摘要:With the growing availability of digital transmission links, progress in signal processing, VL S I technology and image compression research, digital video coding technology has developed from a purely academic research in to a highly commercial business. To follow the accelerated speed of technology development, the standardization process has become sign if i cantly efficient and fast. The actual work of the standardization bodies has changed considerably and has evolved from discussion circles of national delegations in to in ternational collaborative R&D activities. It has to be understood that video coding standards have to rely on comp rom ises betw een w hat is theoretically possible and what is technologically feasible. In contrast to the early standards, recent standards are in tended to be generic and have wide applications.
摘要:This paper proposed a contour/texture-based inter-frame coding method, BBSMF, which is
suited for low bit-rate video coding. The BBSMF method first segments an image into some homgeneous regions, and generates variable size blocks to approximate these regions; Then it applies motion detection technique to separate the image into moving regions and still regions, the moving regions are estimated to get motion information by which a motion field image is constructed; Finally, the motion field image is segmented, the contours and the motion mean errors of textures are coded. Computer simulation shows that the BBSMF method gives better performance than the H.261 method in terms of not only CR but PSNR and subjective quality of reconstruction images. In this paper, some discussions are also made on subsample coding, motion compensation of moving scenes and how to select the number of predicted frames, etc..
摘要:A new source model for image sequence based on detecting and optimization image edge established as well as the coder system is simulated by computer. On the basis of subjective vision characters, the basic idea of this paper is: the edge information of an image sequence is the most significant factor which can describe the image's characters both in space and time domain. By this scheme, neither any priori knowledge of an image is needed nor too much complicated techniques must be applied for model's construction or realization. According to the source model, the method of image matching using geometrical constrains is achieved for motion analysis; means of prediction between frames were extended by the concept of general motion introduced; and the algorithm of general motion analysis is completed. The scheme is also characterized by fitting gray value distribution of images using normal polynomials.
摘要:The segmentation of character and graphics is a fundamental step when processing engineering
drawings using a computer. However, it is hard to deal when the character to uches the graphics. This paper focus on this problem and gives a solving method. The method is called "the technique by shield" . It shields lines in a graphics before segmentation. Some tests have shown good results of segmentation compared to which does not use this method in a graphics with a lot of touching points between the characters and graphics contours.
关键词:Engineering drawing processing;Segmentation of character and graphics;Touching characters,
摘要:Stereo display is one of the key technology adopted by VR(Virture Reality),but it is very important how to organize and store a large number of the colour stereo image pairs.Because of the correlation between the pair,the classic coding algorithm(JPEG,MPEG etc.)is not suitable.The paper brings forward the new scheme for the coding,which usese the correlation between the stereo image pair and combines JPEG,and achieves the higher compression ratio and the higher SNR.The result shows the advantage of the new scheme.
摘要:In this article a new algorithm of three dimensional reconstruction is developed based on voxel.
According to the adjacent of voxel, adjacent relation of voxel surface units is defined using the circuits around voxels. Thus the surface of an object is redescribed. This algorithm was successfully applied in the reconstruction and display of the CT images of human liver. The three dimensional reconstructed audio visual images of the branches of portal vein can be reappeared. The results can render some experimental evidences to the correct localization of CT ,MRI of liver lesions, and invasive radio therapy, as well as regular hepatectomy.
摘要:Given two polygons P and Q in a plane and a translation vector, t∈R2 the area-of-overlap function of P and Q is defined as Ar(t) =Area(P∩(t+Q)), where t+ Q) denotes the new polygon formed by translating Q with t. In order to rapidly compute the maximum value of the area-of-overlap function, an optimal computing strategy is developed in this paper, which includes both combining genetic algorithm with gradient algorithm for rapidly global searching the maximum value of the function and using modified scan-line algorithm for computing the value of the function efficiently in local range.
关键词:Scan-line fill algorithm;genetic algorithm;polygon;Area-of-overlap
摘要:Best first search is a general heuristic search algorithm for shortest path problem. Its applicability is limited by its exponential memory requirement. In this paper, a new data structure for storing traffic network is presented. The search procedure is improved on the advantage of this data structure. On the other hand, a new heuristic estimate function based on the characteristic of GIS is discussed in this paper. The improved algorithm is used in a traffic analysis system.
关键词:Shortest path problem;Best first search;Geographical information system
摘要:Based on analyses of traditional development models of project-oriented geographic information
system, this paper proposes the development model of geographic information system shell(GISS) and discusses its characteristics. Combined with the realization of geographic information query shell(GIQS), the theories and methods about developing GISS are described.
关键词:Project-oriented geographic information system;Development model;Geographic informationsystem shell(GISS);Geographic information query shell(GIQS)
摘要:3D object construction is the foundation of 3D data visualization. In this paper, for the first
time, the CT imaging mechanism is introduced in to 3D object construction. The inverse problem of the slice integral has been studied and a theory and implementation of 3D object restoration has been proposed. The proposed method has great advantage over the traditional 3D reconstruction theory. It can restore the 3D objects accurately even for the thicker slices. The proposed method is a totally new method and of great theorial and practical sign if icance.