最新刊期
卷 6 ， 期 11 ， 2001
- 摘要：In photogrammetry and object recognition, it's necessary to introduce the sensor planning strategies to improve the measurement accuracy. Sensor planning is a new research area in computer vision. Planning the parameters of camera and illumination purposefully can make task be performed effectively. In this paper some affecting factors which are essential to sensor planning are analyzed firstly. They are the parameters of camera and illumination, constrain of feature detectability, the models of sensor and object. After some specific sensor planning systems were summarized, the sensor planning systems are classified based on the methods they used, the goals, the number of sensors and a prior information. Then the problem of sensor planning is concluded to the optimization process which can be done in high dimensional space composed of the parameters of camera,lens, illumination and object models. Finally we show some possible developing directions , namely single camera system in active machine vision,multi camera network design solution, new optimization and combinatorial search method, rebuilding the uncertainty of characters and multi sensor data fusion.2549|329|0更新时间：2024-05-07
- 摘要：Motion estimation of human from a single or a monocular sequence of image is an important research field in Computer Vision and Image Processing. In this paper, based on constraint fusion, we proposed 3D motion estimation from the monocular perspective images without any special markers. We consider the arm as stick model with three joints, and the link between every two neighboring joint points is regarded as rigidity. Based on the knowledge of anatomy and kinematics, we exert coplanar constraint, rigidity constraint, smooth motion constraint on this model. By pre processing the image sequence, we semi automatically yielded 2D correspondences of joint points of arm among images. Then, by constraint fusion, 3D relative structure of arm can be determined in a scale factor. The experiment with real images is included. In order to testify the validity of the proposed method, we compare the 3D motion of human arm obtained by the motion capture system with that obtained by our method. The result shows that the method is very efficient.2349|339|0更新时间：2024-05-07
- 摘要：Smoothing is an important topic in image processing area. However, smoothing and edge preserving are hard to achieve simultaneously. In this paper, we propose an improved adaptive detail preserving smoothing algorithm using multiscale and multidirectional mask to deal with this issue. Before smoothing, complexity of texture is calculated according to the edge map. We ourselves define a measurement of complexity of texture which is simple and fast for operation. For the area with large value of texture complexity, that is the area with details, "stick" is used for smoothing to preserve these minute features, while for the area with small value, block mask is employed to speed and emphasize smoothing. As to using "stick", eight different directions are considered to make the most suitable mask take effect. To achieve that, we take the standard deviation as the criterion, small standard deviation denoting homogeneous area and suiting for being taken as the smoothing mask. This method, to some extent, is effective for both smoothing and edge preserving, mainly because it selects different mask according to the local feature of the image utilizing texture analysis. Compared with other methods, including traditional Median Filter, MWMF, Nagao's Filter, DMF and PDE, this algorithm is easy and faster to implement in computer and has better performance.2358|370|0更新时间：2024-05-07
- 摘要：This paper put forwards a new novel texture image retrieval method by using the advantage feature of tree structured wavelet transform. This method can produce eigenfeature at different scales precisely by decomposing the texture at multi scales and multi directions adaptively under the energy rule. In terms of these image eigenvalues, the proposed method also suggested a modified algorithm, named principal eigenvalues analysis(PEA), which can cut down eigenfeature dimensions to the low space effectively. It was confirmed that on the capability of the hierarchical way provided by this method the use oriented application processing can allow users to carry out different retrieval on accord to users' requirements, which is called a coarse to fine retrieval. It was indicated by experimental results that the modified texture retrieval way has powerful practical merits for it can improve the retrieval accuracy efficiently and speed up the retrieval processing.2278|358|0更新时间：2024-05-07
- 摘要：This paper presents a method of implementing the digital image warping based on control points by combining smooth unequally spaced interpolation and space variant liner image interpolation method. The smooth unequally spaced interpolation is used to establish the locality relationship between the original and the warped image. It maps the coordinates in original image to corresponding position in warped image. The space variant liner image interpolation provides an efficient way to reconstruction an image. The image interpolation method has been proved simple with a good visual result. During the warping process the operation of image coordinates transformation and image interpolation are separated, which replaces the traditional warping operation by row and column gradually. This operation reduces the complexity and saves computing time. With the new method a good warping result has been acquired. This method has been applied to human eyes morphing operation and satisfied result has been achieved. According to the high requirements of keeping regular shape and edge information in eyes morphing, this image warping method proves applicable.2979|416|0更新时间：2024-05-07
- 摘要：The visual detection of bullet holes in a target usually can be formulated as a problem of how to exactly locate in an grayscale image the isolated and overlapping spots, which are treated as image structures without inner detail and texture. In this paper a model based matching algorithm is used to deal with this problem. A non linear model is used to describe gray and position information of those spots in a grayscale background and overlapped spots can be seen as single spots with intersection. Then the model matching process is applied to search and locate isolated and overlapping spots with various sizes, gradients and backgrounds. When we use this algorithm for holes detection and recognition in a military target, according to the characteristic of the time axis distribution of bullet holes, a feed back method is used for better performance. Experiments show it works well and achieves high reliability in real environments.2808|340|0更新时间：2024-05-07
- 摘要：This paper introduced the application of the microscope digital image processing in the modify analysis of blood cell. We deal with the blood cell by the method of digital image processing, and have also thought over the processing step of concerned pattern recognition, so that, we can use the computer voluntarily to judging and distinguishes for the microscope of cell. The viewpoint of artificial intelligence has still been thought over in this paper, for that, the computer have the character of learning, this makes that its recognition result possesses reliability. After we pro processed the digital image of blood cell, by comparing between the round degree, rectangle degree and center quadrature of the geometrical modality, we can get the eigenvectors of the different blood cell, we have discussed their reliability. We calculated the above geometry characteristic from the 56 examples cell samples (every sample includes three kinds of blood cells, and counts 168 images), and by standardization based the standard difference, we computed all values of characteristic weight, finally we got the characteristic vector of three kinds "classification", that are X1,X2 and X3. Using 2 norm X i-X 2 we reated the distinguish function. By train computer to remember the cell characteristic of new kind of cell, we have pledged to differentiate correctly the known classification, and have pledged to distinguishing correctly the new classification.关键词：Microscope image;Digital image procestion;Blood cell;pattern recognition;Modality analysis;Anesthesia2359|327|0更新时间：2024-05-07
- 摘要：As very large collections of images are becoming common, there is a growing interest in image database that can be queried based on image content. Content based image retrieval(CBIR) has become an important research issue for image database. As color histogram is simple to compute yet effective as a feature in detecting image to image similarity, it is an image feature widely used in CBIR. However, using the classical color histogram for indexing has a number of drawbacks. E.g. it does not contain information about the spatial locations or distributions of pixels in an image. To overcome its limitations, the refined histograms techniques are proposed on the basis of joint distribution of color and other features. In this paper, in order to provide a more accurate description of the image content, we propose two histogram models to refine the description of each pixel by some local features. One model is presented to integrate both color distribution and detail signal energy into a single histogram. Another is presented to integrate the edge strength into the definition of the color histogram. The experimental results show our histogram approaches can achieve an increased discriminative power compared to the classical color histogram technique for image retrieval.2577|366|0更新时间：2024-05-07
- 摘要：In this paper, we discuss several problems on fragile watermarking, which is used to ensure the credibility of multimedia. Firstly, we describe some main properties that an efficient fragile watermarking scheme should have. And then we discuss the differences between robust watermarking and fragile watermarking algorithms, which are important for us to develop efficient fragile watermarking scheme. Thirdly, a simple application framework for fragile watermarking, which is called trustworthy digital camera, is proposed. This fragile watermarking framework can be used for ensure facticity for images taken by digital camera. In the end, we present two approaches, pre processing and using integer transform, to solve round off error and overflow error problems for frequency domain fragile watermarking techniques. Experimental results show that these two approaches are efficient for improving other fragile watermarking algorithms.2510|342|0更新时间：2024-05-07
- 摘要：This paper is devoted to seek a new image hiding technology, which only scrambles the detail of images and keeps invariable in the mass, instead of arousing the desire to decrypt the image. Firstly, the concept of knight tour problem and knight tour matrix are introduced. A new scramble transformation is presented, which is called a knight tour transformation(KTT). The principle that KTT can hide the details of images is analyzed. Secondly, several experimental results involving details hiding of images are given. Several edges detection experiments are also given to illustrate the usefulness of our method in hiding images' details. The experiments show that the image is approximately invariable in collectivity and it seems to be polluted by some noises. Therefore, the scrambled image does not easily stimulate other people's desire of trying to decrypt it. In addition, the transformation also can hide text information into image. By means of this the security of image and the information hiding into it are increased. This is propitious to transmit security of image details and confidential texts. Finally, it shows that the security is high and the immunity is strong by analyzing KTT 's characteristics on image hiding and disguising.2914|394|0更新时间：2024-05-07
- 摘要：Managing geospatial information is an important part of geographical information system(GIS). Database system can help GIS designers separate data management task from the huge systems. However, different database systems, such as relational database(RDB), object relational database(ORDB) and object oriented database(OODB), have their native advantages and pitfalls. This paper firstly discussed why database system but not file system is chosen to manage geospatial information, and now there are three major kinds of database systems can be used to manage geospatial data. Then, three typical models--pure relational model, spatial extension model and wrapped model, are introduced, analyzed and compared. The key implementation issues and technologies of wrapped model, which is the most competitive solution, are discussed in details. As a result, ORDB is thought as a more practical way due to current databased technologies, and OODB still remains in prototype step.3136|387|0更新时间：2024-05-07
- 摘要：With the development of remote sensing technology, it has been applied in more and more domains.\nTherefore, it is necessary to efficiently compress remote sensing data because of the rapid increasing of remote sensing data. In this paper the algorithm includes three parts: first, based on the spatial and inter-spectral correlation of multi-spectral remote sensing images, a 3D prediction model is set up, then the context prediction is made; finally, the re-prediction is used to remove the statistic correlation in the images, and it results in the most decreasing of entropy of the residue. The experimental results represent that the put forward algorithm is much more efficient than the Best JPEG, in addition, it is more efficient for AVHRR images than Spatial &Spectral Optimal Linear Prediction (SSOLP).2303|337|0更新时间：2024-05-07
- 摘要：Lifting scheme is honored as the key tool to the construction of second generation wavelets. As for the first generation wavelets, lifting scheme is a new way of implementation faster, simpler and easier than the commonly used Mallat algorithm. Lifting scheme is also the one recommended by JPEG2000, a new image coding international standard scheduled to be formally published this year. In order to apply the lifting scheme to wavelet image coding, we make some improvement to the lifting scheme. We first implement the D9/7 bi orthogonal wavelet transform through lifting scheme and then apply the transform to wavelet image coding. In doing so, we improve the procedure of implementation to make sure that each lifting step is FIR filtering. At the same time, we adjust the value of scale factor under the guide of energy conservation. The scale factor obtained this way is different from that recommended by JPEG2000. The experimental results presented in this paper show that our improved lifting scheme has achieved the image coding gain better than that achieved by Mallat algorithm.2691|338|0更新时间：2024-05-07
- 摘要：After quantization, a lot of image's wavelet coefficients will become zero valued. How to encode the location of these zero valued coefficients with as few symbols as possible is the key to increase the compression ratio of image coding. Shapiro presented the so called zerotree structure to encode these zero valued coefficients, in which the root of a zerotree is the only symbol to be encoded. One can obtain from the root's location the locations of all zero valued coefficients within the zerotree. Shapiro's zerotree coding has been accepted by JPEG2000, a new image coding international standard. In this paper we present an improvement to Shapiro's zerotree coding. We find in practice that after quantization and zerotree coding of image's wavelet coefficients, the roots of zerotrees will appear consecutively if they are scanned following a certain order. Thus, we add the runlength coding of consecutive roots of zerotrees before entropy coding to increase the compression ratio of image coding. Theoretical analysis and experimental results presented in this paper show that our algorithm improves the performance of Shapiro's zerotree coding.2215|390|0更新时间：2024-05-07
- 摘要：Motion estimation is a very important part in the inter prediction technical of video coding, and the performance of a video coder largely depends on the efficiency and precision of motion estimation. The full search algorithm is seldom adopted because of its enormous complexity of computation, and now the fast algorithms of motion estimation such as three step algorithm and cross search algorithm are used widely. But the matching precision of these algorithms is low and in some cases they can not produce a good result. This paper first discusses the spatial correlation and time continuity of the motion object in video coding. Then, a new fast motion estimation algorithm based on the spatial correlation and time continuity of the motion object is presented. The average search times for computing motion vectors per macroblock of this algorithm are less than those of the three step algorithm and the matching precision of this algorithm is very close to that of the full search algorithm by experiments. Moreover, the output bits of the coder using this algorithm are less than those of the coder using the full search algorithm.2397|302|0更新时间：2024-05-07
- 摘要：Wavelet transform based image fusion method have been used broadly these years. In this article, in order to adequately make use of all kinds of remote sensing images' information, it puts forward a new energy based wavelet fusion method which also uses image restoration and enhancement model to merge the images of CBERS 1(China Brazi1 Earth Resource Satellite) and SPOT PAN(Systeme Pour 1'Observation de la Terre Panchromatic) image. Image restoration technique can both increase image acutance and improve image quality. By processing CBERS 1 imgae with image restoration technique, we can not only overcome the blurred image disadvantage, but also improve the fusion result image's quality finally. Compared with the result of IHS?PCA and HPF traditional fusion methods, it is proved that this method can get a perfect result for the application aim of "enhancing the result image's spatial resolution with preserving spectral information and effectively overcome the blurred disadvantage of CBERS 1 image".2545|345|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