最新刊期
卷 20 ， 期 3 ， 2015
- 摘要：Hemodynamic simulation is an important tool for revealing the pathological mechanisms of the initiation, development, and onset of vascular disease (e.g., arterial aneurysm, atherosclerotic plaque, etc.), as well as for disease diagnosis and vascular surgery evaluation.Therefore, this type of simulation has become a popular research hotspot in the diagnosis and prediction of vascular disease. Macro-scale and multi-scale simulation of blood flow dynamics areadopted for thisreview on the basis of the different feature scales of blood flow in hemodynamic simulation. State-of-the-art research methods and key technologies of blood flow dynamics simulation methods that target different feature scales of blood flow are discussed.Subsequently,existing research difficulties and problems are summarized.Finally, trends in the further development of blood flow dynamics simulation methodsare examined. With the rising number of patients with vascular disease, the multiscale simulation of blood flow dynamics combined with the image-based modeling of vascular and physiological blood flow data is expected tobecome ahot topic of future research. Such studies are of great significance to improvethe level of vascular disease treatment in China.关键词：simulation of hemodynamics;vascular disease;multiscale modeling;numerical simulation;boundary conditions4374|271|0更新时间：2024-05-07
- 摘要：A novel algorithm that combines color feature and edge information is proposed to detect and remove vehicle shadows in complex traffic scenes. First, a background model is built with the classical Gaussian mixture background modeling method, and the moving vehicle foreground is obtained through frame difference. Second, a serial fusion strategy that combines color feature and edge information is applied to detect and eliminate vehicle shadows. Based on vehicle shadow detection by edge information method of the moving target foreground, the RGB color feature detection method is implemented to detect the shadow area further and to obtain a precise result. Edge difference and morphological processing methods are used during the operations to detect and eliminate shadows effectively. Shadow assessment is periodically evaluated on the foreground area to improve the efficiency of the algorithm by determining the necessity of applying the proposed algorithm. By comparision with SP、 SNP、 DNM1 and DNM2 algorithm, the proposed method realizes about 10% advance on shadow detection rate and shadow recogmition rate. The high accuracy and robustness of the proposed shadow removal method are confirmed by the test results, and the effectiveness of the method is validated. The proposed method that combines color feature and edge information outperforms those based on a single feature because of their unicity. In addition, the false detection rate caused by complex edges in shadow regions and color similarity between vehicles and shadows is effectively decreased.关键词：color feature;edge feature;multifeature fusion;shadow assessment;shadow detection;shadow removal3857|341|0更新时间：2024-05-07
- 摘要：Human action recognition aims to detect and analyze human behavior intelligently on the basis of information captured by cameras. The applications of this technology include surveillance, video content retrieval robotics, and human-computer interfaces. Human behavior description is a key problem of behavior recognition. To utilize training data fully and to ensure a highly descriptive feature descriptor of behavior, a new human activity recognition method is proposed in this study. First, the brightness gradient was decomposed into three directions () to describe the behavior from different perspectives.Second, the standard visual vocabulary codebooks of the three directions for different behaviors could be obtained by directly constructing a visual vocabulary. Moreover, the standard visual vocabulary codebooks of the three directions for each behavior serve as bases to calculate the corresponding vocabulary distributions of the test video separately. The behavior of the test video might be recognized by using the weighted similarity measure between the standard vocabulary distribution of each behavior and the vocabulary distribution of the test video. The performance was investigated in the KTH and Weizmann action datasets. We obtained an average recognition rate of 96.04% accuracy in the Weizmann action dataset and 96.93% accuracy in the KTH action dataset. Our method could generate a comprehensive and effective representation of action videos. Furthermore, this approach can reduce clustering time by producing the codebooks of each direction. Experimental results show that the proposed method significantly improves action recognition performance and is superior to all available identification methods.关键词：behavior recognition;local space-time features;visual vocabulary;directional weighting3434|416|0更新时间：2024-05-07
- 摘要：Given the non-intrusive nature and broad surveillance application of face recognition, this technology has drawn considerable attention in the fields of pattern recognition and computer vision. However, expression variation is one of the main challenges in 3D face recognition because the geometric shape of a face changes drastically under expression variation. For instance, an open mouth can significantly change the topology of the facial surface, which can degrade the performance of a 3D face recognition system. To handle facial expressions, a novel 3D face recognition method based on facial profiles is proposed. First,the pose of a cropped face is automatically corrected on the basis of principal component analysis,and all facial scans are transformed into a uniform pose coordinate system.A set of vertical facial profiles in the upper half face region is then extracted to represent a 3D facial scan.Hence, the shapes of two facial scans can be matched by fitting the shapes of the corresponding facial profiles. Open curve analysis algorithm is applied to calculate the geodesic distance between a pair of facial profiles extracted from different facial scans.The geodesic distance is used as a similarity measure. Finally, two facial scans can be matched by using the weighted sum of all levels of the corresponding geodesic distance. One of the large stavailable public domain 2D and 3D human face datasets is the Face Recognition Grand Challenge (FRGC)v2.0, which has been widely used in the literature. Two experiments are conducted using the FRGC v2.0 dataset: recognition and expression robustness experiments. In the recognition experiment, the earliest neutral 3D facial scan of every individual is selected to create a gallery of 466 facial scans, and the rest are used as probes. We test three dataset partition methods that are commonly used in existing 3D face recognition systems,which also use FRGC v2.0 as the testing dataset(i.e., non-neutral vs. neutral, all vs. neutral, and neutral vs. neutral). The Rank-1 recognition rates of our proposed approach in the cases of non-neutral vs. neutral, all vs. neutral, and neutral vs. neutral are 95.2%, 97.1%, and 98%, respectively. In the expression-robustness experiment,we consider the gallery in the recognition experiment,and 816 facial scans with an open mouth from the FRGC v2.0 dataset are used as the testing set for face recognition. When the facial profiles are extracted from all the facial regions as features, the Rank-1 recognition rate is 82.8%,whereas that of our proposed method is 93.5%. Achieving high accuracy in the presence of expression variation is one of the most challenging aspects of 3D face recognition.To address this problem,a 3D face recognition method based on facial profiles is proposed.A set of vertical facial profiles are then extracted to represent facial surface. Given that these facial profiles are extracted from the semi-rigid region of a face, our proposed approach weakens the adverse effects caused by facial expression, especially large facial expression deformation, and consequently improves the efficiency of face matching. Experiments are performed using the FRGC v2.0 dataset to demonstrate the effectiveness of our algorithm. Results confirm the expression-robustness of the proposed method.关键词：3D face recognition;expression variation;facial profiles;pre-shape space;geodesic distance3484|688|0更新时间：2024-05-07
- 摘要：Image saliency detection is a method used to eliminate the redundant image information. Moreover, this method is used in many computer vision applications, such as adaptive compression of images, content-aware image editing, and image retrieval. In this study, a new image saliency detection method is proposed to compute for image saliency from different perspectives. In fact, many methods are used to compute for saliency, and most of these approaches use different types of features to detect saliency in single regional representation. However, only a few methods consider the adaptability between the feature and image representation. According to the different characters of different types of regional representations, we compute image saliency from different angles by using a wide variety of information, including color. The method consists of three basic steps. First, the image is mapped from the pixel space to a two-layer regional representation space on the basis of connectivity and edge information. The first layer is related to the spatial structure of the image, whereas the second one is superior in describing color information. Then, on the basis of the diverse properties of the constructed two-layer representations, we adopt a number of features to abstract image saliency. In the first layer, we use the spatial distribution of region in the image and the structure feature to obtain the spatial structure saliency. In the second layer, we use the color feature to compute for color saliency. Given the complementarity between the two kinds of saliency, the last step is to integrate the two kinds to obtain the final saliency map. In practice, color saliency has higher significance and discriminative power than spatial structure saliency. Thus, we use an exponential function to combine the two kinds of saliency while highlighting color saliency. In addition, the boundary prior is also a reasonable and popular method for enhancing saliency detection and has thus been widely used for image saliency detection. In contrast to existing methods that set a region containing boundary pixels directly to the background, we employ the percentage of boundary pixels in each region to adjust saliency values. Given that the extracted saliency clues correspond to the attributes of the local image regions quite well, our method has several advantages over existing methods. To verify the efficiency of the proposed method, experiments are performed using the MSRA-1000 dataset, which is one of the largest publicly available datasets. Results show that our method outperforms state-of-the-art methods in terms of precision, recall, F-measure, and mean absolute error. Image saliency detection is a promising approach in the field of image processing and analysis. This study presents a new saliency detection method based on two-layer regional representation through both color saliency and spatial structure saliency. Experimental evaluation results suggest that our method outperforms other methods in image saliency detection.关键词：saliency detection;two-layer representation;color;spatial structure3749|395|0更新时间：2024-05-07
- 摘要：Detecting moving targets in moving camera environments from video stream is a fundamental step in many computer vision applications, such as intelligent visual surveillance, human-machine interaction, and content-based video coding. Background subtraction is generally regarded as an effective method for extracting foreground objects. Given the existence of extensive literature regarding background subtraction, most existing methods assume a stationary camera. This assumption limits the applicability of these methods to moving camera scenarios. Background is a complex environment that usually includes distracting motions, which make the task more challenging.In addition, with the development of coding techniques, high-definition videos have become widely used. Thus, extracting and updating background images become complex and time-consuming processes. To solve this problem, a fast target detection method is proposed on the basis of motion vectors. The data format and decoding features of surveillance videos are analyzed in this study. Subsequently, methods by which to obtainmotion vectors directly from a video stream are determined, and results confirm the method validity by comparing the motion vectors extracted through this method with those extracted using the H.264 Visa business software.Moreover, the motion vectors are normalized by considering the reference frames, thus preventing the emergence of singular motion vectors and making the distribution of the motion vectors more reasonable. Global motion is detected according to 3 theory, and different compensation algorithms are proposed for various scenarios. Finally, targets are extracted by analyzing the statistical property of motion vectors. Detecting moving targets from a moving camera is difficult. For conventional methods, moving background is not always considered, and systems that have a delayed response to background changes are unsuitable for this situation. To meet the real-time performance requirements of the system, motion vectors are used to detect global motion and extract motion targets, thus reducing redundant computation effectively. Test results show that the proposed method is superior to conventional methods. To prove the validity of this algorithm, an extensive set of experiments are performed, and the proposed method is compared with some of the most recent approaches in the field by using publicly available datasets and a new annotated dataset. Results prove that the proposed approach can effectively extract high-speed moving targets from a moving camera.关键词：moving camera;high-speed moving target;motion vector;3;theory;target detection2973|336|0更新时间：2024-05-07
- 摘要：Part absence is a difficult problem in Synthetic Aperture Radar (SAR)Automatic Target Recognition. However, such problem can be resolved to some extent by detecting the single scattering part of the target in the measured SAR data. This study proposes a matched filter-based approach to detect single scattering part on the basis of the part-level 3-D scattering model of the target;this approach can well predict the scattering properties of the whole target and the single scattering part comprising the target. The following steps are used in applying the matched filter approach in detecting single scattering part:First, a matched filter is built according to the image of the single scattering part predicted by the scattering model. Second,the matched filter slides in the 3-D parameter domain detects the best position of the scattering part that is appropriate for the measured SAR image and determines the maximum correlation coefficientto enhance detection precision. Moreover, in terms of the imprecision of azimuth estimation, the premiumposition is explored in the neighborhood to obtain the best result. According to the intensity of all visible scattering parts under the present position, the strongest to the weakest parts are detected sequentially. The interference from the neighboring scattering part disturbs the detection of the present scattering part. The CLEAN algorithm is employed to overcome this problem and to eliminate the interference of the other scattering parts. A part-level 3-D scattering model of a simple tank is used in this study. Experiments are conducted on the basis of electromagnetically simulated data, dark-room measured data of the simple tank, and MSTAR (moving and stationary target acquisition and recognition) data, which are used as interference. Results demonstrate that the proposed method can detect existing scattering parts in the measured image,as well as find the absent parts. In addition,interfering data can be excluded from the experiments. This study proposes a single scattering part detection method for SAR target. Experiment results confirm the efficiency of the method.关键词：scattering part;part-based scattering model;detection;matched filter;3-D sliding;CLEAN2703|278|0更新时间：2024-05-07
- 摘要：In the creation of stereoscopic 3D images, the use of parallax adjustment to alleviate the visual discomfort caused by large perceived depth range is of great importance. This process is called perceived depth mapping or parallax mapping. Most existing depth mapping methods control the horizontal parallax by changing the camera parameters, such as the inter-axis distance and focal length, or by warping local images during the post process. However, these methods cannot be used in real-time rendering systems. This study proposes an efficient and non-uniform depth compression model for real-time stereoscopic systems. On the basis of the geometry deduction on a horizontal parallax produced by two cameras, we use a single camera to capture two image views with different projection matrices that correspond to two viewing points. A horizontal parallax can be created through simple shearing transformation. Depth compression can cause object deformations (i.e., objects appear flattened). To reduce the artifacts caused by object deformations, different compression ratios are applied at various viewing depths. Given that the areas around the screen are the most comfortable zones and that humans tend to focus on these zones, we apply a small compression ratio for objects near the projection plane and a large compression ratio for far objects. For a smooth change in compression ratio, the relation of the inter-axis distance with the view depth of the camera is defined as a continuous linear function, and the projection transformation for the two views is derived on the basis of the inter-axis distance function. Non-uniform depth mapping is modeled with three processes: horizontal warping, shearing, and normal projection transformation. Experimental results show that our depth compression method can ensure a smooth variation in the compression ratio.The proposed method effectively improves stereoscopic image quality. The warping of objects before projection cannot be identified in the final stereoscopic images. Our study shows that depth mapping can be realized through geometry transformation for virtual 3D scenes.The non-uniform depth compression method is simple, highly efficient, and can be applied to real-time stereoscopic systems, such as games and virtual reality.关键词：stereoscopic display;perceived depth;depth mapping;visual fatigue3157|669|0更新时间：2024-05-07
- 摘要：The geometric correction of SAR imagery is an important step in SAR image processing. This process has a certain degree of computational complexity and requires a certain geometric positioning model. A GPU-supported massively parallel processing method is presented for the geometric correction of space-borne SAR imagery. This method exploits the RPC model. The method takes full advantage of two facts. A GPU has large computational resources,and the processing steps are the same for each pixel in geometric correction. In the course of massively parallel processing, a large amount of pixels are imported into the GPU at each time, and one thread is allocated for one pixel. Each thread performs the steps,which include the calculation of rational function, transformation for projection, resampling, and so on,with high computational complexity. The optimal configuration of two parameters, i.e., dimGrid and dimBlock, improves parallel performance. Large SAR image frames with different sizes can be processed by block partition. Experimental results show that the proposed method can achieve computational speedups ranging from 38 to 44. Meanwhile, the speedup for the whole procedure is recorded to analyze the features of the GPU-based parallel computing objectively and thoroughly. The factors affecting the speedup for the whole procedure are discussed according to the results of several experiments, and an optimal approach that reads and writes a large block to promote I/O performance is proposed. The straightforward method has broad applicability and can be used for most space-borne SAR sensors and for different image frame sizes.The method also achieves obvious acceleration.关键词：synthetic aperture radar (SAR);geometric correction;parallel computing;graphics processing unit (GPU);rational polynomial coefficients (RPC)3898|335|0更新时间：2024-05-07
- 摘要：Image degradation during remote photography severely affects high-resolution imaging and high-accuracy detection. To improve the quality of remote sensing images, a multi-scale image deblurring method for remote sensing viaregularization constraintsis proposed in this paper. At the beginning of deblurring, bilateral and shock filtersare used to handle blurred images.Subsequently, a variational Bayesian iterative model is applied to determine the optimal solution by considering prior knowledge of the sparsity feature of the blur kernel.Finally, the deblurring result can be obtained by non-blind deconvolution based on gradient sparsity. In addition, the effect of scale information on the deblurring result is studied for the case of a serious blur,and a multi-scale iterative method is proposed. Our algorithm is implemented for deblurring numerous remote sensing images. Experimental results show that the proposed method can effectively remove fuzzy sections, maintain edges,and recover details of blurred images. Other methods are compared with the proposed algorithm.Indices such as entropy, contrast ratio(CR), edge strength level(ESL), and HSV (hue, saturation, value)model are used in the objective evaluation. The ESL average of the images increases by 28.7%,where as the CR average increases by 17.6% after using our method. Subjective visual experience and objective evaluation indices show that the proposed method can effectively improve the quality of remote sensing images.关键词：remote sensing image;deblurring;regularization;blur kernel;multi-scale4308|666|0更新时间：2024-05-07
- 摘要：Existing motion deblurring methods fail to consider the fact that blurring occurs in irradiance images. Thus, an efficient deblurring method for reducing the effect of naturally appearing saturated pixel blocks remains lacking. To overcome these two limitations, this study proposes an irradiance-based motion deblurring algorithm. A new estimation method for the camera response function (CRF) that converts irradiance to intensity is proposed.Intensity reflects the energy accumulated during the motion process of capturing the blurred image. A block-based detection algorithm for saturated pixels is presented.This algorithm can automatically detect saturated pixels. The CRF estimation method and saturated pixel detection algorithm are incorporated into the proposed deblurring method to achieve robust irradiance-based motion deblurring without the effect of saturated pixels. Qualitative experiments for reveal that the proposed method can obtain single-image deblurring obtain latent images more accurately with less ringing and better noise suppression than existing methods. Quantitative experiments return higher peak signal-to-noise ratios than existing methods. The proposed method and algorithm results are effective for improving motion deblurring performance.关键词：camera response function;saturation process;irradiance;motion deblurring4751|367|0更新时间：2024-05-07
- 摘要：In theory, color-to-gray transformation is a process of dimensionality reduction, thus making information loss inevitable. Therefore, the goal of decolorization is to use the limited gray level range to preserve as much information of the original color image as possible. Researchers have proposed many related algorithms. However, the algorithms fail in the simultaneous preservation of the local and global contrast, as well as the contrast, color consistency, and grayscale pixel, of the original image. To solve these problems, we propose a new approach that can maximally maintain the features of the original color image. To preserve the structure and local information, we use a bimodal Gaussian distribution, followed by the difference between the pixel and its neighbors, to construct the error energy function. For global color consistency, we use locally linear embedding to build the energy function, which causes the same color pixels to exhibit the same gray levels in the result. For grayscale feature preservation, we distinguish grayscale pixels and specify that the gray values of grayscale pixels are known quantities and are initially unchanged during conversion. We then construct the energy function between grayscale pixels and other pixels. Thereafter, we build the objective function with the linear combination of the three energy functions and the gray image is obtained by solving the objective function using the iterative method. Experimental results show that our algorithm can preserve salient structure and fine detail, as well ensure that the same color can be transformed to the same grayscale and that the gray value of the grayscale pixel would be unchanged after conversion. Our algorithm conforms to the degree of perception about contrast change in image and can preserve detailed information and global structure. The algorithm can be applied to digital printing, pattern recognition, and so on, and thus has great application value.关键词：color-to-gray conversion;contrast preservation;dimensionality reduction;grayscale preservation;color consistency6077|347|0更新时间：2024-05-07
- 摘要：An increasing number of people usually need to work or take a holiday with their families in another city. Therefore, having an accurate map is necessary when they arrive in an unfamiliar city. Some online maps, such as Baidu or Google, can provide sufficient but redundant information. Users only need some information about their destinations, but excessive details on other places give rise to difficulty in reading maps. Meanwhile, users often need to perform many tedious operations, such as zoom in and zoom out, to obtain useful information. Therefore, we present a factor graph-based method to generate multiple destinations maps. This method can provide users with only useful information so that they can obtain help easily. Our method consists of three steps. First, users select multiple destinations of interest. The detail of some areas that are far from the user destinations is unnecessary. Second, the proposed system automatically selects the most relevant subset of roads that are related to the destinations according to certain predefined guidelines. These guidelines ensure that the selected road network maintains conciseness and connectivity so that users can arrive at any destination. Finally, the layout is adjusted to present an accurate map. The map should ensure that the details of the information on the destinations are clear while maintaining the original topology of the map. Moreover, we should maximize screen usage to display more information. Thus, we identify a set of design rules to constrain the map layout. We conduct some preprocessing of the road network to implement layout optimization. We segment the road network into several rectangular areas according to the user destinations. Every rectangle contains only one destination. During optimization, we can choose a rectangular area-based perturbation or point-based perturbation. Furthermore, we use factor graph, a type of graphical model, to improve algorithm efficiency by encoding constraints as factors. We obtain the desired map layout with the Metropolis-Hastings algorithm by sampling from the target distribution constructed by the factor graph. Online maps fail to provide selective information, whereas other maps, such as hand-drawn maps, often produce map deformation that affects map reading. The multi-destination maps generated by the proposed method not only provide users with detailed road information of multiple destinations in the same view while maintaining the topology among destinations. Moreover, unnecessary information is automatically omitted. In this study, we present a new method to generate maps with a factor graph. The factor graph can clearly describe the relationships of objects. Experiment results indicate that the proposed approach can efficiently solve the problem confronting online maps, which fail to provide users with sufficient and detailed information on multi-destinations in the same view.关键词：multi-destinations;map visualization;factor graph;Metropolis-Hastings algorithm4810|300|0更新时间：2024-05-07
- 摘要：The widely used B-spline method is limited by its incapability to realize high-order continuity and high local adjustment simultaneously. The shape of the B-spline curve and surface is uniquely determined by the control points and knot vector. These factors influence the design effects and convenience of the B-spline method. This study aims to overcome these limitations. We construct a new curve and surface that possess the high order continuity of the high-order B-spline method, the high local adjustment of the low-order B-spline method, and the shape adjustment property of the rational B-spline method as determined by the weight factor. We first construct a set of blending functions with a parameter on the trigonometric function space. We then define the new curve and tensor product surface, which have the same structure as those of the cubic B-spline curve and surface, respectively. The new curve and surface inherit many of the B-spline method properties, such as convex hull, symmetry, and geometric invariability. The difference is that although both the new and cubic B-spline curves are based on a four-point segment, the latter is C continuous, while for equidistant knots, the new curve is generally C continuous and can reach C continuity when taking a special parameter. A new curve that is C continuous has one shape parameter, which can be used to adjust the shape of the curve without changing the control points. In addition, taking the special parameter, the new curve can automatically interpolate the given set of points. The new surface has properties corresponding to the new curve. This study presents a kind of shape-adjustable piecewise combination curve and surface, which can achieve high-order continuity under strong local control capability. These features can be used for the possible design of a high-order smooth curve and surface. The new curve realizes the uniform representation of approximation and interpolation. The construction method of the blending function is general and can be used to construct other functions with similar properties.关键词：curve and surface design;spline curve and surface;trigonometric function;shape parameter;continuity4516|412|0更新时间：2024-05-07
- 摘要：Existing algorithms for natural image sketch production easily cause such problems as loss of image details and singleness of stroke line. To solve these problems, this study proposes an improved algorithm for natural image pencil drawing production. By simulating the artist creation process of sketch, we adopt a multi-level and multi-scale method to extract details from different image levels. Furthermore, we use strokes of different lengths and thicknesses to simulate the painting effect of different strengths in the process of artist creation, as well as to draw different levels image detail hierarchically. Then, the final sketch stroke effect is obtained by fusing all stroke effect levels. We finally merge the sketch stroke effects and image background to obtain the output image, which exhibits content that is consistent with that of the original input image, tonal effect that is the same as an artist's drawn sketch, as well as stroke and background form that are similar to those of the actual sketch. The output image has a strong sense of reality in terms of sketch effect. A large number of experimental results indicate that the proposed method proposed can generate high-quality natural image pencil drawings. Moreover, the proposed method has strong comparability with the existing mature pencil drawing generation technique.关键词：pencil drawing;line drawing;nonphotorealistic rendering;tone mapping5323|473|0更新时间：2024-05-07
- 摘要：Three-dimensional (3D) city visualization is fundamental for information display in smarter cities. Existing 3D city visualization systems are mainly limited in two aspects. First, the data model of a 3D city is unsuitable for handling massive building data and does not support real-time network transmission. Second, given that the whole city is rendered in a homogeneous graphic style and that most buildings are highly similar in texture, structure, and height, the rendering results usually exhibit high visual complexity. In this study, we propose a technique for online 3D city visualization on the basis of human visual perception and spatial cognition theory. During preprocessing, our system creates a multiple representation of buildings on the basis of the building generalization technique. Then, at runtime, the system can dynamically choose an appropriate level of buildings on the basis of human interaction. The algorithm is tested on several large 3D city models. Experiment results confirm the applicability of our algorithm to interactive visualization and information intensity reduction of a 3D city. For the Leverkusen city model with 5 530 buidings, we have improred the frame rate to 19.4 frame/s. The proposed method can improve the efficiency of 3D city rendering and thus enable users to obtain relevant information through efficient interaction.关键词：smarter cities;online 3D city visualization;building;human perception5530|313|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