Search

The color reproduction is based on visual perception. The psychophysics experiments present the light source; illuminants and view background affect the results of color matching. Paper is a typical substrate for printed color matching and effects printed color greatly. Now almost all of print papers contain optical brightening agents, which changes significantly the color appearance of printed color and makes color shifting of printed color matching. This paper used CAT02 chromatic adaptation models to transform the measurements of printed color in paper with and without OBA, compared the effect of printed color matching between before and after transforming with display, present the method using chromatic adaptation transforms to improve the accurate of printed color matching.

The prediction of color appearance is the basic theory and technology in modern color science field. The aim of this study is to evaluate the color appearance of packaging and printing atlas. This paper took high gloss double-sided coated paper color printing atlas as the experimental samples. The color samples are measured using ACS color-matching instrument under different light sources (D65, A and TL84), then based on color science and visual psychophysics methods, establishing the datasets of the samples’ color appearance. We use mathematical statistics methods, correlation coefficient R2 and coefficient of variation CV, to evaluate the visual prediction results in the color atlas, under three different light sources.

With the development of the concept of Industry 4.0, research relating to robots is being paid more and more attention, among which the humanoid robot is a very important research topic. The humanoid robot is a robot with a bipedal mechanism. Due to the physical mechanism, humanoid robots can maneuver more easily in complex terrains, such as going up and down the stairs. However, humanoid robots often fall from imbalance. Whether or not the robot can stand up on its own after a fall is a key research issue. However, the often used method of hand tuning to allow robots to stand on its own is very inefficient. In order to solve the above problems, this paper proposes an automatic learning system based on Particle Swarm Optimization (PSO). This system allows the robot to learn how to achieve the motion of rebalancing after a fall. To allow the robot to have the capability of object recognition, this paper also applies the Convolutional Neural Network (CNN) to let the robot perform image recognition and successfully distinguish between 10 types of objects. The effectiveness and feasibility of the motion learning algorithm and the CNN based image classification for vision system proposed in this paper has been confirmed in the experimental results.

Much progress has been made in the field of digital imaging technology with the widespread use of CMOS and CCD devices and the high computing power of IC chips. However, there still are many technological problems to overcome to achieve lightweight and affordable high density 3D lossless imaging for both industrial and scientific applications. This paper will focus on the latest development of freeform optical technology and powerful image analysis techniques to develop a compound eye camera system so as to extract the 3D relative position coordinates of the sub-object in the field of view, and to compensate the shadow effect in the images caused by the non-uniform illumination to obtain high quality lossless images. Such a novel compound eye system will find applications in delivering 3D image of digital camera, increasing capability of scientific profiling measuring instrument, and lowering cost and enhancing resolution of wide-field, video-rate, gigapixel cameras such as the AWARE designed by Duke University.The main objective of this paper is to design a novel compound eye system from high definition 3D lossless digital imaging technology to achieve the realistic three-dimensional positional information of the object in both near and far fields. In the imaging process, the ultimate imaging quality is affected and limited by the resolution of the detector, the field of view, wavefront aberration, resolution of the optical imaging system and the ambient light.The novel compound eye imaging system includes two modules: one is the optical imaging module with compound eye optical elements with other secondary optical lenses, while the other one is the image analysis module. By combining the above optical hardware with the image analysis software, the relative position information of the sub-objects in the picture and the lighting condition of the environment can be deduced accurately.

The paper describes an experimental determination of kinematic quantities using high-speed digital image correlation system. It deals with the analysis of minimal sampling frequency needed for correctly performed high-speed correlation analysis. The authors also describe the influence of cameras sampling frequency on the quality of the obtained results. Mentioned analysis was performed using a rotational object. For the purposes of the results comparison the data obtained from correlation system in a form of displacements in three mutually perpendicular directions were processed in Matlab and in the paper are depicted in a form of graphical visualizations.

It is important to obtain the 3d coordinate in the field of measuring. How accurate, fast, convenient to obtain the 3d coordinate affects the accuracy and reliability of measurement directly. Through studying the basic theories of machine vision this paper focus on the study of a stereo vision measurement model based on the intersecting axis. In view of the parameters in the model of binocular stereo vision, this paper uses the Zhang Zhengyou calibration method to calibrate the system of Stereo vision. In order to test the accuracy of the system, this paper measures the distance of two standard circular. Finally, the machining experiment validates the proposed method.

Golf has become an increasingly popular sport. Determination of motion parameters of golfball can give quantitative suggestion for players. This paper introduces the principles of determinationbased on Doppler radar. The parameters of golf motion to be measured include velocity, angle, spinangular velocity and position. A golf realtime tracking system is designed with one transmitting antennaand three receiving antennas in this paper. Simulations validate the feasibility and high accuracyof this system. This system makes referential significance for golf realtime tracking system.

This paper analysis the characteristics of High Spatial Resolution Remote Sensing Image (HSRRSI), consider the drawback of pixel-wise Classification, an Object-based method was taken to classify the image. The paper firstly preprocess and coarsely segment the image, secondly optimize the segmented results for deeply processing, then extracting the image feature from the original image and segmented results, finally classify the image and assess the results.

In tracking object by a camera, the jolting of the vehicle will increase the deviation of the skewing value of the air object in the image. a method of image stability based on digital inclinometer is presented in this paper. The incline angles caused by the jolting of the vehicle are measured by the angle sensor, and the incline angles are used to eliminate the object image deviation caused by jolting and to stabilize the object image. The method for acquiring the height of object by a single camera is also discussed in the paper during the process of extracting the image deviation. The extraction errors of the deviation are analyzed and simulated; the simulation results show the efficiency of the method.

The paper analysis mean filter, median filter and morphology filter. The paper studied order morphology theory and its filter character. Based on this, the paper proposed multi-scale and multi-structure filter algorithm based on order morphology. The result of simulation shows that the algorithm can not only keep the edge and detail of medical image but also can enhance the contrast of recovery image.

Stereo matching methods are widely used in computer vision and stereo reconstruction, from the perspective of improving the matching accuracy, this paper focuses on the global optimization algorithm. An improved Belief Propagation method is proposed in this paper, by involving more pixels into information transmission, our method improves the accuracy ofstereo matching. The experimental results verify the efficiencyand reliability of our method.

In this paper ,a color printing defect automatic online detection method based on digital image processing technique is proposed. The main idea of this method is comparison of defect product and template and it makes up of following key models. Firstly, multi-scale segmentation is applied to composed image which is overlaid by detecting product and template image. Secondly, an automatic region similarity analysis calculation is taken to segmentation obtained in multi-scale segmentation. The color difference between detecting product and template can be calculated accurately. Thirdly, defect detection results can be obtained according to threshold segmentation. Finally, the characteristics and advantages are approved by experimental analysis and discussion. Algorithm parameters are adjusted and modified to improve the stability and effectiveness. Experimental results approve that color printing defect automatic detection method in this paper has the characteristics of effectiveness and applicability. And experimental results indicate that this method has the advantage of judging the defect types automatically.

Synthetic Aperture Radar (SAR) has become one of the important means for the ocean remote sensing detection of oil spills. The existing SAR image segmentation method has the issues of edge blur, poor contrast, non-uniform intensity image, so the segmentation effect is not ideal. This paper presents a variational level set SAR image of oil spill detection method based on fuzzy clustering. First of all, apply the threshold method on initial segmentation of the original SAR image to transform the initial segmented image as fuzzy clustering. Secondly, introduce the clustering results into the initial level set function to achieve the initial contour. Finally, add fuzzy clustering model in the level set energy function to complete the level set evolution process and get the final segmented image. This paper uses the threshold segmentation results to achieve the initialization of the variational level set function profile. In theory, it could improve the level set method for efficiency, and reduce the wrong segmentation phenomenon. The experimental results show that the SAR image segmentation method of oil spill has good segmentation qualities and is suitable for the edge complex image segmentation.

Gesture has different speed in the process of movement. To reflect the different speed of the user, this paper presents a gesture speed estimation method. Firstly, we use data glove and camera to establish the relation between the variation of gesture contour and that of gesture speed. Secondly, we build the gesture speed estimation model by stages. Finally, we get the real-time speed of hand motion through this model and complete interactive task. The main innovation of this paper is that we reveal the relation between the gesture contour and speed to lay the foundation for further capture the interaction of user intention. Experimental results indicate that the time cost of our method decreased by 31% compared with freehand tracking based on behavioral models and the 3D interactive system based on our model is of high the user experience.

Traffic flow video processing is an important direction of the development of intelligent transportation, Gaussian mixture background modeling is one of commonly algorithms to build background modeling that is widely used in image processing field. In order to analysis the algorithm comprehensively ,in this paper, it calculus on formula as well as tests and compares some traffic flow videos, the solution based on the K-means, EM parameters model are presented, it gets the mathematical meaning and evolution process of the algorithm clearly by calculating formulas to provide theoretical guidance for practical application; the traffic flow videos processing can get the treatment ability of this algorithm according to the different situation of videos, analyze the mixed Gauss background modeling for time migration, the light changing, moving object shadow processing ability. In this paper, the algorithm of mixed Gauss background modeling is analyzed from two aspects of theory and application.

With the improvement of computer hardware and software performance, digital multimedia technology has been applied to various fields. The visual effect of ceramic art image is affected mainly by the brightness, texture and color, in which the image defects will affect the visual effect of the ceramic product. This paper designs the digital processing system of a new ceramic art image processing, the system can repair defects in ceramic art image. In order to verify the validity and reliability of digital processing system, the lead-free perovskite structure BZT based ceramics is prepared in this paper, and the use of electron microscope and digital camera take the ceramic products processing image with artistic defects, and then using VB programming digital image processing technology carries out color re coating for the ceramic, the ceramic products electron microscopy images after restoration will be obtained by the image boundary restructuring, which provides a new computer method for the processing art of the ceramic products.

With the development of computer graphics, real-time rendering-based VF: technology has been applied in more and more fields. LOD is the key technology in large-scale terrain rendering. In this paper, the basic concept of LOD is introduced briefly and some algorithms of LOD in use are mentioned and analyzed; secondly as one of algorithms of LOD, View-Dependent Progressive Mesh algorithm is studied and improved, the result of implementing the large-scale terrain’s LOD by using VDPM is presented. There are key technologies in LOD Large-scale terrain real-time rendering are researched. Relative technologies are presented such as: LOD of the terrain, visibility culling, and cracks eliminate, view-dependent refine, LOD error, technologies of texture etc. Using LOD technology, VR system can greatly reduce the; number of polygons produced in real-time rendering procedure. Finally, we do experimental design work based on the methods and techniques presented by this paper.

This paper presents a novel approach to detect unattended and removed objects from a single fixed camera video for surveillance applications. In this paper two backgrounds rather than any tracking information are employed to detect static foregrounds. Each of the backgrounds is modeled by three Gaussian mixtures. We make a subtraction between two foregrounds extracted by the background subtraction method to initially determine the static regions. Then the region-level density analysis and texture information are combined to remove false static foreground regions caused by illumination changes and objects that are in the static-moving transitions. Finally, classification of the unattended and removed is determined by a method that compares color histograms of the static foreground zone and its extended areas in current frame. Experimental results show that our method can work well in simple scenario as well as complex environments with an unsmooth background.

Image fusion algorithm is very important in image fusion process. Image fusion algorithm based on pyramid decomposition was reviewed in this paper. Pyramid decomposition algorithm mainly includes Contrast pyramid, Gradient Pyramid, Laplacian Pyramid and Ratio Pyramid. The fusion algorithms based on pyramid decomposition were respectively applied in multi-focus images, advantage and disadvantage were summarized and application was given. Fusion results were given by MATLAB simulation. Objective evaluation index including of mean, standard deviation, entropy and average gradient was calculated in this paper. Image fusion algorithm should be selected according to the information extracted and the aim of fusion.

This paper aims at the difficulty that lack of observation model and high-dimensional sampling in video tooning, proposes a method based on key frame matching and dual-directional Markov chain Monte Carlo sampling of video motion redirection. At first, after extracting the key frame of a given video, By affine transformation and linear superposition, the subject initializes the video’s space-time parameters and forms the observation model; Secondly, in each space-time, based on the bi-directional Markov property of each frame, This paper proposed a dual-directional Markov chain Monte Carlo sampling particle filter structure and takes full advantage of the relationship of the front and back frame of the parameters to estimate motion redirection parameters. At the same time, for high-dimensional sampling problem, the subject according to the directional parameters’ correlation implements classification of skeleton parameters-morphological parameters-physical parameters, proposes a hierarchical genetic strategy to optimize the output parameters and improves the efficiency of the algorithm. The research of this paper will produce an efficient and prominent animation expressive video motion redirection method and play an important role on video animation of the development.

This paper presents a 3D gesture recognition technology based on machine vision as the center. Based on a large number of experiments, this paper sums up and introduces the existing gesture recognition technology, the key research contents of gesture recognition, as well as the history of development of gesture recognition technology. Then, the paper does research in the main technology of gesture recognition .The experimental results show that the method can realize 3D gesture recognition in video sequences with real-time and stability, even more; it can get better recognition result.

The paper studies the dancing gestures evaluation method based on images features comparison. The current methods for dancing gestures are according to the people’s vision, which are usually not accurate. Once the dancing is more complex, the conventional methods have the defects of low accuracy. Therefore, this paper proposes an optimized algorithm based on image features comparison for dancing gestures evaluation in which the dancing gestures images are collected. The features in the dancing images are extracted and compared based on the Scale Invariant Feature Transform algorithm (SIFT). The experimental results show that the improved algorithm can increase the accuracy for dancing gestures evaluation greatly.

The gray image can be viewed as a 3D terrain consisting of intensity waves. The Wave Equalization introduced in the paper is to weaken the influence of uneven illumination on the gray image binarization. The 2D wave is decomposed into several 1D waves corresponding to different directions for 1D equalization. Then PCA algorithm is used to compress all 1D results to the final image with uniform illumination. The extensive experiments in the paper had proved that our method has excellent performance and adaptability for various uneven illumination environments.

According to the array size,underwater positioning system can be divided into long baseline array, short baseline array and Ultra-short baseline array (USBL). This paper introduces USBL space positioning principle, describes the essence of direction estimation based on the phase measurement , analyses the factors that affect the positioning error, in view of the phase error, this paper introduces the method of improving positioning accuracy from extension formation.

In this paper, frame differential method will be used to detect moving targets in a static background video file, and pre-contour can be obtained by binarizing the detected targets . However,, the result are not what was expected, so the expansion and corrosion of mathematical morphology are used to extract the final contour of moving targets. in the progress of dealing with massive data , mathematical morphology is not good enough to achieve the need of the real-time in video surveillance. Considering the dilation and erosion is a kind of parallel processing operations, in order to improve the speed of mathematical morphology operations, this paper offers detailed implementation process of the dilation algorithm for parallel computing on GPU. Experimental results showed that GPU parallel processing on mathematical morphology algorithm faster than the CPU serial processing.

In this paper, the average reprojection geometrical errors on all images of a spatial straight line is taken as the restructuring and optimization target function to ensure the optimal result to be acquired; a spatial straight line is expressed by use of the Plücker coordinates for the proposal of the analytical method of correcting the double linear restriction on the Plücker coordinates of noise-included spatial straight lines; during the optimization process, to ensure parameters that meet double linear restriction, the iteration renewal process is expressed by at least 4 parameters, to increase the precision of restructuring results. Experiments results derived from simulation data and real images have all demonstrated the high efficiency and precision of the algorithm proposed in this paper.

In order to improve the accuracy of recognition system for fatigue facial expression of driver, driver fatigue expression of this paper, the detection method for key feature points in the fatigue facial expression image of driver is applied in the paper to establish a fatigue expression image recognition model based on attention mechanism. Experimental results show that the algorithm can improve the recognition rate of driver's expression image, so as to record fatigue expression image of driver more accurate.

The objective of this paper is to build a system for pedestrian detection in an outdoor environment. The contribution of this paper is the detection method that integrates the silhouette edge information with patterns of motion in each frame of the image sequence. The test results show that the high-precision, good adaptability, and real-time performance of the proposed method.

This paper first introduced the geometric correction method for remote sensing image used in the software, put forward the method precision visualization comprehensively and systematically describes the quality of remote sensing image rectification. This paper illustrates an instance method and its application through a IKONOS image rectification,and some useful conclusions are obtained. Keywords: Remote sensing image; image rectification; accuracy; spatial interpolation; error

Aiming at the shortage of SIFT algorithm in time comsumption and getting less matching points. On the one hand, the paper improves the original SIFT algorithm, it proposes regional growth algorithm based on SIFT, so you can get many matching points which are good for generating the disparity map; on the other hand, the paper uses the CPU and GPU heterogeneous platforms and analyses the CUDA programming model and memory model. This paper analyses the algorithm in detail, so the algorithm can be carried out on CUDA. Experimental results show that, compared with the original algorithm, the algorithm is about 10 times faster, and generate good disparity map.

Due to the development of image segmentation and reconstruction technology, it provides a larger space for development of the image enhancement technology. Under the promoting of the image processing calculation, the capture and recognition function of digital X-ray photography technology are stronger, and the image processing precision is higher. Based on the variation principle, this paper uses the function approximation to improve the X-ray photography image processing technology, and obtains the new boundary value reconstruction condition of X-ray photography. In order to verify the effectiveness and reliability of the mathematical model and algorithm of the boundary value reconstruction, this paper uses MATLAB software and C language to debug the algorithm, and realizes the digital color rendering for the images at the terminal of track and field, obtains the image reconstruction algorithm under different boundary values. It provides a new computer method for the research on image enhancement technology.

This paper presents an algorithm for hiding information’s in raw video steams using art of steganography using discrete wavelet transform. While mostly applied to still images in the past, it has become very popular for video streams recently. When steganographic methods are applied to digital video streams, the selection of target pixels, which are used to store the secret data, is especially crucial for an effective and successful-embedding process; if pixels are not selected carefully, undesired spatial and temporal perception problems occur in the stegno-video. Typically, an irrecoverable steganography algorithm is the algorithm that makes it hard for malicious third parties to discover how it works and how to recover the secret data out of the carrier file. In this paper, a new embedding algorithm is proposed to hide the secret data in moving videos. The 2D-DCT of the video is taken and the secret message is embedded. The performance measures are evaluated for the quality of the video after the data hiding and show good results.

Compressed sensing is referred to as the CS technology; it can realize image compression and reconstruction process in low sample rate. It has great potential to reduce the sampling rate and improve the quality of image processing. In this paper, we introduce the structure prior model into the compressed sensing and image processing, and make the image reconstruction of high dimensional optimization process simplified into a series of low dimensional optimization process, which improves the processing speed and image quality. In order to verify the effectiveness and reliability of the proposed algorithm, this paper uses combined control form of C language and MATLAB software to design the programming of structure prior model, and use the Simulink environment to debug the program. Through the calculation we get the image block and the reconstruction result. It provides the technical reference for the research on image compressed sensing technology.

This paper designed and implemented a desktop application of safety lock based on face recognition. The application is developed by C# and used open source visual library OpenCV and SQLite lightweight database. Principal component analysis (PCA) and Rijndael algorithm are respectively used in face recognition and algorithm of encryption and decryption. This paper illustrates design and implementation in detail

Nowadays with the blooming development of new media and cultural creative industries, the video animation has drawn so many researchers’ attention due to animation production efficiency and animation expression. This paper aims at the difficulty that lack of observation model, proposes a method of video time-space body filtering based on dual-directional Markov chain Monte Carlo particle filter for video motion redirection. At first, after extracting the key frame of a given video, by affine transformation and linear superposition, the subject initializes the video’s time-space parameters and forms the observation model; Secondly, in each time-space body, based on the bi-directional Markov property of each frame, this paper proposed a dual-directional Markov chain Monte Carlo sampling particle filter structure and takes full advantage of the relationship of the front and back frame of the parameters to estimate motion redirection parameters. The research of this paper will produce an efficient and prominent animation expressive video motion redirection method and play an important role on video animation of the development.

Genetic algorithm is a search algorithm based on genetic mechanism and natural selection. It has been widely applied to research fields including image processing field. The paper improves standard genetic algorithm and improves the arithmetic speed of the algorithm, which achieves better image restoration effect. And the paper compares the image restoration quality of traditional algorithm, standard genetic algorithm and improved genetic algorithm to prove the feasibility of applying genetic algorithm to image restoration.

A biometric identification system may be a pc application for mechanically distinctive or confirmative of an individual from a digital image or a video frame from a video supply. One in all the ways that to try and do this can be by examination designated face expression from the image and a facial information. This paper planned dynamic face recognition from near-infrared images by exploitation sparse representation classifier. Most of the prevailing datasets for facial expressions are captured in a very visible light spectrum. However, the visible light (VIS) will modify with time and placement, causing important variations in look and texture. This new framework was designed to attain strength to pose variation and occlusion and to resolve uncontrolled environmental illumination for reliable biometric identification. This paper gift a unique analysis on a dynamic facial features recognition, exploitation near-infrared (NIR) datasets and LBP(Local binary patterns) feature descriptors. It shows sensible and strong results against illumination variations by exploitation infrared imaging system.

Disparity estimation is a basic analytic method for images and video and is highly used in fields like stereo image coding, video retrieval and stereo image segmentation. This paper proposes a new disparity estimation method. In this paper, the superiority of the multi-block matching is analyzed. The new method based on the idea of block matching constructs a multi-block matching algorithm, thereby improving matching accuracy especially in the disparity discontinuous regions. A adaptive window method based on edge detect is proposed to select the best window.The experimental results based on the Middlebury stereo datasets demonstrate that our method can achieve a better performance.

The purpose of this research is to improve the landslide mapping using unmanned aerial vehicle (UAV) for the area of slope displacement. It further presents the UAV namely multi-motor that being used to capture images at the research area located in Parit, Perak. The objective of this research paper is to develop a three dimensional of landslide area produced from the UAV images. For the whole process of image processing, thirty six control points are established using global positioning system (GPS) staic mehtod using Agisoft Photoscan. The results show that the digital elevation model (DEM), aspect Model, slope model, and digital orthophoto can be obtained using the procedure and method used in the study. The information is obtained through accurate assessment results and used to create a 3D model which is then used to monitor technique applications. The restitution stereo model is also by three dimensional rotations or transformation in 3D surface. From here, the landslide can be detected by calculation of three difference epoch data achieved from Digital Elevation Model (DEM) generation. Prior to that, this paper focuses on the monitoring of that area based on DEM area and volume generated from 3D surface analysis. To conclude this study shows that UAV can be used for producing digital map.

The Mean-Shift algorithm has very good tracking effect when the background is in a simple; but for a complex environment, tracking effect is not very ideal. Therefore, a new gray feature modeling method is proposed in this paper. Firstly, target in the tracking window is uniformly divided into even pieces. Then the pixel gray value of each block is calculated with subtraction of certain rules. Finally, the gray value of gray difference and the whole object value fusion are fused and established the object model. The object model that established not only contains the whole gray value information, but also contains the gray value differences between regions, has a more accurate description of the target, and then distinguish target from background better. The experiment results show that: the target model using the method in this paper to track based on the Mean-Shift algorithm, has good adaptability when the target is partially occluded and has better robustness for complex background.

In this paper, there is an existing map-matching algorithm on the lack of accuracy, combined with the specific short-message communication function which developed by China's own BeiDou navigation satellites, and then there is proposed a new map matching algorithm for vehicle based on BeiDou Short-Message communication. The algorithm makes up for the shortcomings of existing map matching algorithms, and effectively reduce the matching error, especially for city complex sections, including the main and auxiliary road, the viaduct etc. Finally, this paper validates by comparing the new algorithm with the mature matching algorithm based on fuzzy logic. After verification, it proves that the proposed algorithm can improve the precision of map matching, especially for city complex sections of the map, and the effect is more significant.

The traditional virtral target aiming training uses indoor large screen projection more. Recently, the system also uses the virtual reality scene as background. According to the situation, using real-time video frame as the virtual target aiming training background for the authenticity of the virtual target aiming training scene is presented in this paper. This method fully shows the maturity of modern video acquisition technology and the innovation of virtual reality scene. This paper prensents using DirectShow tools for video frame acquisition and taking the advantages of OpenFight API for the model built by military simulation modeling software MultiGen Creator development based on OpenGL development platform and completes the building of virtual target aiming training scene based on real-time three-dimensional video scene and the consistency of virtual target and scene.

Image segmentation technique was used widely for computer vision and image processing. A robust technique of image segmentation plays a crucial role in identification problem. In this paper, a nonparametric and unsupervised method of automatic threshold for segmenting image was proposed, i.e. the optimal threshold is approximated by global average gray and local average gray, and this method was compared with other methods by using standard image. The experimental results show that our method proposed in this paper is robust. In addition, an image database of road traffic marking (www.ananth.in/RoadMarkingdetection.html) is provided to do this experiment for testing our method, the results show that our method is excellent.

This paper proposes the re-built human body movement model with multiple cameras. In the tracking frame of the non-linear optimization strategy, the paper builds the body dynamic model to dynamically simulate the human movement which effectively solves the issues of the body parts overlap and tracking errors accumulate. Compared with traditional methods, the required equipment is very economic and the matching accuracy of the algorithm is quite high. The paper applies the athletes as the experimental examples which illustrate the proposed algorithm can effectively increase the 3D image tracking matching accuracy in dynamic videos as the analysis basis.

Currently, most mapping researches are deeply dependent on long range sensors such as sonar, laser range finders or cameras, but due to power or cost constraints, which often cannot be mounted in the sensor module. In response to the no long range sensors situation, this paper presents a method to complete indoor mapping tasks in real-time only using short range sensors installed on the multiple micro-robots. The key mapping problem needed to address in the paper is that each single robot created a map based on odometer and short distance sensor data, and then maps from multiple robots were consolidated to generate a greater coherence map. A kind of SRV mobile robots were tested in an artificial rectangular environment. The experimental results show that the multiple micro-robot system has good robustness to odometer and environmental noise.

To obtain the characteristics of rock is one of the most significant methods in reconnaissance in order to make the project camouflaged. This paper, which is based on the former study on digital image processing technique, is mainly about the way to acquire the general and detailed information, such as geographic distribution and specific texture of rock. The technique put forward in the paper is fundamental to carry out reconnaissance, design the project and evaluate the camouflage effect.

The paper mainly discusses the analysis method for the psychological impact of computer vision noising technology. Actually, people's psychological acceptance and corresponding memory capacity of computer vision images with lots of noise are relatively poor. The de-noising process to computer vision images can improve the clarity, thus generating passive psychological impact. Therefore, the paper proposes a spatial domain filtering algorithm-based de-noising method for computer vision. It establishes wavelet packet decomposition tree for computer vision images and de-noises accordance with the decomposing results. The experiment results show that the proposed de-noising method has passive psychological influence and improves the memory capacity of computer vision images.

UAV Video is rapidly emerging as a widely used source of imagery for many applications in recent years. This paper presents our research on the mosaic of UAV video for the purpose of harbor surveillance. First, one new framework on estimating video frame transformation with Optical flow is presented in this paper. For this new framework, fewer number of Gaussian pyramid is created for implementing for the multiresolution approach and thus, more details for optical flow computation is well kept; Second, we make a discussion on using Fourier-Mellin Transformation in image frequency domain to estimate initial motion parameter of adjacent video frames and with those initial motion parameters, small displacements for optical flow computation can be achieved; The experimental results demonstrated that the mosaic image generated from aerial video shows satisfied visual quality and its surveillance application for fast response to time-critical event, e.g., flood, is descried.

This paper researches on the detection and tracking technology of ship intrusion and remaining based on the special surveillance video scenes of sea. A ship detection and tracking method that combines the adaptive background subtraction with centroid algorithm is proposed. Compared with the method of edge orientation histogram, the method proposed in this paper shows a lower computational complexity and a better real-time performance. The experiment demonstrate that this method has a high degree of accuracy. And it makes a realization of detecting and tracking of ship intrusion and remaining within the surveillance sea.

This paper proposed an automatic exposure algorithm based on luminance histogram and region segmentation. The method ensures well exposure to the main object under the backlighting or excessive front-lighting conditions, and uses the characteristics of the luminance histogram to judge the scene condition. According to the luminance histogram, if the image under normal light conditions, the method compares the average luminance value of the image with the corresponding reference target brightness value to adjust exposure compensation. If the image under special conditions, the average brightness value will be calculated using fixed-partition method and exposure compensation will be adjusted. The paper combines auto-exposure and auto-gain control to adjust the exposure compensation. As it turns out, this method in a variety of light conditions can achieve well exposure.