最新刊期
卷 4 ， 期 3 ， 1999
- 摘要：Functional image processing techniques with positron emission tomography (PET) have been widely used for clinical diagnosis and research.In this paper, the basic principles of the functional imaging and its application are introduced.Six most commonly used functional imaging algorithms (NLS,AR,PGA,WIM,LLS,GLLS)are discussed and compared.2805|150|0更新时间：2024-05-08
- 摘要：The Progress of the research of evaluating image segmentation algorithms is discussed in this paper.First,three basic theoretical frameworks,which form the important research contents of evaluation are reviewed and classified as follows: Image metrics,Testing image set,Performance evaluation and analysis of experiments.Then,the evaluation methods of image segmentation are summarized: Statistical approaches,Analytical approaches,AI based evaluation,Hybrid approaches.Finally,The comprehensive evaluation model,a new idea and way to comprehensive evaluating of image segmentation is discussed.2426|132|0更新时间：2024-05-08
- 摘要：This is an overview on fractal image coding. On the base of collecting and reading related documents, we describe the current research status of fractal image coding and make a remark on its prospect.3224|120|0更新时间：2024-05-08
- 摘要：Traditional DPIV (Digital Particle velocimetry) methods mostly are based on area correlation. Though proven to be very time consuming and much error prone, they are widely adopted because of their conceptual simple and easy to be implemented, and also because there is few alternative. This paper provides a non correlation, conceptual new, fast and efficient approach for DPIV which takes the nature of flow into consideration. An Incompressible Affine Flow Model(IAFM) is introduced to describe a flow that incorporates rational constraint directly into the computation. This IAFM, combining with a modified optical flow method named Total Optical Flow Computation(TOFC), provides a linear system solution to DPIV. Experimental results on real images demonstrate our method to be a very promising approach for DPIV.2289|174|0更新时间：2024-05-08
- 摘要：This paper analyzes the possibility of phase unwrapping using Green’s second identity. Assuming that the normal deviation along the border is known, this paper addresses a new phase unwrapping algorithm based on use of Green’s second identity and compares the performance of the new algorithm with the algorithm based on use of Green’s first identity as well as LMS.Results have proved that the new algorithm is equivalent that of LMS as well as Green’s first identity.关键词：Green’s second identity;phase unwrapping;Poison’s equation;Least-squares3297|140|0更新时间：2024-05-08
- 摘要：With recognition methods of circular and elliptic objects taken into consideration, this paper presents a new algorithm for recognition irregular quasi circular object, takes log cross section images as examples. The new algorithm combines two methods of cluster analysis and fuzzy recognition, recognizes step by step from coarse to fine, sets up classifier at every level. It solves the problem of uncertain objects classification successfully, implements recognition of irregular quasi circular object accurately.关键词：pattern recognition;cluster analysis;fuzzy recognition;Membership grade2833|135|0更新时间：2024-05-08
- 摘要：Invariants can be divided into algebraic invariants and deferential invariants. This thesis gives out an algorithm to recognize objects based on algebraic invariants. It has two parts: principle and algorithm. In the principle part, the concept of‘bases’and‘coordinate’is given under recognition using algebraic invariants. The algorithm part includes‘training’and ‘recognition’. At last the thesis gives out the experimental results of this algorithm.2148|155|0更新时间：2024-05-08
- 摘要：In this paper a method of extraction vegetational informationfrom serious scattered IR image is described. The method is basedon spectral analysis and color regulation of IR image. The methodis enhancing near infrared band and weakening other bands by some rules, and making the color saturation to a suitablerange.Theexperimental results show this method is effective.2151|143|0更新时间：2024-05-08
- 摘要：Because fingerprint image is composed of strip, it is difficult to increase compress ratio by tranditional fractal compress method based on block IFS. A new fractal compress method is presented. At first the image block data is formed into a serial data according to the direction of the strip. And then the data is coded by fractal image compress method in one dimension. The theory analysis and experiment results show that this approch can increase image compress ratio and improve the quality of the recovered image. At last a method to eliminate the block effect based on direction filte is presented.关键词：Fingerprint;fractal;Image compression;Strip direction2069|152|0更新时间：2024-05-08
- 摘要：According to the spatial and spectral structural characteristics of multi spectral image data, this paper present a concept of Multi Pattern Prediction : given a principle, any pixel in an image can be predicted by any prediction function selected from an alternative function set in order to decorrelate the image more efficiently ; at the same time, taking the advantage of spectral structural correlation, the spectral adjacent pixels are decorrelated by the same prediction function, so the additional cost of storage in Multi Pattern Prediction would be reduced smartly. We present the Minimum Entropy Principle as the theoretic principle to select the predition function. And we get an equivalent principle named as Maximal Frequency of Minimum-Error Standard. Experiments on TM images show that this method can decorrelate images much more efficiently and lead to higher compression ratios.关键词：Multi pattern prediction;Spectral structural correlation;Minimum Entropy principle;Maximal Frequency of Minimum Error standard2257|117|0更新时间：2024-05-08
- 摘要：This paper presents a new image coding scheme which integrates the wavelet image coding and the fractal image coding together. There are two similarities emerging after an image has been decomposed with wavelet transform, one is the self similarity of each subimage and the other is the co similarity between the subimages with the same orientation at different resolution. It is the co similarity that we make use of in our scheme. The experimental results provided in this paper justifies the good performance as well as efficiency of our scheme.2060|156|0更新时间：2024-05-08
- 摘要：A wavelet based on iteration regularized image restoration algorithm is presented. Consideration is given to both the noise suppression and edge preservation. The selection of regularizing operator and regularization parameter is dependent on the frequence and direction feature of each subband.The experiment results show that the proposed algorithm is superior to the general spatial domain algorithm in the quality of the reconstructed image and iterative convergence speed.2283|133|0更新时间：2024-05-08
- 摘要：A new method of edge evaluation is proposed, based on the good form of the detected edge and different kinds of errors that can occur in edge detection. It combines traditional edge evaluation methods and has the advantage of not requiring ideal edges to be known. This method is proved to be effective and feasible by experiment.2417|147|0更新时间：2024-05-08
- 摘要：This paper presents the powerful ability of the neural network, and puts forward a new structure of the network in edge detection. On the basis of the understanding of Marr algorithm, the characteristics of Marr are taken great consideration into the structure of the network so as to improve the performance of algorithm. Meanwhile, under different noisy environment, among different algorithms in edge detection, comparative analyses are made to reveal that new method owns the powerful ability of anti noise, and that more applications are done well in the robot vision.关键词：Edge detection;neural network;computer vision;Marr feature2293|155|0更新时间：2024-05-08
- 摘要：This paper gives a method of texture matching based on feature symbol random field created by wavelet. On the image wavelet decomposition on some scale,we use the four subimages of the last decomposition to create the feature symbol random field through vector quatization.Based on the FSRF,we make cooccurence of the FSRF,then get the feature vector of the original images.At last,according to distance,we can fast match two texture image.2264|118|0更新时间：2024-05-08
- 摘要：The paper focuses on the stripe noise that only but often occurs in SAR image. The paper analyses the spectrum of the image with noise, discusses the limitations of the repairing methods which always being used and presents a new method, Orientation Filter, to eliminate noise. The correctness of the method is verified in this paper, and the steps of applying the orientation filter are provided. Experiments proved that orientation filter can satisfactory filter the stripe noise and preserves as many target characteristics of original images as possible.关键词：Contend based retrieval;Information retrieval;Multimedia database system2642|140|0更新时间：2024-05-08
- 摘要：The paper focuses on the stripe noise that only but often occurs in SAR image. The paper analyses the spectrum of the image with noise, discusses the limitations of the repairing methods which always being used and presents a new method, Orientation Filter, to eliminate noise. The correctness of the method is verified in this paper, and the steps of applying the orientation filter are provided. Experiments proved that orientation filter can satisfactory filter the stripe noise and preserves as many target characteristics of original images as possible.2454|267|0更新时间：2024-05-08
- 摘要：Based on the synthetic analysis of the planar error band’s development in GIS, the paper derives the general boundary curve’s formula of the“g-band”on the random line segment’s positional uncertainty by means of the envelop theory, analyses the configuration、shape decision and the“g-band”visual figure according to different examples, display the error region size of the line segment’s integral positional uncertainty to the users.2749|121|0更新时间：2024-05-08
- 摘要：In this paper, we present an optimized Freeman chain coding method for bi level line structured images and discuss its application in the storage and automatic plotting of engineering drawings. Compared with the traditional Freeman chain coding that uses single ring or multi ring image quantizers, the proposed method is based on already quantized dot matrix digital images, and is a lossless coding method. We select the code length according to the statistics of the length of straight line segment in the image and thus minimize the average number of bits needed to encode the segment. Theoretical rate analysis and several experimental results show that the proposed method achieves lower bit rate than the existing one and can be easily applied to the storage and plotting of engineering drawings.关键词：Vector chain;Freeman chain code;Bi level images;Image encoding2528|234|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