最新刊期
卷 7 ， 期 12 ， 2002
- 摘要：As one of a serial of reports, this paper is a survey on multimedia researches and applications in China, 2001. Since multimedia is a cross research area, papers about multimedia technology are distributed on various journals. We saw about 2 472 papers published on 9 Chinese journals in 2001, from which we selected 394 ones on multimedia technology and application, then we made analyses on them, and the classified data were compared with that of 1998, 1999 and 2000. Looking into the data we can see that researchers in China pay more and more attention at digital watermarking, virtual reality, multimodal interface, multimedia data retrieval technology, distributed multimedia supporting technology, GIS and Digital Earth, which are also the international evolution trends in a quite long period. We present here an overview on the progress in multimedia technology in China, 2001. This will be convenient for researchers looking up references, and helpful for editors compiling journals and for authors contributing papers.2504|363|0更新时间：2024-05-07
- 摘要：Volume visualization is a method of extracting meaningful information from volume data using interactive graphics and imaging, and it is concerned with volume data representation, modeling, manipulation, and rendering. Medical visualization, one of the most successful scientific visualization research and fields, is heading from diagnosis assistance toward treatment assistance. The aim of volume visualization in medicine is to create accurate realistic views of objects from medical data. The process of volume visualization technology includes data caching, model creating, data mapping, rendering and so on. There are two methods in volume visualization: surface rendering and volume rendering. The intermediate surface representation is used in surface rendering technique. The surface information is fast represented, but the inner information is not displayed. Volume data are directly represented in volume rendering technique, so the inner information can be displayed. The primary illumination models in medical visualization and the basic ideas of various algorithms in image space and object space of volume rendering technique are introduced. The methods for accelerating processing and improving image quality at present are discussed in detail. Some experimental data and results are presented in the end.3097|445|0更新时间：2024-05-07
- 摘要：The related pose between two space objects can be determined if the coordinates of a group marks at two reference frames are measured. The precondition of the above method is that the site relationship among the marks must be changeless at the two reference frames, but this relationship will be destroyed by the computational error. So two model based visual methods include model based monocular vision method and model based binocular stereo method are introduced. The model based monocular vision method starts with the physical meaning of vision, and get the 3D space coordinates of marks by simple restrictive iterative. The model based binocular stereo method inoculates the model based monocular vision and advanced least square method to realize the model restriction. The compare analysis between the two methods and the experiment result will be given at last. Using these method, the model based monocular vision can reach very high measurement precise, and the model based binocular stereo has a little higher measure than the model based monocular vision. For any method, the size of quantization error will effect the measurement precise of visual method directly. These method has been used in mobile robot visual guided experiment at our laboratory successfully, the speed of image grabbing is 0 3 second per frame.3728|356|0更新时间：2024-05-07
- 摘要：Synthetic aperture radar(SAR) images are disturbed by a speckle noise depending on the signal(the ground reflectivity) due to the radar wave coherence. Images have a strong variability from one pixel to another reducing essentially the efficiency of the algorithms of object detection and classification. In this study, we propose to suppress this noise with a multiresolution analysis of the image. Because the Mallat decomposition can not preserve the translation invariance of an image, the filter in this paper is based on stationary wavelet decomposition. This method is used to an ERS-1 SAR image and then it is compared to a filter based on Mallat decomposition and three other classic SAR image filters, the results show that the presented approach has the advantage both in speckle suppression and the preservation of image details.2825|427|0更新时间：2024-05-07
- 摘要：The differences of the spectral curves among different objects are so obvious and important to be extracted through the wavelet transform at different scale. The traditional clustering concepts based on Euclidean distance are also redefined by the wavelet feature correlation coefficient to accurately describe the content of the remotely sensed objects. The fast clustering algorithm for multispectral images based on wavelet feature and simulated annealing increases the number of characteristic points by expanding the spectral bands to enrich the feature of the classes; clustering space is formed by evenly sampled dots; furthermore, simulated annealing leads to the best class centers on the whole scope at a high speed by decreasing the clustering scale and temperature step by step; the class characters is remained strong and durative by choosing the best dot as the class center; it also resolves the initial parameter problem of K means algorithm. The experimental results of Mississippi Thematic Mapper images show that this clustering algorithm is more efficient than other ordinary clustering algorithms such as K means and ISODATA according to the clustering accuracy and speed. Therefore, there are fairly prosperous applications on multispectral images for this fast clustering based on spectral wavelet features extraction and simulated annealing algorithm.2308|347|0更新时间：2024-05-07
- 摘要：The paper explains briefly that the remotely sensed data is non linear, and the practice of its classification by mans eyes is a process of the fuzzy inference. The fuzzy neural networks has a theory dominance, because it accords with the nature rule of classification of remotely sensed images. Analyses the architecture and principles of fuzzy ART, fuzzy ARTMAP. Discusses in detail that FasART is a neural networks based on fuzzy logic system. Put forward a simplified FasART architecture and change the general method of remotely sensed data fuzzification. With the testing of the CBERS -1 data, the results declares that the simple FasART model can be used to supervised classification of the remotely sensed images. The precision of the classification is higher than that of fuzzy ARTMAP and K means. The classification of FasART model has better stabilization and anti jamming, and has capability of dealing with non linear data especially.2805|410|0更新时间：2024-05-07
- 摘要：Discrete Cosine Transform(DCT) can decrease the relativity among the images, but we didn't know the muldimension DCT accurately. In this paper, three dimensional Discrete Cosine Transform(DCT) has been researched, its calculate method, transform and inverse transform were given, and any sizes of three dimensional function f(x,y,z) DCT is proposed, that must be cube before. As we known, it is identity physics model unify reflection into that of color image's three frame, possess same texture, same fringe and same gray change, the information of the each frame almost could reflect the source image except color, the vision similitude are very strong. In other methods, R, G, B data were transformed into Y, Cr, and Cb to decrease the color information. It is clear that if we encode Y Cr Cb data separately, among each component potential relativity can't be decreased enough. So we regard R, G, and B as a whole, transform them in the same time with three dimensional DCT. It can decrease relativity between frames than before. Simulation results demonstrate good reconstruction image quality and high compression ratio. It has good compatible with JPEG and MPEG standard.3128|393|0更新时间：2024-05-07
- 摘要：The technology of content based video analysis is a new research task in the multimedia retrieval field, but the most research is centralized in structurizing video sequence, indexing recognizing and browsing,and video inquiry is seldom focused. This paper analyses the shortcoming of the "window pairs" technology proposed by Nevenka. The way of selecting window pairs is arbitrary in that method, so it is inappropriate sometimes. This paper suggests a new method about row and column coding based on the video coding characteristics, video peculiarities and normal technology of video analysis. The character this method used describes the color and position information of video frame. Most video about different formats could be processed using the frame by frame matching and streaming inquiry technology by this method. This method needs a few bytes to figure a picture, occupying fewer memory, and using less computing. The experiment improved that this technology is robust to frame shift and luminance variable, achieved better experiment results that the Recall and Precision are highly improved.2542|322|0更新时间：2024-05-07
- 摘要：For the further analysis of texture characteristics, considering from the texture direction, a definition formula of texture direction is presented, followed by some useful explanation and discussion. To describe the texture images' direction, a discriminating function is selected properly and an eigenvector, with physical significance, is constructed successfully. About the data processing, the theory of fuzzy information, attribute means clustering and new distance measurement formula are employed, associated with an improved fuzzy clustering method, to directional texture images, satisfied results in the clustering &segmentation experiment had been obtained. Experimental outcomes show that the method has good capability to describe the directionality of texture image. Benefit from this method, stripe analyses, relatively movement analyses and other complexity image analyses regions could give an easy description to the directional objects.2961|359|0更新时间：2024-05-07
- 摘要：In the quantitative analysis of PCR gelation images system, in order to detect PCR correctly and make more accurate diagnosis of pathological situation of the person detected, a method to extract information automatically from PCR images is presented in this paper. Firstly, Gaussian filter and horizontal projection of PCR images' gray information are used to resist the effect of non uniform distribution of brightness and gain the position of the pervasive PCR. Here, a key step is to select a adaptable parameter value for the Gaussian filter. Then, the method uses smooth filtering and vertical projection to reduce the influence caused by the non uniform and moving illegibility of the PCR bands, and then extract PCR band areas accurately. Finally, it makes use of local rectification to improve the precision of the enclosed band. In practice, this method is applied to the PCR image processing and promising results are obtained. The experiment result indicates that with the PCR image been detected accurately, subsequent diagnosis and analysis can achieve fairly efficiency, so the method is of very significant practicality.2636|368|0更新时间：2024-05-07
- 摘要：Soccer robot system is a perfect platform in research and application of artificial intelligence. It is a feedback control system that closed loop by computer vision subsystem. In the vision subsystem, the design of color tag is an important factor which influences the identifying precision, real time performance and anti noise performance of the robot soccer system. Through comparing and analyzing three kinds of representative color tag design, a more reasonable design can be getting. It has many excellences such as steady centroid, non interference from opposite, simply identifying algorithm and less conglutination. Base on the character of the color tag, a fast and exact compensate approximate algorithm is given out. Through real test, the error is less then 1 degree even when the robot speed is more then 2m/s. This algorithm is a very important base for advance research and development on robot soccer.2733|332|0更新时间：2024-05-07
- 摘要：One of the main difficult problems is how to choose interesting point efficiently in 3 dimensional stereo restoration. Although different algorithms are put forward in, these methods cannot restrain disturbance noise availably and are very rigorous on the illumination with large calculation. Based on these, a kind of method to fast extract feature point in 2 dimensional image is presented in this paper. It is proved that the algorithm mentioned in this paper has following remarkable advantages such as strong anti jamming capacity, powerful adaptability to outside illumination, simplicity and fast speed of computation, etc. after comparing the result of this algorithm with that of the algorithm in on extracting feature point by experiment. Moreover, it is testified that this kind of algorithm is valid after 3 dimensional restoration of standard piece with given 3 dimensional coordination. The accuracy of feature point extracted by the algorithm is high.2679|324|0更新时间：2024-05-07
- 摘要：Minutiae extraction is one of the core techniques of automatic fingerprint identification. Routine algorithm for minutiae extraction needs to restore ridges firstly by the way of ridge tracing on thinned fingerprint images and then minutiae extraction is realized. It is trivial and time consuming to restore ridge structure. Aiming at the problem, an improved algorithm for minutiae extraction is brought out. First, the set of original minutiae is directly extracted from thinned fingerprint images. Second, various noises in fingerprint images and their properties are analyzed and distributing regulation of pseudo minutiae is generalized. Last, combining with the information of local ridge direction, special algorithms are designed with respect to various noises to delete pseudo minutiae from original minutiae set. The remainders of original minutiae are viewed as intrinsic minutiae. To verify the performance of the improved algorithm brought out in this paper, contrastive experiment was conducted with routine algorithm. Experimental results indicate that computational time is reduced effectively with the improved algorithm and the accuracy of minutiae extraction can fill the demand of application.3132|365|0更新时间：2024-05-07
- 摘要：To realize automatically detecting and recording system of freight trains, it's necessary to detect whether the locks of freight trains exist by the image of freight trains when they are entering railway stations. An approach of fast lock detection is put forward, which is based on RBF neural network and improved genetic algorithm. In this method, firstly extract the projection features of images, linear moment features of edge images, and the features of gray histograms of images, and these features reflect different characters of targets from different points of view, and then they are normalized to the input vector of RBF neural network, and RBF neural network is used for detection and location. At the same time, improved genetic algorithm is used to search the whole image, and searching process is speeded up because the genetic algorithm is a parallel and robust algorithm. In addition, image preprocessing is not done alone, and gray variation of images is eliminated during the process of feature extraction. Experiment results show that the method can overcome the problems of many types, deformation, and the variation of environmental brightness, and have the fast speed and high success rate of detection, and can be put into practical application. Therefore, the method has significant engineering value.2390|373|0更新时间：2024-05-07
- 摘要：The detecting welding seam has been a difficulty problem in pipe manufacture for a long time, this paper proposed a new method for it based on the wavelet transform theory and its technology of edge detection. The method mainly relates to the pretreatment of the seam image, the wavelet transform based on three order spine function, the scale parameter effect on the transform results and the correlation analysis on the ultimate results. By the analysis on the theory and the simulation on computer, the method shows higher estimated precision and better capability of anti disturb. In the other hand, its complication is not high. So it can be used into real time control system.3079|346|0更新时间：2024-05-07
- 摘要：In order to auto recognize typhoon on satellite pictures, an algorithm is proposed in the paper, which is based on the character of layout of texture direction. The texture direction of typhoon in the satellite pictures disposes with obvious character. A new conception named vector square is introduced to measure the character of texture direction of an image. The process of how to calculate the vector square is presented in detail. According the algorithm, a small window is slid throughout the image, at the same time, the texture direction of every pixel in the window and the vector square of the window are worked out. The size of the sliding window can be adjusted automatically to cope with various typhoon images. Compared the vector square with the threshold, it is clear whether or not it is a map of typhoon. The threshold of vector square would be adjusted according the intensity of typhoon to enhance the adaptability of the algorithm. The result proved that the algorithm is accurate and widely available, which can recognize typhoons in different types and phases.2575|333|0更新时间：2024-05-07
- 摘要：This dissertation expands on a new algorithm, which has the function of analyzing edge form with the chain code. By the introduction of the concepts of relative chain code and absolute chain code, we proposed a simple and direct algorithm to compute chain code sum(average chain code). We found that tangent direction(slope) of edge point can be figured by the absolute chain code sum of three sequential points and curvature of edge can be figured by the difference of chain code sum between the three sequential points ingoing and outgoing. Furthermore, we provided the criterion distinguishing edge inflexion and sleek curve section and the method computing inaccurately curvature radius and approximate perimeter. In the end of the paper, we introduced the use method with cell edge hollow repairing and overlap or conglutination cell segmenting for example. This algorithm has proved high speed and has a good effect of cell segmention on more than twenty groups of conglutinate and absent cells which are gathered from three kinds of cells.关键词：Chain code;Differential chain code;Chain code sum;Form analysis3412|388|0更新时间：2024-05-07
- 摘要：Surface reconstruction from unorganized points has numerous applications, and it is widely studied all over the world nowadays. Crust algorithm, which is based on Voronoi diagram and its dual Delaunay triangulation, can reconstruct the original surface from sufficiently dense sample point set. It is simple and direct in theory and its result is also very fine. However, the algorithm is restricted in the practical application because of its long running time. In practice, the sampling density required by Crust algorithm for successful reconstruction is varied in different area: dense in detailed areas and sparse in featureless ones. Based on this fact, a method for non uniformly sampling the dense data set according to the local feature size is presented in this paper. With the guarantee that the remaining points are sufficient to reconstruction, the amount of points used in reconstruction is largely decreased, and then the speed of reconstruction is improved. The results show that the details are kept well in the reconstructed surface. However, since the points are sparse in featureless region, the triangles approximating the surface are comparatively large there. That makes reconstructed surface look very coarse. Gouraud shading can give an acceptable visual effect. The method of non uniformly down sampling can also be used to decimate the vertices of mesh to realize mesh simplification.3991|369|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