the regularization factor,to improve the accuracy.A joint image-and flow-drive anisotropic smoothing strategy is introduced to design the smooth term,linking the data term and smooth term closely.And then,the speed of the optical flow can be controled by the design of the eigenvalues of the diffusion tensor.Finally,the method utilizes the multi-scale hierarchical refinement strategy to handle large displacements problem.The experiment results prove that the model has a good computation effect of optical flow to cope with the problem of additive illumination changes and the boundaries of the moving objects with complicated backgrounds.关键词：optical flow;variational method;regularization;image-and flow-driven;diffusion3679|225|0更新时间：2024-05-07
- 摘要：By using the local regional information which has the ability to enhance the image, a new SPF function has been defined. The edge stopping function in the GAC model is replaced by the SPF function, and a new region-based active contour model is put forward by improving the GAC model. Therefore, images with intensity inhomogeneities and weak boundaries can be processed. The Selective Binary and Gaussian Filtering Level Set (SBGFRLS) method is continuously used in the new model which is simplified by avoiding the process of reinitializing the level set function. Experiments on real and synthetic images indicate that the new model has the same segmentation results as the LBF model, while the computational efficiencies improve significantly. The new model not only can segment images with intensity inhomogeneities and weak boundaries, but also has the properties such as sub-pixel accuracy, anti-noise nature, selective local or global segmentation, etc.关键词：image segmentation;active contour model;region-based model;SPF function;level set method;intensity inhomogeneity3067|188|0更新时间：2024-05-07
- 摘要：We present an opimized method for multi-retinal image mosaicking.First, the feature-based hierachical, robust estimation method completes a pairwise registration. Then a method is proposed to achieve good registration, which is based on overlaping regions for registration correction. The graph is established by the elationship between the pairwise registrations. The geometric error and similarity are selected as the weight to find the shortest path and to decide the anchor image. The joint estimation is constructed by direct constraints from the anchor image and indirect constraints from non-anchor images. For image fusion, multi-band blending is proposed, which is based on distance transform. The stitching seam is removed and a brightness blending image is achieved. Experimental results show that the method can effectively eliminate the error accumulation, achieving a precise registration and a seamless mosaic for multi-retinal image independent of the input order.关键词：image mosaic;retinal image;shape context;anchor image;global registration model;distance transform;multi-band blending3925|257|0更新时间：2024-05-07
- 摘要：Road extraction is one of important issues for target recognition in remote sensing image. At present, most road extraction methods are based on shape-features.But the threshold selection is very difficult and needs much human intervention,because the shape-features have the complicated information in remote sensing images. Hence,a road-extraction method using the DS (Dempster-Shafer)evidence theory to fuse the shape features is proposed. First, some shape features are selected and optimized. Then, the basic probability assignment functions are designed and the road segments are extracted by using shape features with the DS theory. Finally, the road center lines are obtained by connecting road segments.Typical road images and non-typical road images were selected for experiments, and the results prove that the method can reduce the difficulty of threshold selection and the dependency of human intervention.The method is effective for the typical road images and non-typical road images.关键词：road extraction;Dempster-Shafer evidence;threshold selection;shape features;high resolution remote sensing image3086|254|0更新时间：2024-05-07
- 摘要：A practical way of representing circular arcs with standard-form cubic rational Bézier curves containing a parameter is presented. The parameter affecting the control points, weights and shoulder points is discussed. The inner relation between the parameter and the weights of representing circular arcs with nonstandard-form quadratic rational Bézier curves is analyzed in detail. Modifing the parameter value corresponds to a rational linear parameter transformation. Finally, we discuss the inverse calculation of the standard-form cubic rational Bézier representation of circular arcs.关键词：Rational Bézier curves;representation of circular arcs;shoulder point;weights;rational parameter transformation3407|247|0更新时间：2024-05-07
- 摘要：Because of the stronger noise, the stable performance of optical method degrades obviously in SAR image scale invariant feature extraction. To deal with it, we propose a new SAR image scale invariant feature extraction algorithm based on Harris operator. As well known, Harris operator has well stable performance in single scale image feature point detection. In this paper, we first adapt it to multi-scale image and select the features positions by analyzing the stability of Harris measure local maximum points. Then to improve the speed,the recursive Gaussian filter is designed to replace the convolution filter. At last, the feature descriptor is designed based on the gradient information of feature point neighborhood. Based on real SAR images, the experimental results show that an obviously stable performance improvement can be obtained through our algorithm.关键词：Harris operator;scale invariant feature;recursive filter;SAR image processing3246|246|0更新时间：2024-05-07
- 摘要：This paper improves the traditional LMBP algorithm of remote sensing image classification by extending input vectors which have been changed from low dimension to high dimension with kernel function for full utilization of the first and second derivatives information of error function. Simultaneously, combined with the traditional advantages of the LMBP algorithm, it can accelerate the convergence of network training. The simulation results show that the improved method can be more effective in remote sensing image classification because it needs less iteration for network training and achieves higher classification accuracy.关键词：Kernel function;LMBP algorithm;remote sensing image classification3174|220|0更新时间：2024-05-07
- 摘要：In this paper we aim at symbol sharing. First, we analyze two methods of symbol sharing.Second, we investigate the construction principals of map symbols for the difference issues of symbol design patterns.The PostScript imaging model will be introduced and an XML-based symbol representation mode is being proposed as well. Third, due to the discrepancy of symbol usage environment and symbolic interface, a web services based distributed sharing method for map symbol data and functions is presented. Finally, a prototype system for symbol sharing is being built to validate the feasibility of map symbol service sharing.关键词：map symbol sharing;XML;Web Services3361|216|0更新时间：2024-05-07
- 摘要：Representative algorithms for extracting gradient features include HOG (histogram of oriented gradients), Shapelet and Edgelet. They have been widely used in the fields of object detection, object recognition, image retrieval,and scene classification. A new feature named Multi-Directional Integration of Gradients (MDIG) is proposed after analyzing the HOG and its fast version. By avoiding computing the gradient orientation and using an integral image, the new feature is easier to extract and can be hardware implemented on a DSP.The MDIG is tested on images of human head-shoulders. Using AdaBoost to train a classifier, the result is comparable with HOG, while the computation burden is only one third of that of the HOG. Finally, the use of gradient feature is discussed, and the future work is summarized.关键词：gradient feature;HOG;multi-directional integration;head-shoulder detection3610|245|0更新时间：2024-05-07
- 摘要：Variational models for vectorial image denoising involve couplings of different channels to preserve edges, which lead to problems of complexity and efficiency. Meanwhile, different types of couplings result in different edge preserving effects. The objective of our work is to design fast Split-Bregman algorithms for a couple of variational models which have been proposed in recent years and compare their edge preserving properties and their efficiency. Five variational models for vectorial image denoising using different regularizers are studied: the LTV (layered total variation) regularizer, the MTV (multichannel total variation) regularizer, the CTV (color total variation) regularizer, the PA (polyakov action) regularizer, and the RPA (reduced polyakov action) regularizer. The order of their edge preserving quality and computation efficiency are given based on numerical experiments. It is shown that the CTV model is the best method for edge preserving followed by the PA model, the MTV model, the RPA model, and the LTV model. The LTV model is the fastest model, followed by the MTV model, the RPA model, the CTV model, and the PA model.关键词：vectorial images;variational models;edge preserving;image denoising;Split-Bregman algorithm3516|238|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