最新刊期
卷 14 ， 期 7 ， 2009
- 摘要：Travel information services develop quickly all around the world with increasing public travel demands. This paper introduced the state of the arts of the travel information services industry, proposed the system structure of the technical platforms involved in travel information services, analyzed the roles and trends of communication, positioning and mapping platforms, reviewed the research progress of concerned technologies including traffic information collecting and processing, spatial data modeling, path searching algorithms, short-term traffic forecasting, information visualization and service standards and criteria,and analyzed the scientific problems and technological bottlenecks encountered in developing travel information services industry.关键词：travel information services;traffic information;state of the art5216|291|0更新时间：2024-05-07
- 摘要：The floating car technique is one of the key technologies in ITS (intelligent transportation system)to acquire the traffic information in recent years. In this paper, firstly, the fundamental principle of floating car technique is introduced. Then, the enhanced processing procedures of floating car information system, and some important approaches including map matching, path deriving and traffic information calculating are emphasized. To review the key issues affecting information processing accuracy, such as poor GPS data precision and complex road network, a serious of enhanced approaches are presented. Finally, some experiments are given to evaluate information processing accuracy and future work is pointed out.关键词：floating car;information processing;map matching;path deriving;information fusion6113|286|0更新时间：2024-05-07
- 摘要：With the tremendous development of computer technology and Geoinfomatics, more people are taking vehicle navigation system (VNS)as one of their travelling assistant tools when auto-dependant lifestyle has become more popular worldwide. Digital navigation map is fundamental part for VNS, so map data actuality will directly exert an influence on navigation accuracy. We find that data updating mode just depending on CD or intelligent memory card is a big bottleneck for data updating from present navigation map data production method and procedure, management mode and updating approach in navigation terminal. Therefore, this paper is based on on-demand dynamic data updating in vehicle navigation application and will study navigation data model with the support of updating, change detection approach for incremental navigation data between different data versions and dynamic data updating modest. An experiment system is designed to prove related algorithms and procedure. It will be an important innovative research for map production and efficient use in navigation industry, which can also extend navigation information services, mitigate urban congestion and provide technology for intelligent transportation system.关键词：incremental updating;dynamic update;change detection;vehicle navigation;version management6297|260|0更新时间：2024-05-07
- 摘要：The advances in positioning and communication technologies make it possible to collect and transfer trajectories of moving vehicles. In addition to some basic applications such as navigation, traffic data can be extracted from vehicle trajectories, such as average moving velocity and travel time on a road segment in a given time period. The network and spatio-temporal characteristics of trajectories, however, make it difficult to evaluate queries applied to them. As a result, many studies have focused on developing index structures and methods for trajectory data. However, few of them have presented the corresponding data query language. In this paper, we discussed some issues related to trajectory data query language, including the definition of the connectivity-oriented data query language (CQL), the classification of CQL, and its implementation based on an index structure proposed by previous study. We further discussed a prototype for vehicle trajectory data query based on CQL in terms of its structure and implementation.关键词：trajectory data;index structure;data query language;connectivity6561|272|0更新时间：2024-05-07
- 摘要：Structural patterns of road networks play a vital role in the simplification, multiple representations of road networks. This paper proposes an automated method for identifying grid-like patterns from road networks. The proposed method firstly identifies grid-like patterns according to the shape similarity, and the direction similarity of road networks.Then it clusters grid-like patterns based on a region incremental algorithm. Experiments show that the proposed method is capable of automatically identifying grid-like patterns from road networks.关键词：pattern recognition;road network;grid structure recognition5655|241|0更新时间：2024-05-07
- 摘要：As one of a serial of reports, this paper is a survey on multimedia researches and applications in China, 2008. Since multimedia is a cross research area, papers about multimedia technology are distributed on various journals. We checked 2 918 papers published on 9 Chinese journals in 2008, from which we have selected 382 related to multimedia technology and application. Then we conducted analysis on all the selected papers and for those from 2004 to 2007 statistical analyses were conducted. The categories are the same as last year. Looking into the data, we can find that coding standards, digital watermarking, virtual reality, multimedia data retrieval, multicast, streaming media, interactive mode and interface, identification using biometrics earn high attention of researchers in China. Meanwhile, multimedia application systems are diversified and intergraded with daily life gradually. We present here an overview on the progress in multimedia technology in China, 2008. This article can be used by researchers as a thoroughl reference, and also will be helpful for people in technical planning and management.关键词：multimedia;coding standards;digital watermarking;virtual reality;multimedia data retrieval;multicast;streaming media;interactive mode and interface;biometrics3639|301|0更新时间：2024-05-07
- 摘要：The sparse representation of image is the key to the image coding technique in transform domain. In this paper, based on the analysis of the limitation of ridgelet function, a new construction method of multiscale ridgelet dictionary is firstly proposed. The speed of search of the most matching atoms is improved based on the tree structured organization of atoms. And then a tree structure multiscale ridgelet dictionary based matching pursuit image coding scheme is proposed. In the last the decomposed coefficients are adaptively quantized and encoded based on the distribution of coefficients and the rate-distortion analysis. Experiment results show that the new multiscale ridgelet dictionary can represent the images sparsely. On the other hand, the performances of the new method are shown to compare favorably against those of JPEG2000 scheme, especially at low bit rate.关键词：image coding;multiscale ridgelet dictionary;sparse representation;matching pursuit3397|296|0更新时间：2024-05-07
- 摘要：Compared to the traditional threshold stack filters, mirrored threshold stack filters have been empowered not only with lowpass filtering characteristics but with bandpass and highpass characteristics as well, but their positive Boolean functions length leads to an increasing restriction during calculation. In order to solve the above problems, this paper proposes a mirrored adaptive weight (MAW) algorithm, which calculates cost vector based on adaptive neighbor weight error criterion (ANWMAE). After cost vector stacking is restricted and its astringency is estimated, the optimal positive Boolean function of stack filters is confirmed to construct adaptive weight mirrored threshold stack filters (AWMSF). In order to testify the filtering capability, AWMSF was simulated, the results show that it can suppress noise and protect the details of image effectively, the number of iterations is reduced and the computing complication is decreased rapidly too.关键词：mirrored threshold decomposition;stack filters;positive Boolean function;adaptive weight;image processing3688|299|0更新时间：2024-05-07
- 摘要：In standard wavelet methods, the empirical wavelet coefficients are thresholded term by term, on the basis of their individual magnitudes. Information on other coefficients between the different scales and in the same scale has no influence on the treatment of particular coefficients, resulting in the lower accuracy of signal estimation. A translation-invariant (TI)neighboring-coefficients thresholding is designed by incorporating the different evolution of signal and noise along the scales of wavelet domain and information in the same scale. Considering the particularity of noise in magnetic resonance (MR)images, a novelty MR complex denoising algorithm based on TI neighboring-coefficients thesholding is developed by employing the complex entity method in MR complex images. The results of the simulated experiments show that the proposed algorithm has the higher accuracy of signal estimation, and outperforms previous MRI denoising methods about denoising capability.关键词：magnetic resonance;images;denoising;translation-invariant;neighboring-coefficients thresholding;complex entity3228|268|0更新时间：2024-05-07
- 摘要：In order to improve the denoising effect, a bivariate statistical modeling for dual-tree wavelet coefficient was proposed. This new denoising method used a parametric bivariate generalized Gaussian distribution (GGD)to describe the statistical distribution for Dual-tree complex wavelet coefficients of images. Then, based on maximum likelihood estimate (MLE), we can obtain the estimated parameters for GGD. With the estimated parameters, maximum a posteriori (MAP)estimator can be used to restore the wavelet coefficients from the noisy observations. Results of our experiments show that image noise can be reduced effectively while most image details can be kept. The proposed method outperforms many denoising algorithms both statistically and visually.关键词：image denoising;bivariate statistical model;wavelet transform3614|240|0更新时间：2024-05-07
- 摘要：We propose a new watermarking method in the time domain, which utilizes a local set of information of each vertex to embed watermarking information and blind-detect. During the embedding process, appropriate vertices are selected to be embedded with the watermarking information according to local set area of each vertex on purpose of improving imperceptivity. As the detection progresses, the same local set area information is obtained to build the synchronization between embedding and detecting. This improves the robustness of the proposed. The property of affine invariant of first order moment makes the algorithm resisting attacks of rotation, translation, uniform scaling easily. The experiment has shown that, while the attacks are not so strong to damage the visual quality of 3D object, the algorithm acquires high robustness to resist attacks of adding noise and smooth-filtering.关键词：mesh;watermarking;local set;blind-detect3236|258|0更新时间：2024-05-07
- 摘要：In order to better protect the copyright of the color image, this paper explained theoretically why the green color image component should be chosen to embed into a watermark, and thus proposed a new blind watermark algorithm for a color image. First, the green component of the original image was divided block by block with size 8×8 pixels, and transformed with discrete cosine transformation (DCT). Second, the binary watermark image was spread spectrum modulated by two uncorrelated random sequences. At last, in order to maximize the enhanced watermark strength without affecting transparency, human visual masking factor was imported. And the modulated watermark with different intensities was embedded in mid-frequency region coefficients of DCT domain based on visual masking factor magnitude. The watermark was extracted by comparing the relativity between two random sequences with watermarked image, and the process did not need the original image, which means a blind watermark algorithm. Experimental results demonstrated that this method was invisible and robust against some usual attacks such as JPEG compression, cropping, adding noise, and embedding watermark in a green component is more robust against lossy compression than embedding watermark in a red or blue component.关键词：visual masking;spread spectrum modulation;green component;blind watermark2916|247|0更新时间：2024-05-07
- 摘要：Marching cubes (MC) algorithm is a typical representative for surface reconstruction method, however, there are still some shortcomings about the topology structure and the represent accuracy of the isosurface which extracted by MC algorithm, and so does the algorithm’s execution efficiency. Thus, based on the continual characteristic when the isosurface past the cubes, one improved MC algorithm is proposed, which can shorten the execution time obviously. And its realized in visualization tool kit (VTK). The experiment showed that the stomatognathic reconstruction time can decline to 58s. A new medical 3D reconstruction system for stomatognathic is also developed based on the improved MC algorithm. The application instances show that the improved MC algorithm is efficient and robust, which can be applied successfully in 3D reconstruction for complex curved surface of stomatognathic.关键词：3D reconstruction;surface rendering;marching cubes;visualization tool kit (VTK);stomatognathic3211|259|0更新时间：2024-05-07
- 摘要：To overcome the sensitivity to the dimensions of reduced space, and performance degradation with wrong dimension estimation of neighborhood preserving embedding (NPE) method, an orthogonal neighborhood preserving embedding (ONPE) method is proposed for manifold dimension reduction. ONPE uses neighborhood information to construct the adjacent graph, and assuming that each data point can be represented by linear combination of its neighbor points. ONPE then extracts local geometry information embedded in reconstruction weights, and obtains the low dimensional coordinates by iteratively computes the mutually orthogonal basis functions. Moreover, utilizing the local geometry during ONPE dimension reduction, a new classification method (ONPC) based on a label propagation method (LNP) is proposed. The reasonable assumption is that local neighbor information in high dimensional space is also preserved in reduced space, and the class label of a data point can be obtained through the class labels of its neighbors. Several experiments on artificial datasets and face database demonstrate the effectiveness of the algorithm.关键词：Manifold learning;neighborhood preserving embedding;linear neighborhood propagation3832|287|0更新时间：2024-05-07
- 摘要：Most of the current moving object segmentation algorithms in compressed domain were focused on the MPEGx video standards which were complexity intensive due to the use of complex mathematical models. In order to resolve these problems, a new temporal-spatio method for moving object segmentation (TSMOS) in H.264/AVC compressed domain was proposed. DCT coefficients and motion vectors (MV) in bits stream were mainly used in this algorithm. Firstly, the differential mask of moving object was abstracted using the DCT coefficients difference (QDCD) between adjacent frames and followed by morphological processing. Then the methods of normalizing MV, weighted extended vector median (WEVM) filter on the MV fields and the existent segmentation results projection were utilized to obtain the moving mask of object. Finally, accurate moving objects segmentation results were achieved by combining the differential mask and the moving mask. The experiments on H.264/AVC test sequences demonstrated the validity of the proposed methods.关键词：compressed domain;DCT coefficients;motion vectors;weighted extended vector median filter;object segmentation3107|225|0更新时间：2024-05-07
- 摘要：In order to better study plant black rot, the segmentation of black rot lesion image of cruciferous plant was carried out in this study. And the image fuzzy threshold method was used to fulfill the segmentation according to the characteristic of the lesions. First, due to the problem of automatic selection of window size for image thresholding by index of fuzziness, an adaptive window size selection method for image thresholding by index of fuzziness was put forward under a predetermined membership function and given the class numbers of image pixels.Then, image whose histogram possesses one mode or two modes with great disparity was difficult to be segmented using thresholding by index of fuzziness were studied, a histogram transformation method was presented. At last, the image could be segmented through the transformed histogram using the adaptive thresholding by index of fuzziness. The experiment results of lesion segmentation using the collected leaves show that the method proposed in this paper is of effective and robust.关键词：black rot lesion of cruciferous plant;image segmentation;fuzziness theory;adaptive threshold;histogram transformation3682|273|0更新时间：2024-05-07
- 摘要：The study of texture features extraction as a hot but difficult research topic is an important approach to image retrieval and scenery recognize. In recent years, image processing and analysis based on wavelet territory receives many attentions. But the study of image features based on wavelet territory is relatively slow. In this paper we firstly studied the texture features of image and discussed the process of incomplete tree-structured wavelet transform. Then we studied the texture features extraction based on incomplete tree-structured wavelet transform. Some new types of texture features extraction is proposed in this paper especially. Two methods of direction features extraction has been put forward. We also gave the process and scope of the application. The experiments show the effectiveness of these methods.关键词：incomplete tree-structured wavelet transform;texture features;direction features;image retrieval3221|263|0更新时间：2024-05-07
- 摘要：According to the characteristic of Gauss which smoothes the image without reference to SNR, a new algorithm with adaptive σ(space coefficient) is put forward to in this paper for purpose of both precise image detection and effective noise restraining. Firstly, inertia of moment of gray level co-occurrence matrix is used to design the σ which is suitable to the current image. Secondly, high-pass filters and low-pass filters are designed according to the σ and the next σ is determined out in accordance with the image which is filtered by low-pass filters. The process is repeated till noise is removed basically. At last, images of all levels extracted with different σ are fused to obtain the final image edge with only one pixel wide in accordance by certain rules. Simulation results indicate that when comparing with traditional algorithms and B-spline wavelet, SNR is improved 0.47%～6.07%, and computing time is increased 0.29%～6.36%.The new algorithm proposed in this paper is more efficient in precise image detection and effective noise restraining especially for low-resolution image(256×256).关键词：Gauss wavelet;gray level co-occurrence matrix;adaptive σ;edge detection3359|252|0更新时间：2024-05-07
- 摘要：Aiming at extracting the center position of light stripe accurately and efficiently in linear structured light measurement system, a method which extracts sub-pixel center position of light stripe was proposed, according to the property of the light stripes intensity gradient and non-normal distribution. The method applies low pass filter and power transformation to decrease the error caused by image noise and non-normal distribution of stripe intensity, it uses adaptive threshold method to obtain the light stripes gray edge, and extracts the sub-pixel center position of light stripe by gradient barycenter. Our experiment results proved that gradient barycenter method not only extracts the stripe center accurately, but also has good anti-noise and robust performance. The method improved the measurement precision of the 3D measurement system remarkably.关键词：3D measurement;linear structured light;light stripe center;gradient barycenter method;sub-pixel4532|276|0更新时间：2024-05-07
- 摘要：For the purpose of rapidity and flexibility，this paper proposed a coupled functional framework for target tracking and motion estimation based on geometric active contour and level-set method. Our model estimates the displacement of object while tracking an objects contour and uses this displacement estimation to constrain active contour evaluating. Anglicizing image sequences needs models to have high efficiency and precision, we solve this problem with two methods：(1) Improving tracking effect and range as well as the accelerated convergence speed by modifying the active contours external force; （2）Proposed a new method for effective local motion estimation based on level-set information which is acquired from a coupled functional model. Because our entire formulation is based on geometric active contour, and formulation provide level set information of object in image naturally, therefore we can make use of level set information to assist constructing rapid motion estimation method. The second method deals with both rigid and non-rigid motion. Experiments on image sequences of varietals types such as MRI and video demonstrate efficiency and robustness of the proposed coupled model. One can use this model to get multi-object contour and motion estimation at the same time quickly. This model also provides solid foundation for further analysis and processing.关键词：objective tracking;motion estimation;Mumford-Shah model;level-set method3546|254|0更新时间：2024-05-07
- 摘要：SIFT(scale invariant feature transform) descriptor usually leads to mismatching because it uses the gradient information in the neighborhood of one feature point, when the extracted feature points locate in some similar structures of one image. So a method to correct SIFT mismatching based on a kind of spatial distribution descriptor is proposed. Because the spatial distribution of pixels on the image contour are different aiming at different matching points, each matching point obtained by SIFT can be described again to generate a more distinctive descriptor. Then the method corrects two kinds of mismatching using corresponding correction strategies by the new descriptor. Through the experiments on the real images, the comparing results between the algorithm and RANSAC indicate that the correction method can improve the percentage of correct matching under the condition of remaining the original right matching.关键词：spatial distribution descriptor;SIFT descriptor;distinctiveness;correction mismatching4017|265|0更新时间：2024-05-07
- 摘要：Establishing stereo image correspondence is a vital and the most difficult problem for binocular stereo measure system. An adaptive relaxation algorithm for dealing with feature point stereo matching is proposed. In this algorithm, a correspondence support function is constructed based on disparity gradient limit, and the matching can be achieved through a relaxation optimizing procedure. A new scheme for dynamically updating the relaxation parameter during the matching procedure is introduced, by which the FAR (false acceptance rate) & FRR (false rejection rate) can be significantly reduced compared with other algorithms with parameters being fixed. After the relaxation procedure, the disparity gradient limit is re-imposed to further filter out false correspondences. It has been validated by experiments that this strategy can efficiently reduce FAR at a modest cost of increase in FRR, which meets the requirement of restrict FAR limitation in many industrial applications. The algorithm has been used in a binocular stereo measurement prototype system, and its robustness and effectiveness is affirmed by subsequent stereo reconstruction.关键词：stereo correspondence;epipolar constraint;relaxation;binocular stereo vision;3D reconstruction3695|292|0更新时间：2024-05-07
- 摘要：The point pattern matching algorithm based on the spectral method performs poorly when the two point patterns to be matched are not isomorphic. To improve the matching performance of the spectral method for non-isomorphic point patterns，it is combined with Delaunay triangularization process and a new inexact point pattern matching algorithm is proposed. The algorithm defines the point patterns’ local structures based on Delaunay triangularization to reduce the influenced area of the points that are in one point pattern and have no correspondent in the other. The local structures from the two point pattern are matched with each other using the spectral method. After the most similar local structure pair is found it is used to guide the matching of the remaining points within the two point patterns. Simulation experiments show that the proposed algorithm is superior to the existing point pattern matching algorithms which are based on the spectral method.关键词：Delaunay triangularization;spectral method;local structure;point pattern matching3301|233|0更新时间：2024-05-07
- 摘要：This paper presents a design of the hidden surface removal hardware module based on the Cache and Hierarchical Z-Buffer algorithm. The hardware module can save bandwidth while increasing speed and improving efficiency of hidden surface removing, which is suitable for bandwidth-limited on-chip 3D graphics rendering system. The design optimizes the Z Cache structure to acquire high hit rate, and uses one-level Hierarchical Z-Buffer algorithm to enhance the effect, meanwhile affiliates the Fast Z Clear structure to save bandwidth. The design has been described the RTL models and has passed the simulation. Experimental results show savings of about 30％ of the bandwidth, speed and efficiency of removal up to 20％ at best.3735|247|0更新时间：2024-05-07
- 摘要：A novel speech rate related lip movement model is proposed in this paper. The model is based on the research results on the viscoelasticity of skin-muscle tissue and the quantitative relationship between lip muscle force and speech rate. In order to show the validity of the model, we have applied it to our Chinese speech animation system. The experimental results show that our system can synthesize the individualized speech animation with high naturalness at different speech rates. Finally, the perceptual evaluation experiment is designed to evaluate the quality and intelligibility of the synthesized speech animation.关键词：speech rate;skin-muscle tissue;speech animation;perception evaluation experiment3543|266|0更新时间：2024-05-07
- 摘要：Animation remapping is an important way to produce realistic facial animation. We introduce spherical parameterization to prevent overlap and utilize barycentric coordinate as the interpolation weight of motion vectors, which achieves fast remapping of realistic animation and requires fewer key points. Combined with automatic division of lip contact line and lip region division, it can correct errors of the mouth animation. The method is applicable to real-time facial animation remapping for sparse markers in Motion Capture.3629|253|0更新时间：2024-05-07
- 摘要：In order to extract object from images correctly and to ensure good results under illumination changing between images, a new background cut algorithm is proposed. First, an improved graph cut method is used to enhance the cutting results. Then the color and contrast attenuation models are presented to replace the color term and the contrast term in energy function which will reduce the infection due to illumination. Finally, we use morphology operation and holes filling methods to increase the final effect. Experimental results indicate that this algorithm is more effective than traditional background cut algorithms especially when the source image and the background image hold different illuminations or when light is covered by the object. It can satisfy the system requirements of extracting object in complex environments.关键词：background cut;illumination changing;graph cut;energy minimization3870|264|0更新时间：2024-05-07
- 摘要：This paper presents a novel watermarking scheme based on singular value decomposition(SVD) for 3D meshes, it embeds the watermark into the singular values which equal zero. This watermarking scheme is non-invertible, since the singular value is zero. This algorithm can quickly embed the watermark to large-scale 3D meshes and is robust to various attacks. Finally we analyze some necessary conditions to construct the non-invertible scheme.关键词：3D meshes;digital watermark;interpretation attack;non-invertible algorithm3453|211|0更新时间：2024-05-07
- 摘要：2D barcodes are two-dimensional graphical patterns that encode information. During encoding there is an important stage named mask modulation the purpose of which is to limit the average gray level of each area of resulting image and to make it suitable for printing and recognition. Since the traditional schemes are mostly based on perceptual template pattern selection and lack formal method and objective evaluation is based on quantitative analysis, resulting non-optimal selection during application. Our task is to objectively and mathematically evaluate the mask modulation. In this paper, based on a 2-dimensional continuous matrix bar code with uncertain capacity, we introduce an evaluation method of graphical patterns based on mathematic statistical model. This evaluation is more impersonal and optimal than the traditional method.关键词：two-dimension barcode;mask pattern modulation3379|232|0更新时间：2024-05-07
- 摘要：The detection and location of commercials in news video has been a challenging problem because of the diversity of the production styles of commercial programs. In this paper, the authors present a novel algorithm for the location of commercial segments in news program.With the proposed method, firstly shot transition detection is conducted, which is based on the imbalanced data classification methodology, and then clustering analysis is employed to label commercial blocks roughly. Finally the accurate boundaries between the commercial segments and news story are located by analyzing a new set of features, including the average duration of preceding and subsequent shots and the visual features of the shots, such as color, saturation and edge distribution. The experiment results show that the proposed algorithm is effective with high precision.关键词：commercial segments location;clustering;temporal and visual features;shot detection;imbalanced data3219|225|0更新时间：2024-05-07
- 摘要：The most important problem in the dynamic urban planning simulation system is abstracting and extracting the temporal-spatial logic relationships between those objects involving in the system, and then reasoning and computing the dynamic simulation process using those relationships. To establish the temporal-spatial reasoning system in CAUPS, the temporal-spatial reasoning rules for the simulation of computer aided urban planning system with specific planning are proposed aiming to provide the basic principles for these computer aided urban planning system. Those basic principles are established based on the traditional temporal-spatial logic relationships and they also concern the non-rigid objects in urban planning, so our approach could support the temporal-spatial reasoning between non-rigid objects such as plants and water. The resolution to execute these temporal-spatial reasoning principles in multiple-agent system is presented which includes two reasoning algorithms, the temporal reasoning executing algorithm and the spatial reasoning executing algorithm respectively. These two algorithms have been implemented in our core CAUPS system which uses these two algorithms to adjust the multiple agents relationships.关键词：temporal-spatial reasoning;multiple-agent system;qualitative reasoning3441|207|0更新时间：2024-05-07
- 摘要：Wavelet transform is a powerful method in image denoising, and it is of great significance to select an appropriate kind of wavelet transform. V-system of degree k is a class of complete orthogonal function system on L2［0,1］,which is composed of sub-function of degree k, with the characteristics of the multi-resolution of multi-wavelet. Particularly, its basic function contains both continuous and discontinuous functions, which makes it different from classic orthogonal function system. In this article, a new kind of orthogonal discrete transform called DVT is proposed, and the application of DVT in image denoising is studied. The comparisons of the DVT with wavelet transform in image denoising are given. Using PSNR as evaluation criteria and using MATLAB processing to carry out a lot of experiments, the comparison results indicate that the linear DVT has obvious advantages in the denoising of Gaussian noise, salt & pepper noise and speckle noise.关键词：orthogonal function system;V-system;discrete V-transform;image denoising;wavelet3394|272|0更新时间：2024-05-07
- 摘要：With the development of virtual reality，3D reconstruction based on image has drawn increasing attentions. The biggest merit of it is that the posture and texture information can be directly extracted from images to reconstruct geometric models with photo-reality simply and rapidly. The common method calibrates camera in the first step, the calculates space points position in a world coordinate. This paper shows a brand new method which reconstructs models from two images. Firstly, using properties of vanishing points to calculate the position of camera center in the world coordinate. In addition, according to the geometric restriction of the camera,the model posture and position can be acquired. Finally,we extract and rectify the texture from images. In this approach any extrinsic parameters is not needed. The result of experiments proves that the accuracy of this algorithm, and that it can meet the requirement of common object modeling.关键词：virtual reality;image baesd modeling;geometry model;vanishing point;camera calibration3593|213|0更新时间：2024-05-07
- 摘要：Traditional approaches to generate digital elevation model（DEM) from aerial imagery consist of two steps. The first step establishes feature correspondences and determines their height, and the second interpolates height to generate dense DEM. Because the first step does not apply global optimization and the second step usually introduces interpolation error, they impair the quality of DEM. This paper describes DEM as Markov random fields, formulates DEM generation as pixel labeling. It generates DEM in a global optimization framework and does not need interpolation. Then, this paper constructs multi-resolution Height Fields and proposes a multi-level pixel labeling strategy. It determines the Height Fields on the highest level at first, and then determines the Height Fields on the rest levels step by step. It improves efficiency greatly. At last, this paper modified Belief Propagation algorithm to determine Height Fields on a specific level. It passes Height Fields on the higher level to the lower level, restricts the possible height and reduces the search space greatly. As a result, it improves both efficiency and quality. Experimental results have shown that high quality DEM have been generated by the proposed approach.关键词：DEM generation;stereo matching;Markov random fields;global optimization3957|227|0更新时间：2024-05-07
- 摘要：Augmented reality is a technique to fuse virtual and real worlds. The proposed augmented reality system aims to use an off-the-shelf web camera, and augments related virtual media information onto the Chinese-character-markers surrounded by black rectangles, which provides a new learning experience for Chinese character. The proposed method enhances the marker detection method with edge extraction, and performs well under complicated shading condition. To reliably solve the camera parameters, the proposed system first uses the homography to estimate the initial values, and then incorporates adaptive smoothness constraint to reduce the image noise. This constraint makes the solved camera parameters very accurate and stable. The experimental results demonstrate the effectiveness and robustness of the proposed system.关键词：augmented reality;edge extraction;Chinese-character-marker;camera tracking4100|263|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