最新刊期
卷 8 ， 期 11 ， 2003
- 摘要：Research on cardiac MR images analysis is a primary topic in the field of medical image for many years. The tagged MR imaging technique proposed at the end of 80's, last century, it was very attractive in that the tag line and tag plane provided a particular approach to track the motion of the left ventricle. It is a hot topic in the field of medical image at present to analyze the tagged cardiac MR images. In this paper, a brief introduction to tag-based motion tracking technique is presented at first, and sequentially, the existing methods for tagged image segmentation, including the epicardium and endocardium segmentation of left ventricle and tag stripes tracking, are inspected, as for shape recovery and motion reconstruction of left ventricle (LV), five approaches are discussed with comparison and summary of their merits and demerits, these methods are deformable model, stochastic model, B-spline based model, harmonic phase method and optical flow method. Another aspect of this topic is the strain analysis of LV which provides intuitive quantities for physicians. Here, some important results are also presented. In the last section of this paper, several problems in the existing methods are discussed and they are helpful for further work.关键词：Medical imaging;Cardiac magnetic resonance imaging;Tag line;Image segmentation;Shape recovery;Motion reconstruction;Strain analysis;left ventricle2866|190|0更新时间：2024-05-08
- 摘要：Image enhancement techniques are used to process an image so that the final results are more suitable than the original image for human perceptual. Based on the analysis of color space and color components, we present a novel algorithm based on wavelet decomposition for color image enhancement. The perceptual color space HSV (Hue-Saturation-Value) is chosen according to human visual properties, in this color space, a hybrid contrast enhancement algorithm for color image, which used to adjust the luminance of image adaptively, in step of what follows, the saturation component of the image is stretching too. Color image enhancement exists two goals: image naturally and looking vividly as well as the distinguished details. Compared with other enhancement methods, experimental results confirm that our method improves color image quality from details as well as color components.2791|288|0更新时间：2024-05-08
- 摘要：One of the main tasks in multi-scale edge detection is to seek the best compromising between removing noise and remaining fine edges. At present, many adaptive multi-scale edge detection algorithms have been developed. But there is one problem in these methods in that some detected edge points are moved actually from their exact positions. In order to obtain edge points as exactly as possible, in this paper, a new adaptive multi-scale edge detection method is developed, in which the edge positions are kept invariant to the most in the case of large scale. Moreover, firstly, it is proved that, within an apt scale range, with a special class of wavelet basis, the positions of edge points based on zero crossing of two order derivatives won't be changed after wavelet transform even with ordinary edge detection operator; secondly, according to the property of multi-scale analysis and the relation between differentiating and integration operation in which a wavelet function is taken as a kernel function, a multi-scale self-adaptive multi-scale edge detection algorithm was put forward in which the local maximum scale in that the positions of local edge points won't be changed is developed; finally, two group of experiments are carried out with different kinds of wavelet basis. The experiments show that, under the restriction that the specific edges should be kept as good as possible, the positions of edge will not be changed in large-scale case.2394|158|0更新时间：2024-05-08
- 摘要：A new method based on two dimensional histogram and block clustering is proposed in this paper, in order to extract bubbles in the aerated water flows images. Firstly, the original image was divided into sub-images with different sizes in this method, to solve the uneven illumination of the image, which was result from the high intensity of bubbles defilade the low intensity bubbles in the image, and then the binary image of the sub-images were obtained. Secondly, block clustering is used to recognize the pure background blocks and dispose them in a special way. Thirdly, the bubble information is analyzed based on the definition of an evaluation function and an evaluation block in the binary image. Finally, a statistical characteristic analysis method is used to compensatethe overlapping bubbles, which were appeared in the original image, and it is also impossible to get high-accurate bubbles size by the method of image segmentation, and at the same time, optimal quantificational estimation of the bubble area square distribution was obtained. Many experimental results show that the method is efficient.2240|196|0更新时间：2024-05-08
- 摘要：In this paper a novel image mosaicking approach is proposed, which includes three steps. Firstly multi-scale wavelet decompositions are performed on each image segment that is partitioned from large reference image and each tile image in image library respectively. Then standard deviations of image wavelet decomposition coefficients and similarity distance between image segments and tile images are calculated in each level, which latterly are used as a criterion for tile image selection that images in library best matched to the mosaic area in reference image are selected. Finally to make the color of mosaic image identical to that of reference image, color correction on each pixel of mosaic image is employed. Due to human visual character that what human perceives is only the whole intensity of observed region by rounding and meaning of the intensity values in it, the final work of mosaicing has realized the effect that the tile images only can be discern nearly and the overall figure looks similar to that of the reference image from a distant view. An automatic computer mosaic image generation routine using above algorithm has been implemented, which shows the feasibility of this approach.2608|175|0更新时间：2024-05-08
- 摘要：In order to improve the effectiveness of Snake active model, the deformable template approach of contour-tracking algorithm, in color images, this paper mainly focuses on resolving two problems. First, to reduce the difficulty of handcrafting initial contour for Snake, the authors introduce the pretreatment procedure. They first use color-based clustering algorithm to divide an image into several regions, then use improved graph-searching algorithm to get the contour of independent regions which may be important for further recognition, and take this result as the initial contour. Second, to avoid the distortion contour produced by Snake model when applying it to tracking the contour in some color images, they redesign the external energy function of Snake. They use neighborhood pixels' Euclidean distance of the colors in weighted HSI color space, instead of grey level difference in traditional algorithms, to approximate the grads of image, and take the square of the distance as external energy function to keep the active contour staying at the place where the real contour locates. At last, comparing prototype of Snake model with the new method, authors illustrate that the improvement is considerable, especially when color images are involved.关键词：Computer Image Processing;Snake model;Color-based clustering;Contour-tracking algorithm;Euclidean distance;Weighted HSI color space2555|218|0更新时间：2024-05-08
- 摘要：Automatic center-path finding in real 3-D data set is the key problem in realization of automatic navigation. Many methods have been proposed, but most of them are either quality depressed or time expensive. In this paper, after summarize the main existent methods, a new 3D center path finding algorithm base on two distance fields is proposed: Between any given start-point and end-point which can be connected, the source distance field base on the start-point and the boundary distance field base on the boundary are established. Through the co-restriction of the two-distance field, a center path that connects the start-point and end-pint can be found. In order to enhance the effect of navigation, a cubic B-spline curve is used to smooth the path. The algorithm is realized on PC platform and two medicine image date sets are used for test. Computer time and path finding results are shown in Section 4. As compared with onion peeling algorithm and Dijkstra's Single Source Shortest Path Algorithm, the result shows that the algorithm not only can get fast speed and high quality result, but also has more flexibility. The application in Virtual endoscopy is also shown in the end of the paper.关键词：3D center path;Dist field;Automatic navigation2508|188|0更新时间：2024-05-08
- 摘要：In view of industrial spot and the particular requirements of the segmentation for the end image when the bar steel were bundled up, a new image segmentation approach is presented. At first. the image is preliminary segmented with threshold method and edge based method, by using the fast 2-D entropic thresholding algorithm and the simple gradient operator respectively. Then the results from the two methods are fused and integrated, the target areas are identified and the adhesions are removed, in which genetic algorithm is implemented. The fusion-based technique is shown to perform very well in terms of robustness to noise and adhesions, and the calculating time is not evidently added. The experimental result shows that the proposed approach is efficient in improving the quality of segmentation and fitting the industrial requirement.2415|185|0更新时间：2024-05-08
- 摘要：This paper proposes a wavelet-domain Least-square (LS) based algorithm for image superresolution. Beginning with presenting the edge models popularly accepted in the literature, it is demonstrated in this paper that the edges in different scales are similar to each other in form. This property is called the self-similarity of the multiscale edges. Due to the property, it is possible to predict the three subbands of wavelet coefficients. In order to guarantee the stability and effectiveness of the prediction, the least-square method is adopted. The wavelet coefficients obtained so far are not correct where the multiscale edges are not self-similar. So, the correlation correction method is used to reduce the kind of distortion. Once the wavelet coefficients are obtained, the high resolution image can be reconstructed. Because the algorithm properly preserves the geometrical regularity around the edges, the induced image is of high visual quality. Besides, since only the wavelet coefficients near edges need to be predicted, the algorithm is computationally efficient. Simulations demonstrate the performance of the method.2086|158|0更新时间：2024-05-08
- 摘要：It is well known that edge is the basic feature of image and the important property for image analysis and recognition. Using traditional edge extraction, strong edge can be efficiently extracted but detail edge information may be lost, however, these detail edge information are often important features in some real applications. This paper proposes a new method to extract image detail edge based on the combination of gray-morphology and image decomposition. First gray-morphological operators are used to detect edge image and remove part background and noise, then it is decomposed into several areas using quad-tree method, continuous decomposition is terminated when the area size is equal or smaller than the minimal area size parameter, finally, different thresholds for different areas are selected to ensure the integrality of edge extraction, in order to void smooth background are involved and some detail edges are lost, global minimal and maximal thresholds are set beforehand to limit the scope of selected threshold, when area threshold is smaller than global minimal threshold, it will be replaced by the global minimal threshold, inversely, by global maximal threshold. Simulations show that this method can efficiently extract detail edges from both noiseless images and noise images.2390|183|0更新时间：2024-05-08
- 摘要：Without digital data, how to fully make use of the old optical analogue seismic sections and films and to renewedly constrast, reinterpret, especially to reprocess and restore them (including those on which have been manually interpreted), how to solve the problem that the hard copied seismic sections can only be read but not be further reprocessed, an effective approach has been put forward in this paper. Which makes it possible to reprocess and reinterpret old seismic sections and films. According to the graphical image processing technology, the seismic waveforms on the rasterized images scanned from paper seismic sections can be refined, identified and extracted and be converted into digital data, then be reformatted as the standard SEG-Y format, which can be used for being further processed and interpreted. Analyzing the results of applying this approach to old paper seismic sections, we realize that in the digitalized seismic sections their original characteristics can be effectively retrieved. Therefore this approach is feasible to reconstruct old paper seismic sections.2882|181|0更新时间：2024-05-08
- 摘要：There are two mainly error rules for the generation of multi-resolution model of terrain, one is screen space error, the other is terrain roughness, however, both of them are liable to reduce the area near or not very far from the viewer, because they discard 2 of 3 dimensions of a vertex. In this paper, first, a rule named remaining energy is proposed which is suitable for reducing the area far or very far from the viewer, the principium of the rule is that the light eradiated from surface will attenuate during its transmission, so the light energy eradiate from the surfaces far from the viewer will become too faintness to attract the viewer's attention, thus the surfaces far from the viewer can be emerged into larger surface, then the rule based on vision principle is formed by combining remaining energy rule and terrain roughness, in the end, an algorithm and its implementation based on quadtree for generating terrain multi-resolution model under the vision-dependent rule is described in detail, the algorithm can simplify the terrain effectively and is proven to be very valid for the reduction of large scale terrain at the same time.2618|194|0更新时间：2024-05-08
- 摘要：With a consideration of the influence of consecutive multiple bands of images with a low signal to noise (S/N) ratio obtained from imaging spectrometer on the reconstruction of reflectance spectrum, an amending method is developed from the digital number (DN) value of trend-face of low S/N ratio band. On the basis of properly estimated the trend-face of DN average values, the highly correlated signal variances are introduced from adjacent band to amending band under the constraint condition of energy ratio of DN in this method. As a result, not only the influence of original noise band on the reflectance spectrum reconstruction is eliminated accordingly, but also the quality of composed images generated from low-ratio S/N band is improved by this amending method. The results show that the application aim of amending defect bands and restoring spectrum characteristics is realized with the new concept of lower-passfilter.2683|196|0更新时间：2024-05-08
- 摘要：Crown cover fraction is an important ecological and environmental parameter. It's very important for the study of global environment change and observation. How to estimate crown cover fraction from remote sensing data and how to improve its precision are key work for the developing of global or regional climatic and ecological models. In the study, the authors developed a new method to estimate crown cover fraction according to the characters of soil and vegetation spectra and the definition of crown cover fraction. Furthermore, they also put forward a more simple method on that basis, which is called maximal gradient difference method. Once the crown cover fraction is known, the background spectra for pixels that are partly covered by vegetation can be rebuilt easily under the assumption of linear spectra mixing. Using the above method, the authors have successfully retrieved background information of a TM image of Colorado State. The introduced methods are simple, easy to performance, and can separate the effect of soil and vegetation effectively, so it's possible for them to replace the method of estimating crown cover fraction from NDVI, which is used widely at present. The authors also pointed out that if noise is removed from data before performing the method, the result will be much perfect.3472|262|0更新时间：2024-05-08
- 摘要：This paper discusses the reason of triangle metamorphosis in Marching Cube Algorithm.According to there as on wegivea method to protect the metamorphosis.This paper also diseusses a new data structure-Sliee-objeet-based(SOB) strueture，which advances the algorithm.The experiments show that SOB method takes the same rendering time when there is only a objeet in dataset.But if there is more than one objeet，octree alogrithm takes few times rendering time than SOB algorithm，so if the datasetdoes not contain so many voxels，it is meaningful that less iso-surface rendering time is taken by the lost of the approximate memory spaee.That is to say SOB algorithm excel so ther methods when the dataset contains more than one object，e.g.octree algorithm an drun-length encoding algorithm，in the executing efficiency.When the dataset contains more than one object，octree algorithm only use one octree data strueture for all objects，so when we want to render one or several of the objects，this algorithm need to create the octree strueture ain. SOB algorithm is not so，it creates the SOB structure for each of the objects from the beginning of rendering，so we can render any one of objects selectively according to user's interest，so SOB algorithm 15 adapt to interactive rendering.2755|187|0更新时间：2024-05-08
- 摘要：Theory and method in digital communication have been applied effectively to the embedding and detection of digital watermark. The method is efficient to embed the watermark bits into the digital media by the principle of CDMA communication due to its many advantage, such as, multiple-accessing, excellent secrecy, standing firmly against disturbing, and resisting noise attacks etc. In this paper, a DWT-based image watermarking algorithm applying DS-CDMA is proposed. The orthogonal Gold sequences sets generated via the preferred-pair of m sequences are well improved, and then are used to spread spectrum of the watermark. The CDMA encoded watermark is embedded into the DWT domain by modulating selected DWT coefficients of the image in a fashion similar to CDMA. Blind recovery of the embedded data is achieved by analyzing the DWT coefficients of the watermarked image and the auto-correlation of orthogonal codes. The proposed algorithm is compared with the existing watermarking schemes based on CDMA. The experiment results demonstrate that the watermark with the proposed algorithm is robust to JPEG compression, Gaussian noise and other typical attacks. The hidden information is secure well and hard to resist by unauthorized detection and attacks because of the same private key.2910|180|0更新时间：2024-05-08
- 摘要：The management and retrieval of Image/Video database based on their content has became an important search area in recent years. MPEG-7 standard provides a rich set of tools for completely describing multimedia content. In this paper, we explore the issue on content-based video manage and retrieval system. A video database described by MPEG-7 video content description and MPEG-7 semantic description is realized. Within this system, natural language comprehension technique is incorporated which makes it possible for user to query conveniently by more natural language, and synonymy lexicon is used to reduce the redundancy of features. We combine Low-level visual feature and high-level semantic description together, and adopt relevance feedback mechanism which enables user to find his target video step by step. Also we use semiautomatic intelligent weight innovation algorithm , which can make the features more and more reasonable. Experiment result shows that we can manage and retrieval video by content expediently and effectively using this system.2360|173|0更新时间：2024-05-08
- 摘要：Because image database is very huge, the feature extraction and retrieval process are usually time consuming. In order to effectively use existing text information retrieval methods in content based image retrieval, especially the index mechanism of the product tf * idf by term frequency (tf) and inverse document frequency (idf) for each text document, this paper cooperates tf * idf model with triangle tree to improve the retrieval performance. First, after pixel-based histogram features of sub-block in certain image class are mapped to color concept space through independent component analysis (ICA), we would obtain all of independent keyblock of such image class; then well-trained fuzzy support vector machine is used to recognize all of independent keyblocks contained by each image. Similar to text retrieval, in which the whole text document is indexed by , the recognized independent keyblock is used to index each image in database. Because independent component features are naturally high order independent with each other, compared to principle component analysis (PCA) method, this algorithm achieves higher performance. At last, triangle tree is used to hierachically index image database and thereof speed up retrieval.2111|167|0更新时间：2024-05-08
- 摘要：Video caption could be used to index video stream with high-level semantics since it implied lots of semantics inherently. The prior work of caption location and extraction considers how to define good caption features and neglects the self-generalization of classifier machine thereof. In order to overcome this limitation, an algorithm firstly localization and extraction video caption using support vector machine (SVM) and independent component analysis (ICA) is presented. In this algorithm, the raw video frame is segmented into N * N sub-blocks, and each block is identified either a caption block or a non-caption block; then mutually high-order independent ICA features are used to train a support vector machine classifier; finally the location and extraction of video caption can be finished automatically with pyramid model and de-noising techniques by each trained support vector machine classifier. Because support vector machine holds excellent generalization of classification with non-enough samples and independent component features are naturally high order independent each other, compared to other algorithms, the experiment data shows this method works well.3053|193|0更新时间：2024-05-08
- 摘要：Video object segmentation has important application in content-based video encoding and video retrieval. In this paper, a new approach based on temporal and spatial union information is proposed. In spatial domain, morphological operators such as morphological filters, morphological gradient operator and watershed algorithm are used, morphological filters are used to simplify the image, morphological gradient is used to mark homogeneous regions, the watershed operator is used to decide spatial boundaries. In temporal segmentation domain, the initial change detection mask is gotten by using F-test, then by combining it with the result of spatial segmentation, the final result is gotten. The procedure toward complete segmentation consists of the following main steps; global motion estimation and compensation (if exist background motion), change detection, marker extraction, watershed transfer, temporal and spatio information fusion. The experimental result shows the effectiveness of the method: it can separate the foreground from still or motion, simple or complex background satisfactorily, preferably locate and segment the video object.2382|176|0更新时间：2024-05-08
- 摘要：The video coding standard MPEG-4 enables content-based functionality for various applications. To support this functionality and improve coding efficiency, MPEG-4 decomposes each frame of a video sequence into video object planes (VOPs). The technique is also used in content-based video retrieval and video monitoring system to improve intelligence degree; and its applications are extended to pattern recognition, computer vision et al. Thus, segmentation and tracking of video moving objects become a key related research topic. A novel algorithm for automatic segmentation and tracking of moving objects is proposed in this paper. On the base of different image, we utilize a method of combining Canny edge detection and fourth-orders statistic hypothesis testing in improving the moving areas separation results. Then we use region growing with color information in the moving areas to achieve precise boundary of moving objects; and make use of the last segmentation result, extract edge and color information of moving objects, apply edge model matching and color model matching approach to track the changes of moving objects, and detect temporarily still parts. The experimental results for several video sequences demonstrate that our approach is effective in precise segmentation and detect the temporarily still parts of the video moving objects.2947|174|0更新时间：2024-05-08
- 摘要：Because of buffer overflow or delay timeout, packet loss is very common in current internet data transmission. The situation is even worse for real time internet video communications. Due to the prediction property of most state of the art video coding methods, data loss in a single frame can also deteriorate several successive frames. Intra update is well known as an effective method to reduce this error propagation. In this paper, we propose a novel adaptive intra update method. Taking decoder error concealment into account, the distortion at the decoder can be precisely estimated at the encoder by using the Recursive Optimal Per-pixel Estimate (ROPE). Then, image regions with maximum distortion are intra updated. Thus the damaged image regions at the decoder can be recovered in time. Simulation results demonstrate that our method achieves substantial and consistent gains in PSNR(0. 3~2. 5dB) over the non-adaptive methods such as raster-order intra update and random intra update for different video coding rates and packet loss rates. The complexity of this method is also modest, and it can be easily incorporated into current video communication system while maintaining compatibility.2450|180|0更新时间：2024-05-08
- 摘要：Video projection is in widespread use for multimedia presentations in classrooms and in conferences. Unfortunately, the usual human-computer interaction (HCI) with keyboard and mouse confines the speaker to the computer devices. In this paper, we present a system for remote HCI using a laser pointer. In addition to the projection equipments, the system only requires a video camera and a video capture card. We use the video camera to capture the projection plane, detect the laser point (that is, the region induced by the laser pointer) in the captured frames. We then track and recognize the trajectory of the laser point and translate it into messages sent to the computer system. Compared with the previous work in this area, we develop some novel ideas in the system: first, a training process is introduced to improve the system adaptability; second, multiple cues such as color appearance, motion, and shape, are integrated in laser point detection to improve the system robustness. The experiments demonstrate the usability (robustness) and adaptability of our system.2838|189|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