最新刊期
卷 13 ， 期 5 ， 2008
- 摘要：This is the thirteenth in the survey series of the yearly bibliographies on image engineering in China. The main purpose of this survey work is to report the up to date development of image engineering in China, to provide a convenient means of literature searching facility for readers working in related areas, and to supply a useful reference for editors of journals and potential authors of papers. Considering the wide distribution of related publications in China, 895 references on image engineering research and technique are selected from 3 312 research papers published on a set of 15 Chinese journals around 118 issues. These 15 journals are important ones in which papers concerning image engineering have higher quality. Those selected references are classified into 5 categories （image processing, image analysis, image understanding, technique application and survey）, and then into 23 specialized classes according to their main contents （similar like last year）. Some analysis and discussions about the statistics made on the results of classifications by journal and by category are also presented. This work shows a general and off the shelf picture of the various progresses of image engineering in China in 2007. It should be pointed out particularly that, in 2007 the number of research papers for image engineering, published in the above 15 journals has greatly increased and staged in a new era in the history. A tendency of continuous development of image engineering in China shows great promise.关键词：image engineering;image processing;image analysis;image understanding;technique application;survey; statistics;classification3124|293|0更新时间：2024-05-08
- 摘要：An improved JPEG2000 algorithm is proposed to meet the realtime requirements of video supervisory system under the limited bandwidth. Firstly, a novel symmetric biorthogonal 97 wavelet is constructed based on the Cohen Daubechies Feauveau Theorem, vanishing moments condition and the lifting scheme. Then the constructed 97 wavelet is introduced to substitute CDF97 wavelet of JPEG2000 algorithm in order to improve JPEG2000 performance. Compared with traditional CDF97, the new 97 wavelet has a faster processing performance for its remarkably less computational complexity of image compression, while the compression performance is almost the same as that of CDF97 wavelet. At last, the experimental results of applying the proposed algorithm in a video surveillance system validate the predominance.关键词：97 wavelet;lifting scheme;JPEG2000;web robot;image compression2951|210|0更新时间：2024-05-08
- 摘要：Noise removal from images is an important part of images pre treatment,and its result has a great effect on the final result of images processing. How to take full advantage of local characteristics and details in images to improve noise removal,detail preserving and edge preserving ability of filters? A nonlinear filtering algorithm using probability statistic and main texture direction analysis is proposed.This algorithm utilizes Radon transform to determine texture direction probability density distributions of local areas of images and then applies probability statistic model to estimate the middle pixel’s gray value according to its neighbour pixels. The performance of the proposed method is evaluated with several sets of images contaminated by pulse noise and Gaussian noise. The experimental results show the superiority of this method including the ability of de noising and preserving edges and details of images especially for images with pulse noise and Gaussian noise. This algorithm makes richly use of local characteristics and details of images. Especially unlike some recent algorithms applying only for salt and pepper impulse or Gaussian noise removal,our method,is applicable to images contaminated by any kind of noise.关键词：nonlinear filtering;probability statistic;RADON transform;texture analysis;main direction2782|255|0更新时间：2024-05-08
- 摘要：In this paper, a new algorithm based on a statistical model of the coefficients of the Dual tree Complex Wavelet Transform (DT CWT) is proposed for image denoising. The DT CWT is approximately shift invariance and has good directionality, which are properties suitable for image de noising. Neighborhoods of coefficients at adjacent positions and scales are modeled as Gaussian Scale Mixture. Under this model, subband coefficients are estimated by Bayesian Least Square estimator. Experimental results show an improved de noising performance of PSNR and human vision in comparison with other methods.关键词：image de noising;dual tree complex wavelet transform;Gaussian scale mixture;Bayesian least squares estimate3088|195|0更新时间：2024-05-08
- 摘要：This paper proposes a bran new discrete transform called AP Contourlet (All Phase Contourlet), the multi scale decomposition and reconstruction of which are based on the APDCT (All Phase DCT) sub band filtering and the APIDCT (All Phase IDCT) interpolation for optimizing denoising effection of Contourlet. In addition, a kind of all phase directional filter bank (APDFB) based on DCT is used for directional filtering of the AP Contourlet. The APDFB is possessed of excellent directional selectivity, and reconstruction algorithm is so simple that only adding the directional images is needed. Experiments in image de noising have shown that the performance of the proposed AP Contourlet is obviously superior to the conventional Contourlet both in vision and in signal to noise ratio (SNR).关键词：all phase Contourlet;directional filtering;subband decomposition;interpolation2552|221|0更新时间：2024-05-08
- 摘要：The high radiation dosage of computed tomography limits its further applications to mass screening. Clinically,lowdose protocol has been used in data acquisition for this situation. This will increase the image noise and degrade the image quality,and thus result in difficulties in diagnosis. To improve the image quality of low dose CT,a statistically based wavelet denoising method in sinogram domain is proposed. The noise properties of low dose projection data were first analyzed and modeled. It could be regarded as approximately Gaussian distributed with a nonlinear signal dependent variance. Then the property of non stationary noise in the stationary wavelet domain was analyzed,and the wavelet coefficients were reconstructed with the adaptive filtering based on minimum mean squared error combined with Bayesian estimation for an optimal noise treatment. After proposed sinogram filtering,the image was reconstructed using the conventional filtered backprojection (FBP) method. Experimental results have shown that the algorithm is effective in removing noise while maintaining the diagnostic image details.关键词：computed tomography;sinogram domain;noise reduction;wavelet transform;Bayesian estimation3263|246|0更新时间：2024-05-08
- 摘要：The function of de interlace is to convert interlaced images to progressive ones. In this paper, a motion adaptive de interlacing algorithm based on edge direction is presented. It consists of motion estimation, low angle edge detection, and spatio temporal weight adaptive interpolation. It estimates the motion by comparing the sum of absolute block difference (SAD) between same parity fields with the threshold of motion. The motion information obtained by motion estimation is filtered by a mathematical morphology filter. The low angle edge detection uses an adaptive searching radius and parallel search strategy in which the 6° edge can be detected. Experimental results show that the proposed method achieves high image quality with low hardware complexity.关键词：de interlacing;motion estimation;mathematical morphology filter;edge detection;motion adaptive interpolation3406|235|0更新时间：2024-05-08
- 摘要：Scene visibility is very low in foggy days We need to defog the surveillance video to make sure the intelligent visual surveillance system work normally According to optics theory, the reduction of scene visibility is exponential to the scene depth The defog effect of existing model based methods are not good enough In this paper, a new physics based method of image restoration is presented This method models scene points in foggy day firstly Then the scene depth is calculated for two images which are captured in clear day and foggy day respectively Image or video taken in foggy day is restored using depth information finally The experiments demonstrate the effectiveness and practicality of this method关键词：defog;intelligent visual surveillance;image restoration;physics model;scene4245|330|0更新时间：2024-05-08
- 摘要：To develop a security multipurpose image watermarking system，we use the 3D unified chaotic system sequences [WTHX]Y[WTBZ] and [WTHX]Z[WTBZ] to generate embedding coefficients of visible watermarks, and embed the visible watermarks in the image spatial domain to achieve the goal of copyright notification. Then we embed the first invisible watermarks in the image with embedded visible watermarks to obtain the aim of copyright protection. After removing the visible watermarks, the second invisible watermarks are embedded in the image to obtain the aim of unauthorized copy tracing. Two kinds of invisible watermarks embedding positions are selected randomly by unified chaotic system sequences [WTHX]X[WTBZ] and [WTHX]Y[WTBZ], based on a novel dual parameters quantization technique. Both watermarks that can be blindly extracted are embedded in the lifting wavelet transform (LWT) domain of those selected bocks. Simulation results show that the visible watermarks can be removed entirely by authorized users and can hardly be removed by unauthorized users. The impact of the invisible watermarks on the image quality is slight, and the invisible watermarks are robust to common image processing operations.关键词：watermarking model;lifting wavelet;unified chaotic system;visible watermark;invisible watermark3145|213|0更新时间：2024-05-08
- 摘要：Watermarking scheme is a well known method for copyright protection of images. However, we can not identify who the real owner is if multiple watermarkings are embedded in an image or the embedding algorithm of watermarking has made public because some attacks can destroy the watermarking. In this paper, a robust copyright protection scheme for digital images that combines trace transform and timestamp technique is proposed. Firstly, the features of the image are extracted using the trace transform. Then the message is registered at the fair third party in order to gain the timestamp. Finally, we can use the timestamp to prove who the real owner is. Experimental results are given and compared with the results of the method based on the geometric moment invariants. The results show that the scheme can well withstand geometrical attacks such as rotation, scaling, shearing, combined attack, affine transform, remove lines, filter and JPEG lossy compression.关键词：copyright protection;digital watermarking;geometric attacks;Trace transform2525|193|0更新时间：2024-05-08
- 摘要：Fuzzy C means (FCM) clustering is one of well known unsupervised clustering techniques, which has been widely used in automated image segmentation However, when the classical FCM algorithm is used for image segmentation, there are also some problems, such as weak robustness of distance measure,reguire ments of setting the initial number of clusters in advance, without considering local image feature In this paper, an adaptive FCM image segmentation algorithm based on the feature divergence is proposed, which can accomplish image segmentation by importing the feature divergence vector into distance measure, incorporating the cluster validity exponent to ascertain the initial number of clusters automatically and extracting the image feature according to Laws texture measure Experimental results show that the proposed method is simple and work well for most images (especially for texture images), and has better segmentation effect than the existing FCM image segmentation.关键词：image segmentation;fuzzy C means clustering;Laws texture measure;Cluster validity3300|229|0更新时间：2024-05-08
- 摘要：In this paper, an improved fuzzy clustering based segmentation algorithm for noise images is presented. Being sensitive to noise is one of the popular fuzzy C means (FCM) algorithm’〖KG-*3〗s drawbacks. The new objective function used in fuzzy clustering is modified to obtain different membership functions by rewarding crisp membership degrees, that is, using distances to the Voronoi cell instead of using distances to the cluster prototypes. Thus the proposed algorithm enhances robustness to noise. Experimental segmentation results for noise images demonstrate the effectiveness and robustness capability of the proposed algorithm.2838|264|0更新时间：2024-05-08
- 摘要：In casting defect recognition of an X Ray image，due to the complex image background，it is difficult to do segmentation with traditional segmentation technology for interested regions，in which the defects are intense. In this paper，fractal feature of the edges is introduced to distinguish which edge belongs to the casting defects in an edge image and the regions containing mass defect edges，e.g. interest regions are segmented by clustering. Directed to the aluminum alloy casting of vehicle wheel X Ray images，the fractal based region segmentation method of this paper is used for recognition experiments，and the results show that the approach is effective.关键词：image segmentation;region of interest;fractal;casting;X ray image2758|217|0更新时间：2024-05-08
- 摘要：Based on the laws of Gestalt psychology, a geometric active contour model for image segmentation is proposed in this paper. The energy function is mainly made of inter region dissimilar properties based on object background principle when contour curve is far from the object edges. The energy function is mainly made of intra region similar properties based on proximity principle when contour curve is situated near the object edge. The developed model, which incorporates the information of foreground and background regions, produces perceptual characteristics and is a general form of the geometric active contour model. Experiment on Lateral Ventricle and tumor images show that in general the model can deal with automatic segmentation of weak edge image efficiently. Furthermore, the experiment on immune cell shows that cytoplasm of immune cell is segmented out from two background regions of cell nucleolus and body fluid with the method.关键词：laws of gestalt psychology;geometric active contour model;image segmentation3070|188|0更新时间：2024-05-08
- 摘要：A new weighted adaptive algorithm of face recognition based on class matrix and feature fusion was proposed. Firstly, global features and local features of six key parts of faces were extracted respectively. Dynamic method of how to choose the weights of local features was given. Different weights could be gained for different training sets according to this method. So, the adaptive ability of algorithm was enhanced. Then, global and local features were fused with weights to get the eigen matrix of samples. Secondly, a new weighted principal component analysis (PCA) method was designed to lower dimension for sample matrixes. Thirdly, the concept of class matrix was proposed, and formula of how to obtain the class matrix was given and proved. According to class matrix, a new projected rule was given. Finally, class matrix and tested samples were projected respectively through the proposed rules. Then, the final class that tested faces belonged to was declared according to the Euclidean distance. Experiments show that the proposed algorithm can deal with small sample problems in LDA effectively, and the results also indicate that it has good performance on speed and recognition rate.关键词：face recognition;feature extraction;Gabor wavelet;principal component analysis (PCA);linear discriminant analysis(LDA);class matrix2706|194|0更新时间：2024-05-08
- 摘要：For tracking object more robustly and rapidly, a novel approach for object tracking using Monte Carlo method is presented, which is motivated by the method of tracking with particle filter. In this approach, the locations and scales of the candidates in the next frame are sampled through Monte Carlo technique. Then the similitude degree between the samples and the reference objects are calculated. At last, the object states are estimated and the object under tracking is attained. This method does not need the motion information of the object, so it is especially good at tracking flexible object. The method is easier to implement than other present methods. Furthermore, it is proved to be quite robust and versatile by the numerical experiments.关键词：object tracking;Monte Carlo methods;Bayesian filtering;particle filtering2965|236|0更新时间：2024-05-08
- 摘要：A local image registration method is proposed specially for the conditions when images, in medical image registration, are largely similar or the differences are not significant. The transformation function is a radial basis function with compact support and its locality can be conveniently controlled by distributing the feature points into the desired regions, which especially allows us to deal with local changes in medical images. Mutual information is chosen as cost function in order that the transformation function can be achieved accurately. In the process of the optimization, the image registration is treated as optimal problem and niche genetic algorithm is employed to optimize the transformation function parameters because it can overcome the drawbacks of premature and weak exploitation capabilities compared with genetic algorithm. The experiments on the simulated image with the known transformation function and the real image are conducted by using the proposed method. The results show that the optimal transformation function can be found and its action domain is controlled within a relatively small region. The presented method, which is a feasible and robust medical image registration approach, exploits the advantages of both feature points and intensity information and can obtain the accurate transformation function by the efficient optimization strategy.关键词：medical image registration;feature points;mutual information;niche genetic algorithm2804|233|0更新时间：2024-05-08
- 摘要：The images are processed with Q shift DT CWT. The low frequency coefficients and high frequency coefficients are fused with NGMS (Neighborhood Gradient Maximum Selectivity) and SI MVMS (Synthesis Image Module Value Maximum Selectivity) separately because of their different characteristics. For the low frequency coefficients, the ones having maximal neighborhood gradient are selected. For the high frequency coefficients, the ones having maximal absolute values are selected, and verify consistency of the fused coefficients . Experience results show that the images fused by the proposed algorithm are of better quality than that produced through the algorithms with wavelet transform. The proposed algorithm not only solves the problems such as low contrast and blocking effects caused by fusion algorithms in space domain, but also avoids the artifacts and ringing artifacts exhibited by conventional wavelet based fusion algorithms, which benefit from the approximate shift invariant and good directional selectivity of Q shift DT CWT.2846|255|0更新时间：2024-05-08
- 摘要：In this paper, we propose a sketch based synthesis approach to create 3D animation of a target character. Our approach consists of the following steps: (a) a sketch based mapping between the sources and the target; (b) deforming the target based on the deformations of the corresponding sources by means of differential mean value coordinates; (c) stitching the deformed portions of the target together smoothly with constrained minimization method to form a key frame of the target animation; and (d) interpolating between key frames created from (c) to produce a complete 3D animation of the target. We provide new research contributions on all these topics, and integrate them into our newly developed prototype animating system. Our approach does not require the sources and target to share the same number of vertices or triangles, or to have matching connectivity. In our system, source animations can be 3D animations as well as 2D videos or macromedia flashes; and the target can even be unstructured or point cloud. Our approach is intuitive and is able to produce highly authentic 3D animations. We demonstrate our approach by constructing a full cart animation from dog, cat, snake and human animations as well as applying a 2D video of the wheel of a car onto the wheels of the cart. Other examples produced from our prototype system are also given in the paper to illustrate the generality and robust of our approach.关键词：sketch;deformations;Animation;differential mean value coordinates;interactive animating tools2684|208|0更新时间：2024-05-08
- 摘要：Irradiance environmental maps are an effective technique to render diffuse objects in arbitrary lighting environments. Based on current computer graphics hardware a technique is proposed for calculating irradiance environmental maps using vertex shader. The approach starts from an analysis of spherical harmonics and finds out a quadratic polynomial form of environment mapping. Using vertex shader instructions the polynomial coefficients and spherical harmonics functions can be fast calculated, and then the corresponding irradiance environment maps are also generated. When the lighting environment changes, the lighting coefficients are fast computed by generating a mipmap of the source environment map, so that real time recalculating irradiance environment is obtained. The experimental result shows it not only renders realistic lighting but also satisfies the interactive requirements in dynamic lighting environments.关键词：irradiance;environmental maps;spherical harmonics;vertex shader;mipmap2884|206|0更新时间：2024-05-08
- 摘要：This paper presents a novel bionic model based on olfactory systems to generate texture image. The model simulates one of the olfactory neural networks. The chaotic characters of Logistic function are used to adjust the parameters of model during iteration. A simple periodic function is used as the activation function of node in the model to generate periodic texture. And a random noise is introduced to simulate the background noise of brain when processing information. The experimental results show that the model can generate plentiful and multivariant texture images. The introduced random noise plays an important role and enriches the variety of texture images obviously. In addition, the model efficiency to generate texture image outperforms the conventional back propagation neural network model.关键词：olfactory neural networks;bionic model;texture image;random noise3013|190|0更新时间：2024-05-08
- 摘要：This paper presents a system for real time modeling and rendering short fur. In the system, multi layer textured slices are employed to represent furry surfaces, and the high rendering performance is achieved. The system utilizes a vector field as well as a scalar field to control the shape variations of furry surfaces, and provides several fur modeling tools, including the comb, the blower, the attractive stick, and the tools for interpolating, smoothing and disturbing fur, by which users can model fur shape conveniently. In addition, a fur texture creator is provided for designing various fur textures that can affect the final rendering results of furry surfaces. The experimental results show that the system has good modeling capability and is easy to use.关键词：fur modeling;fur rendering;multi layer textured slices;surface fields3329|292|0更新时间：2024-05-08
- 摘要：The optimal layout problem of circle group in a circular container with performance constraints of equilibrium belong to NP hard problem. A fast local serach algorithm (LS) is presented for solving this problem. Firstly, the method constructs an equivalent physical model and defines the energy function of the system. Then the energy function is optimized by the steepest descent method and the approximation solution is obtained indirectly. In order to improve the computational efficiency, the accelerating strategy is added to the LS algorithm. Two examples are computed numerically, and the results of the experiment show the feasibility and efficiency of the approach.关键词：constrained layout problem;NP hard;configuration;local search algorithm;accelerating strategy2722|189|0更新时间：2024-05-08
- 摘要：Ray space representation is an effective approach to realizing real time free viewpoint video system with complicated scenes. A feature point based ray space interpolation method is proposed for arbitrary viewpoint image rendering in this paper. At first, feature points of ray space slice or image are extracted, then, considering occluded problem, the feature points are matched based on correlation based criterion, and each ray space slice is segmented into different regions. Finally, an interpolation operation is applied to the segmented regions. Experimental results show that the proposed method can interpolate ray space data fast and render arbitrary viewpoint images, and PSNR of the interpolated ray space data is high, as well as quality of the rendered arbitrary viewpoint images.关键词：Viewpoint image rendering;ray space interpolation;feature point2487|186|0更新时间：2024-05-08
- 摘要：In order to detect the spatial conflicts that rivers deviate from the valleys in the process of topographic database updating, a new algorithm using the vector projection method to quickly extract the valley lines related to the rivers is proposed. Based on the algorithm, a method to detect the inconsistency that rivers deviate from valleys is advanced. The above algorithm has been implemented by VB6.0+MO, and achieved the better efficiency.关键词：spatial consistency;spatial database updating;spatial relation;valley line2486|185|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