最新刊期
卷 18 ， 期 5 ， 2013
- 摘要：This is the eighteenth in the annual survey series of the yearly bibliographies on image engineering in China. The purpose of this survey work is mainly to capture the up-to-date development of image engineering in China, to provide a convenient means of literature searching facility for readers working in related areas, and to supply a useful reference for the editors of journals and potential authors of papers. Considering the wide distribution of related publications in China, 792 references on image engineering research and technique are selected carefully from 2 912 research papers published in 134 issues of a set of 15 Chinese journals. These 15 journals are considered as important journals in which papers concerning image engineering have higher quality and are relatively concentrated. Those selected references are classified first into 5 categories (image processing, image analysis, image understanding, technique application and survey), and then into 23 specialized classes according to their main contents (same as the last 7 years). Some analysis and discussions about the statistics made on the results of classifications by journal and by category are also presented, respectively. This work shows a general and off-the-shelf picture of the various progresses of image engineering in China in 2012.关键词：image engineering;image processing;image analysis;image understanding;technique application;literature survey;literature statistics;literature classification;bibliometrics3362|305|0更新时间：2024-05-07
- 摘要：The automatic recognition of traffic signs in dynamic scenes has attracted a lot of attention due to the development of intelligent vehicles.In recent years, many scholars have made great efforts on the detection, tracking, and recognition of traffic signs. In this paper, we review the key technologies and give an introduction about the publicly a vailable datasets. Finally, we point at some future research directions.关键词：dynamic scene;traffic signs;detection;recognition3635|344|0更新时间：2024-05-07
- 摘要：While using compressed sensing to measure and recover images, the approach to divide an image into blocks may improve the speed of reconstruction but will bring strong blocking effects. In order to solve this problem, we propose an adaptive block compressed sensing based on edge detection at the encoder, and a smoothed projected Landweber(SPL) reconstruction algorithm based on principal components analysis(PCA) at the decoder. The reconstruction algorithm uses PCA to train a dictionary adapting to image structure using hard thresholding, thus the image blocking effects are eliminated effectively and the reconstructed image quality is improved. In order to improve the efficiency of the hard thresholding and to reduce the training complexity, three hard thresholding schemes using patch based PCA are proposed: global PCA, local PCA, and hierarchical PCA. As a comparison between our experimental results and the traditional block compressed sen- sing shows, the peak signal-to-noise ratio(PSNR) of the recovered images using the SPL reconstruction algorithms combined with the adaptive block compressed sensing is improved by 13 dB; Regardless of which method is being used at the encoder, the SPL reconstruction algorithms using PCA based hard thresholding are better than SPL algorithms using directional wavelet based thresholding on PSNR.关键词：block compressed sensing;edge detection;adaptive measurement;principal components analysis;hard thresholding;directional wavelet;smoothed projected Landweber reconstruction3383|274|0更新时间：2024-05-07
- 摘要：Focusing on the problem of object tracking, an object-tracking algorithm based on visual saliency and particle filtering is proposed. Based on the research results of the human visual attention mechanism, this algorithm integrates features such as color, intensity and motion to generate the visual saliency feature. Both, the visual saliency feature and the color distribution model, are used as the representation model of the object. Particle filtering is used to track the object. This algorithm can overcome the instability brought by using a single color feature. It can also solve the difficulty caught by object shape changes, illumination variation, and the problems caused by target objects and background having similar color distributions. The algorithm has been tested on many video sequences. Experiment results and analysis are presented in this paper. The experimental results show that this algorithm is robust and it is effective and valid for object tracking.关键词：object tracking;particle filter;visual attention;visual saliency3530|709|0更新时间：2024-05-07
- 摘要：A regional segmentation algorithm for nature sense images is presented in this paper. The method integrates two local edge cues and aims to reply the inadequacy in single visual cue based methods. Firstly, the logistic regression model is trained by 200 color images and the regression relationship between edge cues and image object boundary is established. Then, the model is employed to predict the boundary confidence of every pixel. Finally, the confidence functions are integrated into the procedure of mean shift segmentation by calculating the adaptive variable weights. Both quantitative and visual inspections operated in about 20 images show that the algorithm has better performance in suppressing over-segmentation and under-segmentation, and tightly localizing the boundaries.关键词：region segmentation;nature scene image;edge detection;mean shift3552|450|0更新时间：2024-05-07
- 摘要：Automatic image annotation(AIA) is an important and challenging job for image analysis and understanding such as Content-Based Image Retrieval(CBIR). In AIA, a model for annotation which represents the relationship between the semantic concept space and the visual feature space, is constructed by learning from annotated image datasets. The performance of the AIA is still poor because the relationship between the key words and visual features is too compli-cated due to the semantic gap. However, with constrains under the scenes, the correlation between them becomes simpler and clearer for better annotation results. In this paper, a method of automatic image annotation based on scene semantic tree is proposed. Image scenes are obtained through the annotated words using PLSA. Scene semantic trees are constructed for each image scene. Un-annotated images are classified into certain scenes with visual features and are annotated using the corresponding scene semantic tree. Using the visual features provided by Duygulu, the experiments get the more effective results on Corel5K database than TM,CMRM,CRM and PLSA-GMM.关键词：automatic image annotation;image scenes;scene semantic tree;image clustering3589|260|0更新时间：2024-05-07
- 摘要：By imitating the color constancy mechanism of the low-level human visual system, we propose a computation method of color constancy based on color adaption mechanism of cone cells by using computation model of color adaption of cone cells instead of traditional simplified Von Kries diagonal transformation of the computation model of Grey-Edge color constancy. The quantitative and qualitative analysis of the experiments demonstrate that the proposed method has better color correction results, and that the intrinsic color of the scene in the image is resumed effectualy and more stable than the algorithms of the original model. In the view of computational neuroscience, our results also provide support for the notion that color adaptation mechanism of cone cells play critical role on the capacity of color constancy.关键词：color constancy;cone cell;color adaptation;Grey-Edge assumption3264|439|0更新时间：2024-05-07
- 摘要：To improve the correct recognition rate of white blood cells images, the effective methods of image segmentation and feature extraction are studied in this article. Because of the existence of grains in some type of white blood cells (granulocyte), the result of image segmentation is seriously affected. Integrating spatial information and kernel function into the fuzzy C-means clustering FCM algorithm, this paper proposes an improved FCM algorithm. Applying this new algorithm to image segmentation and taking the measure of mathematic morphology to process segmented image, the study gets a good segmentation effect and solves the problem of cytoplasm-nucleus of granulocyte segmentation. As for the feature extraction of cells, by fuzzification of the threshold parameter in local binary pattern(LBP), the texture feature extraction method based on local fussy pattern (LFP)is proposed. The employment of the methods above in image segmentation and texture extraction supports vector machine as the classifier and tests the classification of 100 CellAtlas's white blood cells images. The results indicate that the correct recognition rate is up to 93%.关键词：white blood cells classification;image segmentation;fuzzy C-means clustering(FCM);texture feature extraction;local fuzzy pattern (LFP)3075|247|0更新时间：2024-05-07
- 摘要：In order to implement segmentation and bias correction simultaneously for brain Magnetic Resonance (MR) images with intensity inhomogeneity, a multiphase level set method based on local Gaussian distribution fitting (LGDF) model is proposed in this paper. By analyzing the local properties of the bias field model, the multiplicative factor of the bias field is induced into the local intensity means formation and thus it becomes a new variable of the energy function. Therefore, the minimization of the energy function by iteration does not only accomplish the objective tissue segmentation, but also makes an effective estimation to the bias field. Experiments on synthetic images and simulated brain MR images show the proposed method is superior to the state-of-the-art on segmentation results. Moreover, the corrected images using the bias field estimated by our method have a better visual effect.关键词：image segmentation;multiphase level set method;bias correction;intensity inhomogeneity;brain MR image2798|289|0更新时间：2024-05-07
- 摘要：Aiming at the interference of close outliers and the complexity of background features, a new anomaly detection method based on separable local projection in hyperspectral imagery is proposed. After normalizing the data, the test pixel spectrum is selected as the reference spectrum to build the target subspace. Then, the background pixels in close areas are projected onto this subspace and the formula of the abnormal degree is acquired through the modulus of the projection vectors. Finally, comparing the anomalies with the main surface features, parts of the false alarms are eliminated. The experiments were conducted on HyMap hyperspectral data and the results show that the proposed algorithm overcomes the impact of background complexity and interference pixels, especially in situations where the interference pixels and test pixel are in different classes.关键词：hyperspectral images;anomaly detection;local projection;normalized process3058|223|0更新时间：2024-05-07
- 摘要：Because of the uncertainty of imaging, large scales, and rotation difference between the match data, the traditional intensity-based matching methods are not meeting our demands. For the complexity of the remote sensing data, we developed an image matching method based on the features of the structure characteristics, combined with the excellent properties of invariant moments. At first, we are extracting image linear characteristics based on RANSAC (random sample consensus) and further extracting the structure feature. Next, we are getting the optimal matching by the invariant moment and space correlation, realizing the initial image matching. Finally, using relaxation matching method, we realize the global image matching. Experimental results show that this method is not affected by rotation, and zooming factors and it can realize an automatic and quick remote sensing image matching.关键词：image matching;RANSAC;structural features;moment invariants;relaxation method3428|263|0更新时间：2024-05-07
- 摘要：Image quality assessment has attracted increasing attention in recent years. Conventional algorithms focus on measuring the image degradation caused by low-level features like noise, definition, and resolution. However, with the ra- pid development of digital devices, images with low-noise, high definition, and high resolution can be obtained easily, making these low-level features less important in quality assessment. In this paper, we propose an image quality assessment method from the perspective of aesthetics. An image is assessed by two metrics: 1) whether it has a clear topic and 2) whether its composition is friendly to human viewers. We propose an image assessment method combining depth-of-field(DoF) and composition to achieve the above mentioned goal. We first introduce a blur measurement based low-DoF detection approach. A low-DoF image with a clear topic and blurred background is considered to have greater visual impact. Furthermore, we assess the image composition by how well it is consistent with the rule of thirds, which can make the images more compact. Finally, we combine low-DoF and composition to get an objective aesthetic assessment. The experiment results suggest images assessed with high qualities by the proposed method are also good images according to human aesthetics.关键词：image quality;blur measurement;low depth of field;rule of thirds;composition assessment4851|292|0更新时间：2024-05-07
- 摘要：In this paper, we propose a new fast dehazing method from a single image based on an accurate scattering map. First, based on the characteristics of atmospheric scattering light, we take full advantage of the edge-preserving feature of bilateral filtering to estimate the atmospheric scattering light and the local contrast of the image. Then, by introducing the comparison of pixel values and the average gray value, we get a more accurate scattering map to restore the scene radiance using the atmosphere attenuation model. Finally, we optimize the recovered image by tone mapping and local denoising to get a more realistic and clear image. Compared with the exiting state-of-the-art dehazing methods, our method could get better dehazing effect at distant scenes and places where depth changes abruptly. Our method is fast with linear complexity in the number of pixels of the input image. Furthermore, our method can be performed in parallel, thus, it can be further accelerated using GPU, which makes our method applicable for real-time requirements.关键词：dehazing;bilateral filter;tone mapping;denoising;GPU acceleration4650|359|0更新时间：2024-05-07
- 摘要：Nowadays, content based video copy detection has become a widely studied issue. Due to the uncertainty and diversity of video copy transformation, it is difficult to achieve great performance based on single visual features in video copy detection. In this paper, we propose a new method, which uses a multiple visual feature synthesizing method to solve the problem. Besides traditional local and global visual features, we additionally employ nonorthogonal binary subspace(NBS) as special visual feature to represent the video content. On this basis, the normalized cross correlation(NCC) is used to improve the performance of similarity matching of the video content. We also used other measures to improve the detection accuracy. The experiment results show that our system is robust to various video transformations, and achieves better detection accuracy.关键词：multiple visual feature synthesizing;video copy detection;nonorthogonal binary subspace(NBS);normali- zed cross correlation(NCC)4234|227|0更新时间：2024-05-07
- 摘要：Photorealistic rendering of rain streaks is a challenging task. A falling raindrop has complex optic effects due to its rapid oscillation and the reflections & refractions of a ray in itself. Furthermore, it is perceived as a motion-blurred rain streak observed by a human or a camera. In this paper, we present a Monte Carlo ray-tracing based approach to render photorealistic rain streaks. An acceleration structure is built in advance for the triangle mesh of a raindrop at every discrete time point in its oscillation period and an intersection computing is made between its motion bounding box and the sampled ray, which make a large acceleration to the ray-raindrop intersection test. An adaptive sampling technique, which avoids many samples for every pixel, is used to improve the sampling efficiency and to accelerate the image rendering. We integrate the approach into the off-line renderer(PBRT) and produce photorealistic images of rain streaks and rain scene animation. Experimental results show that our approach can be used to correctly simulate the reflections and refractions of a ray in a raindrop and efficiently render photorealistic rain streaks in the scene with any type of light source.关键词：Monte Carlo ray-tracing;rain streak;raindrop;particle system;intersection test;adaptive sampling5562|339|0更新时间：2024-05-07
- 摘要：This article describes a dense 3D reconstruction method based on video sequences. It mainly aims at the repeated projection problem, which exists in the 3D objects reconstruction from multiple depth maps. With the concept of an error cloud, a method called "optimization of fixed 3D points" is proposed to solve this problem. First, it uses depth rectification to improve the consistence. Then, it adopts a projection classification algorithm to classify all the projection points into the different error clouds which are characterized by the different 3D points through the many-to-one mapping relationship. At last, it computes the point coordinates for every cloud in 3D space by Gaussian distribution. Following the above, the point data are polygonized to recover the mesh of 3D objects. The final experimental results show that this method efficiently reduces the bad effect from the many-to-one mapping problem and it makes the reconstruction closer to the real object, which achieve a better effect.关键词：3D reconstruction;video depth maps;error cloud;optimization of fixed 3D points;spatial resolution3752|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