最新刊期
卷 6 ， 期 8 ， 2001
- 摘要：Data mining is an emerging research field in database and artificial intelligence. In this paper, the data mining techniques are introduced broadly including its producing background, its application and its classification. The principal techniques used in the data mining are surveyed also, which include rule induction, decision tree, artificial neural network, genetic algorithm, fuzzy technique, rough set and visualization technique. Association rule mining, classification rule mining, outlier mining and clustering method are discussed in detail. The research achievements in association rule, the shortcomings of association rule measure standards and its improvement, the evaluation methods of classification rules are presented. Existing outlier mining approaches are introduced which include outlier mining approach based on statistics, distance|based outler mining approach, data detection method for deviation, rule|based outlier mining approach and multi|strategy method. Finally, the applications of data mining to science research, financial investment, market, insurance, manufacturing industry and communication network management are introduced. The application prospects of data mining are described.3231|361|0更新时间：2024-05-07
- 摘要：The aim of this research is to realize multiple knowledge sources cooperation of problem solving in real time image processing system under parallel environment. This paper introduces a kind of expert system tool language for image understanding, V language (ver 3). The discussion of V.3 is concentrated on its knowledge representation, inference mode and control mechanism. In V.3, knowledge source is consisted of frames of typical object in image understanding, and is controlled by layered blackboard. The inference mode is mixture. Apart from these, several new ideas are also introduced. The object oriented frames present both the static knowledge (about the object) and the dynamic knowledge (methods/procedures using for image processing) at the same time. The post office may record and transmit messages between frames. Blackboard, served as control mechanism, is layered according to knowledge sources. At the end of this paper, an example, the recognition of bridge over water is discussed in detail.2467|352|0更新时间：2024-05-07
- 摘要：Camera calibration, which is to calculate the intrinsic and extrinsic parameters of a camera in viewing, is one of the basic tasks for analyzing the geometry of a scene and determining the shapes and locations of objects in vision. Since the optical system of a real camera is not correspondence to an ideal pinhole optical system in generating a perspective view, there exist many types of distortions. These distortions must be taken into account in three dimension geometry reconstruction when high precision is needed, and this will cause the relation of perspective transformation nonlinear. Many efforts have been being made to solve the nonlinear calibration problem. This paper deals with camera calibration while radial distortion exists, which is the most common distortion and has a noticeable effect. A new multi step decomposition linear approach of calibration is introduced in this paper, which eliminates the nonlinear iterative calculation processes. The running experiment results show that the new approach is very efficient and the accuracy is good enough.3022|334|0更新时间：2024-05-07
- 摘要：With emerging of new video coding standard MPEG 4, segmentation and tracking of video moving object is a key research topic in the field of interactive multimedia. In order to support content based functions in the visual standard of MPEG 4, each frame of video sequences should be represented in terms of video object planes (VOP's). Segmentation and tracking of video moving object is used for guiding the extraction of video object plane from the video sequence. This paper introduces spatio temporal information to automatically segment and track video moving objects. Firstly, a fourth orders statistic hypothesis testing in inter frame difference is used to automatically separate the moving areas from the background in a general video sequence. Secondly, Change areas are extracted as markers of the corresponding moving object. Using the marker, watershed algorithm can locate the precise boundary of moving object. Lastly, using segment moving object as the initial model, moving object can be tracked by Hausdorff distance on a sequence of video, and the model is updated every frame to accommodate for changes in shape. Experimental results demonstrate that this method is effective to segment and track video moving objects. More, the number of parameters updated is the least.2672|331|0更新时间：2024-05-07
- 摘要：With the complication of great pattern design, the speciality of designer, the dispersal of design location and the integration of computer networking computing, It is necessary and possible to develop computer supported pattern design system. This paper applies CSCW technology in the field of computer art and construct cooperative pattern design model. It focuses on the cooperative pattern design and analyze modes of the cooperative design, the structure, the relative management and control. Firstly, it presents the basic process of pattern design system, analyzes the basic requirement of computer supported multi user pattern design, describes the interactive mode and cooperative mechanism of multi designer, whose tasks are dispersed in temporal and space, and are dependent on each other. It defines the elements in a cooperative pattern design system and builds up a logical model of pattern design system. It studies not only the pattern management model and concurrent control of multi designers, but also presents design conference management model, the synchronous conference model, message management policies model in a pattern design system. Finally, it gives the structure model of pattern design system.2468|395|0更新时间：2024-05-07
- 摘要：The Hough transform is useful for detecting or locating translated two dimensional shapes. However, a weakness of the Hough transform is its storage requirements and hence the acutely increased computational complexity resulting from the increased dimension of the parameter space of the shapes. Significant improvement in speed and storage are needed for efficient use of the Hough transform. The size of the parameter space must be reduced significantly to save storage and to minimise the associated search task. In this paper, the concept of boundary dual points of shapes is present and the existence of dual points of symmetric shapes as well as their invariant under translation,scale and rotation are proved. A quick algorithm of the image detection specially for circles and ellipses is presented by combining the dual points with the Hough transform. And the shortcoming of the formula in Ref. is replenished. The experiment examples indicate that this algorithm is easy in use and insensitive for noise and occlusion as the Hough transform and has high calculation speed. It can be flexibly used for some difficult shape detection problems.关键词：Image detection;Dual points method;Hough transform;Combining transform;Invariability2281|332|0更新时间：2024-05-07
- 摘要：Many techniques in pattern recognition, robot vision, segmentation, feature extraction and etc require edge detection as a basic instrument. Although many methods have been suggested, the performance is quite different for different types of images and there is still not a general method. In this paper, we proposed a novel edge processing approach which makes the detected edge maps more valid and more ideal, instead of introducing a new edge detection method. The proposed method uses genetic algorithm to optimize the edge maps after edge detection. First, it encodes the edge maps into a two|dimensional binary array and determines the fitness based on valid edge structural templates for each individual. Second, the parent population is generated by changing a small part of pixels in edge maps randomly. Then the proposed method re|allocates edge points according to the genetic operators such as crossover and mutation, and forms their offspring population. Finally, elitist section is adopted to drive the genetic procedure approaching convergent state. When the genetic algorithm is converged, the optimized edge maps can be obtained and the noises in edge maps can be effectively reduced. The proposed method has been carried out for both the artificial and natural images, and the experimental results have shown its good performance.2370|335|0更新时间：2024-05-07
- 摘要：Image segmentation based on watershed method always results in over segmentation and makes the contours of the objects buried in the irrelevant watershed lines. In this paper, we first smooth the image while preserving the edge by nonlinear diffusion method, which can reduce the noise in the image and at the same time the numbers of the region minimums of the gradient image that is equal to the region numbers of watershed segmentation. Then from the result of the initial watershed segmentation which is organized by the Region Adjacency Graph (RAG), we execute the bottom up hierarchical region merging according to the region average gray value similarity and the edge strength criterions that can settle the over segmentation problem well. The region average gray value is a rough characteristic about the region while the edge strength criterion is local. Similarity criterion of the average gray value of the region is first used in Merging operation. Then upcast the RAG and continue to merge according to edge strength. The results of the experiment using 2D real images show that this method can provide accurately localized and closed region contours.2633|363|0更新时间：2024-05-07
- 摘要：With the development of the target identification and tracking system,it is very necessary to realize the automatic capture, identification and tracking of the target. In this paper, a target searching algorithm is proposed. The algorithm takes the advantage of the gray level distributions and it automatically searches for the target inside a given region. We suggest that the searching begins at a given region, which contains the target and there are gray level differences between the target and the surrounding backgronds. Such suggestion is valid and the algorithm can be regarded as an active boundary method. At first, it reduces the searching area gradually and it locates the target with the gray level gradient information of the image, at last the boundary of the target is located. During the searching process, only few control points are computed and the initial positions of the control points are not limited strictly. Therefore it can search quickly. The experiment results are satisfactory for the targets with different size and shapes.2685|360|0更新时间：2024-05-07
- 摘要：Scaleless GIS is one of the key technologies of digital earth and web gis. With the development of GIS, the current GIS techniques couldn't meet the requirements of the information society very well. The ability of deriving small scalem ap from large scale map automatically is very important in modern GIS. To derive small scale map from large scale map in scaleless GIS is related to cartographic map generalization technique. Based on the description of the amount rule of selection, quality principle of selection, combination principle and attribute data generalization. A hierarchical organization strategy of polygon map layer, C Tree, and the corresponding algorithm of generalization in scaleless GIS is presented in this paper. The algorithm based on C Tree can complete the polygon selection and combination operation in high efficiency for given large scale polygon map layer and small output scale. This strategy has been implemented into a 863 project of China and the experimental results show that it can complete polygon layer generalization with high efficiency and can maintain the appearance and the topological structure of map well.关键词：GIS,Scaleless GIS;Cartographic generalization;Line generalization;Polygon simplification;Polygon generalization2702|368|0更新时间：2024-05-07
- 摘要：A new algorithm for color image quantization based on the pattern recognition technology is proposed in this paper. First, the color samples in a color image are grouped together, and the initial representative points of the categories are chosen based upon a method of combining maximum frequency degree with maximizing minimum discrepancy, that is , an optimum seeking method of initial value of clustering center. Then both the clustering criteria of Euclidean distance in clustering analysis and the gravitational center method in mechanics are used to determine the vector values of the new clustering region centers, and the satisfying clustering effects can de obtained. This is a fast statistical clustering algorithm based on maximizing minimum discrepancy (FSCAMMD). The presented algorithm can overcome the shortcomings of the seeking method of initial value of the clustering center of SCA algorithm. Both the total mean square deviation and lack fidelity of images quantized by the present algorithm have a relatively big reduction and the effect of color image equalization is better than that of SCA algorithm and other clustering algorithms.2501|308|0更新时间：2024-05-07
- 摘要：The paper presents a new adaptive fractal image IFS encoding algorithm based on classification. The classification is according to the contrast sensitivity of human visual system(HVS). It also gives the concept of generalized creditability (GC) and uses GC of between a certain range block and its domain blocks to determine the matching threshold in quadtree partitioning in partitioned fractal IFS encoding. This adaptive algorithm avoids the problem which is caused by the conventional algorithm based on prefixed threshold: The conventional algorithm can not adapt to input images with different complexity. The adaptive algorithm takes good advantage of the statistical characteristics in them, and calculates the threshold according to the variance of the input range block. In this way, the adaptive algorithm improves the encoding efficiency and makes the IFS image encoding more practical. Experimental results show that this new algorithm can adapt to different input images automatically while the decoded images are still satisfying. It also compares the adaptive algorithm with the conventional fixed threshold algorithm.2243|329|0更新时间：2024-05-07
- 摘要：The classical image restoration filters are deconvolutioninverse filtering),Wiener deconvolution, SVD pseudoinverse filter, Kalman filter and maximumentropy restoration etc .Because the universality morbidity of image estoration process, the deconvolution is only suitable for high SNR(Signal to Noise Ratio). Wiener deconvolution needs the knowledge of general stationary process and the correlation function and power spectrum, which makes it difficult to be used in practice. SVD pseudoinverse filter, and Kalman filter is very complex and has large computing work, which restricts its use in practive. The other methods based on if them model, Gauss and Gauss--Markov stochastic process have also been restrictedin real image processing bacause of theri complex. Because factors affecting imaging and making image degeneration are fuzzy and uncertain, it is very difficult to build accurate mathematical model of image degeneration and, which therefore make it impossible to restore the image. Due to its capability of nonlinear mapping and synthesis, CMAC can effectively achieve image restoration by learning reverse process of image degeneration, which resolves the shortcomings of traditionary methods. The simulator results showed that CMAC neural networks are able to restore the degeneration image effectively. The learning algorithm of the network is simply. The present method is convenient for real|time restoring image.2309|319|0更新时间：2024-05-07
- 摘要：Coding for binary images based on mathematical morphology is a hotspot of research in recently years. A new method presented here based on ultimate erosion, subset of skeleton, removes more redundancy using morphology analysis. Following a review of the theoretical background and old research work on this area, such as a method based ultimate erosion for coding binary images proposed by kresch etc . We proposed a new theorem to predicate non skeleton pixel after analyzing the old method . It was proposed a new condition to predicate non skeleton pixel in the new theorem. Compar ed with the old theorem , pixels...More pixels can be determined to be non skeleton when being encoded under the new theorem and the efficiency is increased. The corresponding corollary was also presented to realize the theorem in computation. The proposed method is suitable for progressive transmission and can be fast implemented by hardware. A binary image "tools" was tested by this method and a better result, 0.065 bit/pixel, was gotten, compared with other methods, such as runlength, quadtree or chain coders. It is also better than the old method ,0.071 bpp.2682|420|0更新时间：2024-05-07
- 摘要：It is well known that there are a lot of noise data in many digital images. In generally speaking, the embedded noise data in an image are not only harmful to view the image, but also there is less correlation among noise data and original image data. So the noise data are hard to be compressed by general compression methods based on predictive or schotistic coding. Therefore a new method to achieve nearly lossless compression of an image is proposed in this paper. In the first step in the proposed method the noise data in an image are eliminated based on a noise model, and then the resulted image is as better as the original one for the usages. In the second step, the image is encoded by using a region adaptive subband compression algorithm without loss any data. Obviously, after decoding, the reconstruction image is a nearly lossless image without harmful to the usages. Coding time of the proposed algorithm is affordable thanks to fast convergence of the algorithm. Coding could always be performed in real time. The experimental result shows that the compression scheme provides impressive performance such as high signal to ratio and high compression ratio. According to the theory of the algorithm, the noise elimination algorithm based on noise model can also be extended to the other compression algorithms such as DCT, DPCM, JPEG, SPIHT and MPEG.2387|310|0更新时间：2024-05-07
- 摘要：World Wide Web Geographical Information System(WebGIS) is the combination of Internet and modern Geographical Information System. Since its naissance, it has brought great influence to the national economy and people's lives. After the study of WebGIS and its solutions,this article proposes and implements a new MAPGIS -Web solution. This solution is not only based on ASP components technology, but also based on MAPGIS--a Chinese GIS software. It adopts the fashionable three layers Client/Server architecture as its system structure. In this architecture, the GIS Data Server layer take charge of restoring the GIS data, and the WebGIS Server layer take charge of running the GIS data, and then the Browser layer take charge of submitting the command for the GIS data and getting the common HTML as the result. So this solution has a good platform independent quality at the Client, and it's can be reused easily, and it can help people to set up WebGIS Server quickly and easily.2572|321|0更新时间：2024-05-07
- 摘要：A 3D realistic image is a core technique of scientific visualization, computer animation, and virtual reality, and an important technique of spatio temporal GIS. OpenGL is the current international standard of 3D image. This article discussed the implementing method for the representation of terrain realistic image based on OpenGL, such as the digital terrain model mapping, the blend of remote sensing image and terrain data, the blend of thematic information and terrain data, animation, virtual reality, and etc.The mapping methods respective to the DTM format such as TIN, GRID and Contour, the direct and indirect blending methods of remote sensing image and terrain data, and the blending method of thematic information and terrain data based on the 2D transparent texture were presented mainly. Based on aforementioned methods, a flood routing visualization system for Jingjing diversion area was developed and the results fit the anticipant goals well.2676|331|0更新时间：2024-05-07
- 摘要：Based on the detailed analysis of MC(Marching Cubes)algorithm which is widely used for extracting Iso surface from volume data filed, this paper presents a new method for 3D reconstruction and visualization: directional smooothing and polygonal merging on the extracted iso surface. The directional smoothing method decomposes the Iso surface into a 3D coordinate field and a vector field, and then performs a smoothing process in this vector field. By vector field smoothing, the reconstructed results and the rendred anatomical details are greatly improved. The method of Iso surface merging is analyzed in this paper. According to correct merging principles and threshold value, this method transforms the triangle composed iso surface into polygon composed iso surface, and it can greatly speed up the 3D rendering process without impairing reconstruction fidelity and anatomical details. A modified approach of 3D reconstruction and rendering is presented in this pqper, which can improve the results of 3D reconstruction, and at the same time, speed up the 3D rendering process.2877|343|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