最新刊期
卷 18 ， 期 4 ， 2013
- 摘要：As one of the most important areas of research in the domain of computer vision and intelligent vehicles, pedestrian detection of on-board vision systems has attracted extensive interest from the research community. In order to show the general idea of the latest development of this technology, we focus on the improvement of the two major modules since 2005: regions of interest (ROIs) segmentation and object recognition. First, we classify the typical methods in the field of ROIs segmentation according to the information used in segmentation and analyze their respective pros and cons. Then, we summarize the improvement of object recognition on the feature extraction, classifier construction, and the search framework. Finally, future research trends are proposed.关键词：pedestrian detection;driver assistance system;ROIs (regions of interest) segmentation;object recognition7056|569|0更新时间：2024-05-07
- 摘要：Embedding effciency is one of the most important factors to measure the information hiding perfomance. Our proposed method is based on Deng’s scheme in order to improve its embedding efficiency. Our proposed method embeds secret data by using the highly correlated SMVQ index and variable length coding. Experimental results show that the proposed method can increase embedding effciency significantly. Compared with Deng’s scheme, our proposed method achives not only 10.43%~19.80% reduction of bit rate but also 25.61%~46.95% increasment of embedding capacity. As a result, the embedding efficiency is increased to 41.33%~71.95% in total.关键词：information hiding;side-match vector quantization;embedding capacity;bit rate;embedding efficiency3495|279|0更新时间：2024-05-07
- 摘要：For the security of the set partition in hierarchical trees compression, an algorithm of joint image compression and encryption with discrete wavelet transform and secure Hash algorithm-1 (named as DSCE) is proposed in this paper. First, the wavelet transform coefficients of the plain image are divided into a high-frequency and a low-frequency part. Then, the low-frequency coefficients, the initial keys, and the sum of total pixels are the input data of the SHA-1 to output the 160-bit hash value. The 160-bit hash value is used to produce a sub-key and to perturb the initial keys. The high-frequency coefficients are permuted by a sub-key produced with the 160-bit hash value. The initial keys perturbed with 160-bit hash value are used to partially encrypt the high-frequency coefficients encoded with the SPIHT code. Third, the partial encrypted high-frequency coefficients and the initial keys are input to the SHA-1 to output another 160-bit hash value. The hash value is used to produce a sub-key to permute the low-frequency coefficients, and it is also used to perturb the initial keys to encrypt the low-frequency coefficients encoded with Huffman code. This new DSCE algorithm can implement the mutual encryption between low-frequency and high-frequency in the process of compression. The simulation results show that our algorithm not only has a large key space, but also is highly sensitive to both the key and plain-image. Moreover, it can increase the efficiency of image storage and transmission.关键词：SPIHT coding;secure hash algorithm-1;image encryption;chaos mapping;discrete wavelet transform3138|241|0更新时间：2024-05-07
- 摘要：The unstructured road detection is a challenging problem due to the fact that many rural or desert roads are not well-paved, lack prominent lane markers or boundaries. By the constraint of vanishing point, the performance of unstructured road detection can be improved significantly. But the existing vanishing point detection methods based on Gabor filters with multiple scales and multiple orientations have high computational cost, and the disturbance of background always deteriorates their performance. In this paper, we propose a new method for detecting the vanishing point of unstructured road based on Haar texture. By using integral image technique, the complex response of Haar texture is fast calculated according to the carefully designed real and imaginary Haar templates, and then the orthogonal rectification and diversity voting methods are used to realize the vanishing point detection. Compared with the state-of-the-art algorithms in a variety of difficult environments, the experimental results demonstrate the effectiveness and efficiency of the proposed method.关键词：vanishing point detection;unstructured road;Haar texture;diversity voting3846|377|0更新时间：2024-05-07
- 摘要：The use of video surveillance in for people counting public places has an important value in the field of intelligent security. However, there are several factors such as camera perspective, background clutter, and occlusions, which restrict its development and application of the study. An algorithm based on the regression model is proposed for estimating the number of people. First, in order to eliminate the effect of the camera perspective on the image features, the input image is divided into several sub-image blocks according to the change of pedestrian height in the image. Second, the simile classifier is used to improve the advanced local binary patterns (ALBP) texture feature of the blocks. Then, according to the crowd density, we use the support vector regression (SVR), which has two kernel functions to establish the relationship between input features and the number of people. Finally, adding the number of persons of all sub-image blocks gives us the total number of people on the image. Experimental results show that the absolute error of the sparse population is approximately one person using the presented algorithm and the relative error of the testing crowded population is less than 10%. This therefore demonstrates the high accuracy of this algorithm, which can be applied for people counting in video surveillance.关键词：counting people;simile classifier;support vector regression;crowd density estimation4241|421|0更新时间：2024-05-07
- 摘要：To solve the problem of low efficiency and weak stability in searching the -nearest neighbors of a large-scale scattered point cloud, a fast algorithm for finding -nearest neighbors is presented. First, the point cloud data is divided into different sub-spaces by using a space block strategy. Second, the variation of the search step length is controlled dynamically. The accuracy of the algorithm is ensured by the minimum distance from the point to the small cube boundary. Finally, the infinite loop problem due to improper initial values in existing algorithms is avoided by altering the right-side threshold, which controls the number of pre-screening points. The experiment results show that the proposed method obtains not only a good stability for the initial searching step, the step increment, and the sampling density at different topology structures, but also a better performance than the existing algorithms.关键词：-nearest neighbors;surface reconstruction;point cloud;search step4546|280|0更新时间：2024-05-07
- 摘要：Based on the matrix form of triangular Bézier surfaces, the shape modification with geometric constraints is extended from the curve and the tensor product surface to the triangular Bézier surface. The new triangular Bézier surface does not only keep the shape nearly unchanged, but also meets the geometric constraints (multiple position and normal direction constraints). With the help of the Lagrange multiplier method, the conditional extremum problem from the shape modification with geometric constraints is equivalent to solving a system of linear equations. Particularly, the new triangular Bézier surface with the boundary (C, C, C) continuity constraints at three corners also can be obtained. Finally, the numerical examples show the validity and effectiveness in the interactive design in CAD systems.关键词：computer aided geometric design;triangular Bézier surface;shape modification;geometric constraint3376|249|0更新时间：2024-05-07
- 摘要：The optical image grading method is of interest for surveying & mapping satellite data users.For the scale, re- solution, signal-to-noise ratio, modulation Transfer function(MTF) cannot clearly reflect the mapping application capacity of the image. In this paper, we propose an image quality grading method based on characteristic features, which are usable in a the stereoscopic environment. Based on the psychological principles of physics, learn national image interpretability rating scales (NIIRS) criterion generation step, a psychophysical test is designed and implemented in the stereoscopic environment,characteristic features that is labeled in the topographic map specification are selected as candidate criterion,and aviation and aerospace stereo images are selected to build sample database,32 operaters take part in test,and a grading specification composed by the stereo pair, mapping criteria and scores is generated.关键词：feature interpretation;remote-sensing image;image quality;grade3102|272|0更新时间：2024-05-07
- 摘要：The accurate extraction of rivers is important for survey of water resources, time series change detection on water usage, assessment of large-scale water conservancy facilities, and so on. The general methods of river extraction are difficult to be applied widely because of the disruption by clouds, snow, shadow of mountains, and lakes in remotely sensed images. In this paper, we propose a new index calculation model for river extraction, which is based on an improved water index, named Gaussian normalized difference water index (GNDWI). The model can remove the interference factors effectively by the aid of a DEM. The experiment for the extraction of Ili River from Landsat images show that the new model can automatically and rapidly extract the river in very complex environments. Furthermore, shadows and other useless information can also be effectively removed with a high accuracy.关键词：Gaussian normalization;high precision;river extraction;digital elevation model (DEM);water index4350|589|0更新时间：2024-05-07
- 摘要：Point-like natural disaster phenomena, such as earthquakes, landslides, etc., have complex linkages and interactions with their surrounding geographical environment. Disaster environment on the disaster may enlarge or reduce effect. In the process of cartographic generalization, only considering a single point of individual disasters is not enough. Based on Geography and Cartography, but from the perspective of a disaster system, the applicability issues of automatic generalization of point phenomena are explored by combing the layer constraint (LC) and Fuzzy Inference System (FIS). A landslide-disaster automatic-generalization technology, based on LC and FIS, is discussed. Taking landslides as an example, an automated comprehensive application of the constraint layers, including landslide hazard zoning, terrain slope, landform regionalization, long-term seismic intensity zoning, annual torrential rain days, and annual precipitation was developed. The experiment results showed that there are three key landslide areas in China, which are the eastern slope of the Qinghai-Tibet Plateau, the Loess Plateau, and the Taihang Mountain landslide area, Wushan-Wuling Mountain Area landslide. Our research can provide an effective way for automatic generalization of natural disaster risk maps under the constraints of multi-scales and multi-layers. It can also provide a more efficient, more accurate decision-making support and technical support for disaster risk management of different regional scales.关键词：cartographic generalization;layer constraint (LC);risk map of natural hazard;landslides;fuzzy inference system3115|209|0更新时间：2024-05-07
- 摘要：Stencil shadow volumes are the main 3D rendering method for vector maps. Basic geometric graphs, such as lines and polygons, are the main representation of map features. These representation forms were so simple that strictly limited map feature information had been transferred. Consequently, in order to enrich the 3D visualization styles of a vector map, an extended rendering method based on the stencil shadow volumes is presented, which realizes real-time symbol overlaying of a vector map on terrain. First, the basics of vector data visualization based on the stencil shadow volume theory are reviewed. Second, the extended method is introduced, which includes border smoothing with rounded caps, while dealing with outlines and overlaps. Finally, the experiment results show that a real-time display of vector maps on terrain can be achieved and that the visualization effects have been improved by cartographic symbolical overlay.关键词：vector map;virtual terrain environment;stencil buffer;shadow volume;symbolization3181|250|0更新时间：2024-05-07
- 摘要：The identification of copy-paste forgery image is to find the suspicious region via pixel-by-pixel or block-by-block match, the computation costs is very heavy. An efficient blind forgery image detection approach based on scale-inva- riant feature transform (SIFT)is proposed in the paper, which employs SIFT keypoints for positioning the initial suspicious forgery region. The asymmetric search is exploited for refining the suspicious region and determining the forgery area. Experiments demonstrate that the proposed algorithm could significantly decrease the number of candidate search block, accurately identify the copy-paste forgery region regardless of the existence of gauss, salt and pepper noise, and the computational cost is reduced by 1~2 orders of magnitude with compare to the conventional methods.关键词：scale-invariant feature transform;asymmetric match;copy-paste forgery;image splicing;blind detection5215|333|0更新时间：2024-05-07
- 摘要：Reliable estimation of visual perception reflects the understanding of the meaningful structure in an image. In this paper, we describe a complete abstraction framework for images and videos that explicitly responds to this goal. The method stylizes salient images or video parts and non-salient images or video parts differently by an edge preserving filter, and keeps a harmonious transition between the two parts. First, we introduce an automatic salient object segmentation algorithm to distinguish salient regions, and it is a saliency computation based local spatial neighbors. Taking into account the actual needs, we provide an interactive technology, which can be convenient for specifying image salient structure information on purpose. Based on the generated salient structure information mask, we use a single-scale anisotropic filter to process the salient parts, and use a multi-scale anisotropic filter to process the non-salient parts so that we can implement a strong abstraction effect.The proposed method generates a kind of image and video abstraction that does not only represent a preferable visual effect, but also can be applied to the implementation of another non-photorealistic rendering (NPR)result after soft image quantization. Experiments show that our algorithm could get the desired result for processing a certain number of images and videos.4558|402|0更新时间：2024-05-07
- 摘要：An effective grayscale image colorization technique is presented in this paper by annotating the image with a few color scribbles. A cost function from a local linear model optimization assumption on Lab color channels is designed and derived. By taking advantage of the matting Laplacian matrix, the local linear model optimization can produce high quality colorizations as existing methods, while having better performance in color bleeding with sparse constraints. Our local linear model optimization is actually the global optimum of the cost function, which can be solved with a sparse linear system. We further improve the performance of our primary model to use diffusion distances instead of Euclidean distances for the construction of the matting Laplacian matrix. The experimental results show that fewer scribbles are required and better colorizations are produced with the improved diffusion distances based optimization model.关键词：colorization;matting Laplacian matrix;diffusion distance;grayscale image5648|293|0更新时间：2024-05-07
- 摘要：Motion de-blurring is a heated topic in the area of image quality enhancement, where kernel estimation is the key challenge. In this paper, we propose a new approach, which first performs segmentation on the input image and then deblur different regions. The method described in this paper includes the following steps: image decomposition, image segmentation, region-based kernel estimation and non-blind de-convolution,as well as image reconstruction by use of overlapping regions. Experimental results show that our method can outperform single kernel motion de-blurring methods in some cases.关键词：motion blur;image segmentation;deconvolution;overlapping regions5116|517|0更新时间：2024-05-07
- 摘要：Using 3D visualization reconstruction technique to get the three-dimensional structure of the cerebral vessels and its related organizations is helpful for the cerebral vessel disease diagnosis. Considering its special position, morphological complexity, and changing gray information, we hope to display its spatial structure accurately using a reconstruction technique. In this paper, we present a CUDA(compute unified device architecture)based ray-casting volume rendering to express the cerebral vessel model by introducing the factor of depth and thickness such as silhouette enhancement based on curvature, boundary enhancement on depth and the depth cue enhancement based on stereo display and color blending etc. With those real-time rendering methods, the three-dimensional structural information about depth, gradient, and direction of the viewing is clearly visible. The experimental results show that those methods can not only meet the need of accuracy but also improve the rendering quality.关键词：medical visualization;cerebral vessel remodeling;ray-casting volume rendering;illustrative enhancement4219|374|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