最新刊期
卷 15 ， 期 8 ， 2010
- 摘要：The development of interaction technologies is advancing new interaction models and theories. Implicit human computer interaction is a new interaction paradigm, which models a new interaction pattern that people is served without manipulating the interaction devices. The interaction system serves to people based on the understanding of their intention by analyzing the behavior and being aware of context in real time. Based on the analysis of the requirements of implicit interaction, this paper proposed a generalized application oriented service share model (AOSSM) for constructing distributed vision system to support implicit interactions. Two implicit interaction systems on group interaction and ambient kitchen are implemented, which demonstrated the effectiveness and the generalization of AOSSM.关键词：simplicit interaction;vision system;interaction theory;context aware3688|275|0更新时间：2024-05-07
- 摘要：As a brief representation of video content, video summarization can effectively assist users’ browsing and organizing video clips. This paper presents a novel method to generate video summarization based on sketches. Based on the result of video content analysis, this method visualizes video’s main content, taking advantages of sketches’ efficiency and succinctness. Firstly, we extract semantic features and key frames from the video. Secondly, user interaction is adopted to generate sketches from the key frames. Finally, the layout is arranged to get the final summarization result. Experiment result shows that the method can represent the main objects and main events of video materials, and has a quite high user satisfaction.关键词：video summarization;sketch;non-photorealistic rendering3355|227|0更新时间：2024-05-07
- 摘要：A new method is proposed for face pose estimation based on the area ratio in the paper. First, the change of triangle area ratio which consist of feature points is analyzed when a face turns. Then, the BP neural network is applied to train the relationship between position parameters of area ratio of face and the parameters of pose for face pose estimation. Finally, the method is applied to roam in the virtual environment. The experimental results demonstrate that this method can estimate accurately pose using only few characteristic points, and also does well in recognition rate and stability.关键词：pose estimation;artificial neural network;human-computer interface2857|261|0更新时间：2024-05-07
- 摘要：Based on Augmented Reality, we developed a virtual assembly system using Tangible User Interface. In the condition of single camera, it connects real markers and virtual models of the components. Users control the virtual models by manipulating tangible markers to accomplish the assembly procedure. Compared to traditional virtual assembly, it improved the sense of reality during the operation. We designed four kinds of markers according to the demands of the assembly task. Three interaction techniques, including proximity, constraint information judgment and view control are proposed. Besides, we also used collision detection to realize accurate and natural virtual assembly. The system can help designers analyse and improve products, and it’s also easier for users to learn.关键词：tangible user interface;virtual assembly;constraint;proximity;view control3380|222|0更新时间：2024-05-07
- 摘要：Delaunay triangulation reconstruction is a hotspot but hard problem in 3D scene rendering and visualization field. In this paper, a review of Delaunay triangulation development is given, and then three current kinds of Delaunay triangulation methods are summarized: incremental method, triangle expanding method and divide-and-conquer method. Moreover, several kinds of algorithms under the frame of each triangulation method are compared in terms of advantage, disadvantage and complexity. Finally, directions of future work of Delaunay triangulation methods in large-scale scene rendering and terrain visualization field have been discussed, including research on algorithm combination, algorithm supporting technology and distributed parallel algorithm.关键词：Delaunay triangle;triangulation method;triangulation efficiency;comparative research4095|307|0更新时间：2024-05-07
- 摘要：A systematic and deep research is carried out into the camera calibration methods based on the vanishing points theory and planar control fields. Then a practical algorithm to obtain the initial values of the exterior orientation elements is deduced by combining the theory of vanishing points and collinearity equations in planar scenes. Detection methods of signalized points based on 2D direct linear transformation was proposed. Mathematical model of bundle block adjustment with self calibration is given and the method of determining the;weight of every class of observations based on the calculating the QVVP is proposed. The performance between two camera calibration methods is compared and some valuable conclusions are acquired. It can provide some advices in practical applications. But how to realize the full automatic detection and localization of target points should be improved in the future.3848|281|0更新时间：2024-05-07
-
A Novel Second-order Distribution Maintained Steganographic Algorithm Based on Markov Chain Security
摘要：It has been a difficulty and hotspot in steganography research that how to design large payload steganographic algorithm while high-order statistical security is ensured. Based on the Markov Chain security benchmarking and dynamic compensating means, a novel second-order distribution maintained steganographic method was proposed, which preserves the second-order distribution of the cover image and the embedded payload is unreduced. Experimental results showed that the proposed algorithm can maintain the second-order distribution of cover image while larger secret information is embedded, and improve the steganographic security.关键词：information hiding;steganography;Markov Chain;second-order distribution maintain3235|359|0更新时间：2024-05-07 - 摘要：In order to reduce search space efficiently, the codebook must be rearranged in the fast codeword search algorithms of vector quantization. Two kinds of algorithms are summarized in this paper. One is the codebook sorted by 1D characteristic value; the other is the codebook mapped into 2D adjacent ripples. Comparisons and analysis are conducted for these two kinds of fast algorithms in terms of the search space and encoding time. The criteria about how to use these two rearrangement method reasonably is also given in this paper.关键词：vector quantization;fast codeword search;codebook rearrangement;performance evaluation3348|182|0更新时间：2024-05-07
- 摘要：This paper proposes an invertible semi-fragile video watermarking algorithm using a hash function to authenticate the MPEG-4 video contents. The proposed algorithm embeds two watermarks into I frame while encoding the YUV video to MPEG-4 format. One watermark using a hash function aims to authenticate the contents and embed the frame number for tamper localization between frames, and the other one based on direct coefficients and low-frequency coefficients is used for the detection of tamper localization within the frame. The experimental results show that the proposed algorithm is able to authenticate the video contents and detect the tamper localization, and it is robust for MPEG-4 compression. In addition, the proposed algorithm is exactly invertible, which means that the original video data is available as long as the watermarked video is credible.关键词：video watermark;MPEG-4;semi-fragile;invertible;tamper localization3071|220|0更新时间：2024-05-07
- 摘要：Fuzzy feature vector has been used to represent the characteristics of image on different regions. The fuzzy feature vector has been integrated into Mumford-Shah model for the image segmentation. The generalized Mumford-Shah model has contained more information of image, enhanced capability of segmentation image and not increased complexity in comparison with original model. The original model is a particular case of the generalized model. The generalized model provides well segmentation to weak edges， concavo-convex region and complexity background image. Experimental results of applying the scheme to artificial and real images demonstrate its segmentation power.关键词：fuzzy feature;image segmentation;Mumford-Shah model3021|361|0更新时间：2024-05-07
- 摘要：A novel background model based on pixel layer for moving objects detection in a dynamic scene is presented in this paper. Fast mean shift approach is used to cluster into layers where those pixels share similar statistics. The background is then modeled as a group of pixel layers. An incoming pixel is detected as foreground if it does not adhere to these layer-models of the background. The experiments show that the proposed method performs better than the traditional MoG method under dynamic background and especially in the presence of nominal camera motion.关键词：video analysis;moving objects detection;background model;visual surveillance3240|330|0更新时间：2024-05-07
-
A Text Detection Algorithm Based on Local Haar Binary Pattern with Multi-scale Directional Filtering
摘要：The main difficulties for text detection are due to complicated illumination condition and text-background weak contrast in natural scene images. In order to resolve them, this paper presents a local Haar binary pattern(LHBP) based algorithm for text detection. Firstly, LHBP texture description operator is presented. It is insensitive to illumination variance and can effectively describe the text feature. Secondly, the Multi-scale Directional Filtering based on LHBP is proposed for fast filtering to obtain candidate text regions. Finally, the LHBP-histogram-based SVM is presented to refine the text location. Comparing with state-of-the-art algorithms, the experiment results demonstrate the robustness of the proposed method with better accuracy.关键词：text detection;Haar wavelet;local Haar binary pattern(LHBP);support vector machine(SVM)3747|253|0更新时间：2024-05-07 - 摘要：The Beamlet transform is an efficient tool for multi-scale analysis. A new edge detection algorithm is proposed through improving the algorithm of linear feature extraction based on Beamlet transform. A definition of weighted average for the gray values on a Beamlet is proposed, all dyadic squares of an image are searched for edge in all directions of Beamlets. The evaluation of this algorithm is taken from several aspects, such as the continuity of the edge detection, the antinoise performance and so on. The algorithm is applied to detect edge and identify lane from road images. The experiments show that the continuity of edge detection is good, the new method has low rate of wrong detection and miss detection, which also has good performance on line extraction. But the performance of antinoise is inferior and the calculation has low speed. The edge we gained contains information on the position, the orientation and so on, which will be convenient for the target recognition.关键词：multiscale geometry analysis;Beamlet transform;edge detection;lane recognition3505|359|0更新时间：2024-05-07
- 摘要：Distorted fingerprint matching is an important but difficult problem for automatic fingerprint recognition system. To cope with it, we propose a novel approach for distorted fingerprint matching via stable region (SR). By means of SR, the issue to match two fingerprints is tactfully converted to a search and find the corresponding SRs in the fingerprints. Through a three-step procedure which consists of construction, confirmation and expansion, corresponding minutiae are matched in different spatial scales from points to local structures and in different kinds of distorted regions from linear to nonlinear. The proposed algorithm is evaluated with FVC2004 database, and the experimental results show that the proposed algorithm has excellent performance and is strongly robust to nonlinear distortion.关键词：fingerprint matching;nonlinear distortion;stable regions3134|233|0更新时间：2024-05-07
- 摘要：The fidelity in description of pollen exine sculpture, which is one of the key benchmarks to measure the evolutionary extent of angiosperms, is of great academic and practical significance. Following an analysis of the features and deficiencies of the existent methods to measure the pollen exine sculpture, we provide a method based on computer image processing technology. The new method, whose robustness is verified through comparative analysis of experimental data, achieves rapid and accurate measurements of the pollen exine sculpture.关键词：image processing;pollen exine sculpture;radix paeonia4065|294|0更新时间：2024-05-07
- 摘要：Image quality is mainly affected by its structure and content. Traditional image quality evaluation metrics based on structural similarity put emphasis on image structure, but inadequately consider local features of image. So their application fields are limited and performances are unstable. If dividing an image into more meaningful structural blocks, the impaction of local features on image quality can be represented adequately and metric performance can be improved greatly. Based on these considerations, this paper proposes a new quality index using local character. It is implemented by three steps. Firstly, the image is divided into separate meaningful blocks according to a new image division algorithm. Different blocks represent different structures of the image. Secondly, the gradient of the image is used to weigh the influence of different pixels, and then the structural similarities of corresponding blocks between the reference image and distorted image are calculated. Finally, the ultimate image quality is calculated by combining structural similarities of all blocks according to their weights. The experiments show that the proposed metric is more reasonable and stable than traditional methods, and could be used in more application fields.关键词：image division;structural similarity;image quality assessment3309|769|0更新时间：2024-05-07
- 摘要：Linear space-invariant image restoration algorithms often introduce ringing artifacts near sharp intensity transitions. It is shown that these artifacts are attributable to the noise, the different of the restoration parameters and the restoration algorithms. This paper presents a blind deconvolution image ringing metric. According to different ringing artifact types, two ringing metrics are presented to measure the ringing artifacts in restoration images by using Gabor filter and Co-occurrence vector etc. Finally, a whole ringing metric is proposed. The experimental results validate the proposed method perform well over a wide range of restoration image ringing levels and restoration algorithms. And the proposed model is in good agreement with MOS.关键词：image quality assessment;restoration;ringing;edge;metric4678|218|0更新时间：2024-05-07
- 摘要：Target detection is difficult to be realized in complex scenes when there are moving background objects such as trees. In this paper, a new target fusion detection method is proposed based on background model. Firsty, by combining temporal information of per-pixel and the spatial information in the local region, we introduce a variant of multimodal mean model called the spatiotemporal multimodal mean model that is well suited for the non-stationary scenes. Then, the proposed background model is separately used to extract foreground pixels in visible and infrared image sequences, and a fusion detection method based on the confidence map is proposed to get the target detection result. The multi-sensor information can improve the detection precision and handle different environmental conditions. Experiment results demonstrate the effectiveness of the proposed method.关键词：target detection;spatiotemporal multimodal mean model;fusion detection3116|281|0更新时间：2024-05-07
- 摘要：Geodesics are used widely in computer graphics and other field. MMP is a sophisticated method while geodesics are calculated on triangular meshes. We achieved a precision geodesics algorithm based on MMP, and modified the error which occurred when adopting the Vitaly Surazhsky’s method. At the window propagating, lamp-house-radial method was used instead of using an approximate type. In dealing with window intersection, the complex and multiplex sorts were treated with layered enumerate type. A clear and comprehensible algorithm used in backtracing. At last, the result was presented, the run time is little more than Vitaly Surazhsky’s to some exerd, but the proposed method is precise.关键词：geodesics;MMP algorithm;window function;triangular meshes3640|261|0更新时间：2024-05-07
- 摘要：Due to rapid development of remote sensing technology and worldwide application of remotely sensed data, image fusion is an effective way to incorporate data from different remote sensors to create an improved image containing much more spectral and spatial details, and could be used to facilitate visually interpreting of remotely sensed imagery or subsequent mechanism analysis. It has been proved that better fusion images, which contain the spatial information of the panchromatic data and details of the multispectral, might be produced by the integration of wavelet transform with IHS. But the influence of some important parameters has been neglected in the practice of integration, such as levels of wavelet decomposition. Most commercially remote sensor data are composed of hundreds and millions pixels with a number of bands. Thus, the applicability of research conclusions has previously been suspicious. Therefore it is helpful to have a deep study to explore those questions.关键词：wavelet decomposition level;fusion;wavelet transform;IHS;Remotely sensed data3217|268|0更新时间：2024-05-07
- 摘要：The paper summarizes three modes for the use of neural network regression to fuse remote sensing images, while proposing a new pansharpening method, based on neural network regression, to fuse thermal infrared(TIR) image and the panchromatic(Pan) image, which can hardly be done using traditional image fusion techniques. Extreme learning machine algorithm is applied to obtain the regression relationship between remote sensing data, in a rapid and efficient manner, while the pansharpening for TIR focus on the internal physical relations of pixel values recorded as an image, and aiming at a real improvement of the TIR data quality rather than a visual enhancement. TIR data synthesized by this new image fusion method is qualified to be used in physical models. This provides convenience for quantitative remote sensing applications. Experiments on ETM+ images prove the effectiveness of this approach which achieves fairly accurate results, while direct fusing mode achieves dissatisfactory results.关键词：remote sensing;image fusion;Pansharpening;regression;ELM3120|337|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