Table of Contents

31 October 2017, Volume 38 Issue 5 Previous Issue    Next Issue

---

For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Select

Review of Rain Removal Techniques in Videos and Images ZHOU Pucheng 1,2,ZHOU Yuan 1,HAN Yusheng 1,2 2017, 38(5): 629-646.  DOI: 10.11996/JG.j.2095-302X.2017050629 Abstract ( 352 )   PDF (6659KB) ( 1391 )   Removal of rain in video and images is an important issue that interests both digital image processing and computer vision areas, and it has been extensively used in many applications such as video surveillance, automatic transmission, and military reconnaissance. The essential goal of rain removal in video and images is how to detect raindrops or streaks and remove them, and restore the details of the scenes. In order to make more researchers have a more comprehensive understanding in rain removal techniques in video and images, a detailed introduction was reviewed. On the basis of extensive literature research, the basic ideas, methods, characteristics and effect of various kinds of rain removal techniques in videos and images are summarized and compared. According to the implementation ways and disposal methods, rain removal techniques in videos and images are classified into four types, they are imaging parametric-based method, time domain-based method, frequency domain-based method, and sparse domain-based method. At the same time, the analysis of recent years in time domain, frequency domain and sparse domain to the latest developments in the rain is focused. With the improvement of computer hardware performance and the deep research of raindrop characteristics, the method research and application conditions for rain removal in videos 630 前沿探索与综述 2017 年 and images have made great progress. However, due to the interference in rainy dynamic complexity and/or the environment with complex background, rain removal techniques in videos and images are still far from perfect, there are still many open problems that need to be solved. Related Articles | Metrics

Select

Mixed Wave Rendering in the Visual System of a Maritime Simulator CHEN Lining, QIU Wenqin 2017, 38(5): 647-653.  DOI: 10.11996/JG.j.2095-302X.2017050647 Abstract ( 73 )   PDF (2000KB) ( 205 )   Mixed wave is real-time rendered in the visual system of maritime simulator. Inverse fast Fourier transform (IFFT) Gerstner wave model is used to compute the motion of ocean wave. Six parameter spectrum, a frequency spectrum showing energy distribution of mixed wave, is applied to construct the wave number spectrum of mixed wave. The wave number spectrum is contained by the Fourier coefficients of height field and choppy wave. By IFFT, the position of mixed wave surface is gotten. Further, mixed wave is rendered. The rendered mixed wave is affected by the low frequency wave with long wave length. Its shape is smooth and flat, differs from that of wind wave, but is similar with the actual ocean wave. The presented method of mixed wave rendering is used in the visual system of the maritime simulator, and the frame rate meets the request of real-time rendering. Related Articles | Metrics

Select

Research on Generation of Activate Joint Based on 3D-Printable Model GAO Peng 1,LI Lin 1,CHU Xiaoli 1,TAN Jianbo 1,2 2017, 38(5): 654-662.  DOI: 10.11996/JG.j.2095-302X.2017050654 Abstract ( 250 )   PDF (4927KB) ( 424 )   3D model can’t be afforded to 3D printer an activate structure, because it is generally closed mesh which is static. Aimed to creating a printable activate model for non-professional users, this paper studies common parameters design of the active joints and the solving of parameters, while gets the limit of the threshold parameters that would influence the performance of activity. On this basis, an easy to use method that can add active joint to static model is proposed, which can generator some doll’s models with moved limbs and functional models for daily use. The results of 3D printing demonstrate the joints originated from the design of this paper are enrich and valid. The method creating active model is quickly and flexible, and has been accepted by most users. Related Articles | Metrics

Select

Rolling Bearing Fault Diagnosis Based on Improved EEMD and Spectrum Kurtosis MA Zengqiang, ZHANG Junjia, WANG mengqi, RUAN Wanying 2017, 38(5): 663-669.  DOI: 10.11996/JG.j.2095-302X.2017050663 Abstract ( 76 )   PDF (926KB) ( 183 )   In order to solve the problems that the early fault vibration signals of rolling bearing affected by the noise, the parameters of ensemble empirical mode decomposition (EEMD) are not easy to be obtained. A method for fault bearings of rolling bearings based on improved EEMD and spectrum kurtosis (SK) was presented. Firstly, obtaining the amplitude coefficient of added noise by extracting the high frequency information, the number of ensemble members is obtained by calculating the expectation error, the fault signal was decomposed into several intrinsic mode function (IMF) by improved EEMD, the IMFs were reconstructed based on kurtosis criterion. Then, the central frequency and bandwidth of a band-pass filter were determined with spectral kurtosis. Last, the filtered signal was analyzed by using energy operator demodulation spectrum. The results demonstrate that the proposed method can not only solve the problems such as losing fault information and leaving noise due to the mode mixing in the process of denoising, but also extract the fault frequency accurately. Related Articles | Metrics

Select

Gait Analysis and Simulation of Quadruped Robot ZHANG Ben 1,BIAN Xingao 2,ZHU Denglin 2 2017, 38(5): 670-674.  DOI: 10.11996/JG.j.2095-302X.2017050670 Abstract ( 119 )   PDF (759KB) ( 218 )   Aiming at achieving the stable walking of a hydraulic driving quadruped robot, according to the principle of stability margin, a linear and turning gait planning for JQRI00 quadruped robot were researched. The virtual prototype of JQRI00 quadruped robot were established according to the structure characteristics of quadruped robot, and the linear gait simulation of virtual prototype was carried out by using ADAMS. The simulation results show that the gait planning is correct and feasible, which achieves the stable walking and lays a foundation for the control of the robot in the later stage. Related Articles | Metrics

Select

Research on Design Rules of Suitcases Using Rough Sets Algorithm on User Orientation WANG Hanyou 1,2 2017, 38(5): 675-682.  DOI: 10.11996/JG.j.2095-302X.2017050675 Abstract ( 72 )   PDF (373KB) ( 232 )   This paper seeks to apply rough sets algorithm to improve the effectiveness and efficiency in customer satisfaction survey, identify condition attributions that affect product marketing by data mining on raw data acquired through questionnaire surveys, extract decision rules of these attributions combination, provide an innovative research approach on user driven design method and market exploitation. The research is to complete the combination, reduction of the condition attributions and calculation of their core, conduct induction and screening of decision rules and significance computing of attributions, and to obtain design rules of suitcases with the questionnaire survey data for well-known enterprise on product objects, reveal that there are only six key condition attributions affect effectively the product market performance, it is necessary to according to the design rules that the dependence relationship is objective existence between discrepant level properties and market performance to implement R&D and marketing policy as a whole. Suitcase design should return to its origin, pay more attention to overall effect of product, unique and novelty of appearance, selection of high-quality fabrics, and should not focus too much mind and emotions on slick function innovation and extra finishing details. Related Articles | Metrics

Select

Study on the Aesthetic Sense of the Miao Clothing Patterns Based on Psychological and Physiological Measurement YAN Longhua, LIN Li, XIAO Hualiang, SHI Lei 2017, 38(5): 683-688.  DOI: 10.11996/JG.j.2095-302X.2017050683 Abstract ( 88 )   PDF (1338KB) ( 229 )   In order to provide the effective reference for the better application of the Miao clothing patterns to the modern design, making better protection, inheritance and exploration of this intangible cultural heritage, Miao clothing patterns were taken as the research object and 20 masters with design professional background were tests, using comprehensive research methods of psychological measurement combined with physiological measurement, the eye movement characteristics and corresponding aesthetics of different patterns of tests were discussed in depth. Research results showed that the such patterns with dynamic lines, abundant composition and novel visual image have a higher aesthetic feeling. The quantitative research approach can improve the accuracy of the study on aesthetic feeling of Miao clothing patterns and provide a reference for the protection, inheritance, development and utilization of the intangible cultural heritage in visual recognition category. Related Articles | Metrics

Select

Product Form Grouping Design Based on Users Requirements Subdivision HE Xuemei 1,CAO Tinglei 1,SONG Ning 1,JI Xiaomin 2 2017, 38(5): 689-694.  DOI: 10.11996/JG.j.2095-302X.2017050689 Abstract ( 114 )   PDF (887KB) ( 273 )   Based on a variety of design goals for pattern prototypes of specific product, the concepts of user requirements subdivision table and product form grouping were presented, defined and expressed. The user requirements subdivision table act as a template to obtain user requirements subdivision information while the KANO model determines the types of it and the corresponding weights. On this basis, the user demand subdivision information grouping, namely the basic requirements group, the performance requirements group, the attractive requirements group are completed. Based on the user requirements subdivision information group, the three conceptual models of product pattern grouping are defined, and the process model of product form grouping is given. Finally, the electric kettle is used as an example to verify the feasibility of this method and process. Related Articles | Metrics

Select

Dimension Spanned, Emotional Internalization#br# —The Application of Chinese Elements in Product Design HOU Shijiang, CHEN Yuzhi 2017, 38(5): 695-699.  DOI: 10.11996/JG.j.2095-302X.2017050695 Abstract ( 81 )   PDF (509KB) ( 220 )   Characters are one of the necessary tools to realize the information transmission and express the emotion. In numerous types of characters, Chinese characters contain unique cultural charm and aesthetic value. With five thousand years of heritage and development, Chinese characters develop the special proportion and form. In addition to being able to be applied in the field of visual communication, Chinese character elements are able to be applied in product design to express thoughts and feelings. Based on the analysis of morphological characteristics of Chinese characters, this paper points out the characters and principles in the form of emotional expression, accounts for the application of Chinese elements on the concrete methods of product design and verifies that the form of Chinese characters, the symbolic characteristics and thoughts connotation that apply to product design with practical examples. Therefore, it is a kind of important design method for the user to produce emotional resonance. Related Articles | Metrics

Select

Adaptive Color-Style-Aware Image Cloning CHENG Haonan 1,WANG Kai 1,LIU Shiguang 1,2 2017, 38(5): 700-705.  DOI: 10.11996/JG.j.2095-302X.2017050700 Abstract ( 73 )   PDF (4130KB) ( 347 )   Image cloning is a process that makes the clone selection in the source image seamlessly merged with the target image, aiming to generate a new image which appears natural and realistic. While the traditional image cloning technology may introduce significant color changes in the inserted elements. To solve the problem, we propose an efficient color-adjustment algorithm. More specifically, firstly, our approach uses the grayworld normalization (GN) model to prejudge the color style of the target image and get the illumination judge parameter  . Then, we get the composite image with color control by a translational transform in the gamut space of the whole cloned region with different distances which are designed based on the illumination judge parameter  . Finally, we set the color distance and spatial distance as restrictions to optimize the boundaries of cloning constituency after translation and generate the seamless result image. The proposed algorithm can be adapt to images in different color styles with color control and avoid manual adjustment of the parameters in the conventional methods. Various experiments verify the effectiveness of our method. Related Articles | Metrics

Select

The Style Transfer of HDR Image Based on Dictionary Learning XIE Zhifeng 1,DU Sheng 1,GUO Yuchen 1,HUANG Dongjin 1,MA Lizhuang 2 2017, 38(5): 706-714.  DOI: 10.11996/JG.j.2095-302X.2017050706 Abstract ( 104 )   PDF (11938KB) ( 227 )   The style of HDR image includes natural, reliefs, drawings, paintings and so on. It shows the rich artistic qualities, has the characteristics of remarkable style, rich in detail, full color and others. However, the conventional method for generating HDR style is tedious, time consuming and the effect of HDR is instability. Therefore, in order to simplify the interaction and enhance the effect of the result, we proposed a style transfer method for HDR image by giving an HDR reference image and using color transfer and dictionary learning technology to transfer the features of HDR photograph to the source image so that to generate the HDR image automatically. First, with the help of gradient-preserving color transfer technology, the color feature of the HDR reference image was transferred to the source photo. Then, extracting the detail features of the HDR reference image and using K_SVD algorithm to train a dictionary to form an over-complete dictionary set of the details. Next, extracting the detail features of source image and using the dictionary set to sparse reconstruction and then generating the right HDR image with detail features. Finally, combining the result of color transfer with the sparse reconstruction result of the source photo to generate a new style of HDR image. The experiment results show that the method applied to a variety of HDR styles can obtain the same visual effects as the reference image. Related Articles | Metrics

Select

Incremental Discriminant Non-Negative Matrix Factorization and Its Application to Face Recognition CAI Jing 1,2,3,WANG Wanliang 1,ZHENG Jianwei 1,LI Jiming 3 2017, 38(5): 715-721.  DOI: 10.11996/JG.j.2095-302X.2017050715 Abstract ( 82 )   PDF (525KB) ( 255 )   Aiming at the efficiency and efficacy of online feature learning for face recognition, a novel incremental discriminant learning algorithm of non-negative matrix factorization is proposed. Compared with former unsupervised online matrix factorization methods, our approach takes good advantage of the label information of training samples. Coefficient vector means of the same class were used for initialization, then the algorithm iterates with constraints on minimizing Euclidean distance of in-class samples. Which results more discriminative features and less iterations in computation. Experiments on ORL and PIE face datasets demonstrated that the proposed algorithm achieved better classification accuracy and converged faster than former batch based non-negative matrix factorization algorithms. Related Articles | Metrics

Select

Vegetation Classification Method Based on C5.0 Decision Tree of Multiple Classifiers LIU Dan 1,YANG Fengbao 1,WEI Hong 2,LI Dawei 1,HAN Xiaofeng 1 2017, 38(5): 722-728.  DOI: 10.11996/JG.j.2095-302X.201705072 Abstract ( 87 )   PDF (926KB) ( 186 )   Aiming at the problem that the vegetation classification accuracy of spectral angle mapping (SAM) and maximum likelihood classifier (MLC) for AVIRIS hyperspectral remote sensing images both are low, a vegetation classification method based on C5.0 decision tree of multiple classifiers is proposed. During the first stage, the kernel function and its parameters are selected by using support vector machine (SVM) to extract the vegetation information of AVIRIS hyperspectral image. Then, SAM and MLC combined by C5.0 algorithm, as characteristic property of the decision tree, are used to learn samples training and generate classification rules. According to the C5.0 algorithms, the information gain rate of the corresponding classifier in the vegetation samples is calculated, and the attributes with the largest information gain rate are selected to classify the samples. While classification results of leaf sample meeting the stopped growing threshold, outputs the result of sample classification, otherwise, go back to the beginning, recursively tune the above method and goes on classifying leaf sample, until all the subset contains only the sample of a vegetation type, and complete decision. Experiments show that comparing with SAM and MLC, overall accuracy of the 第 5 期 刘 丹，等：基于多分类器的 C5.0 决策树植被分类方法 723 proposed method were improved by 6.04%, 2.92%, respectively, not only confirmed the feasibility and effectiveness of the combination of multiple classifiers, and it is more suitable for vegetation investigation in the AVIRIS hyperspectral image. Related Articles | Metrics

Select

Kinship Verification Through Color Face Images WU Xiaoting 1,FENG Xiaoyi 1,SUN Kai 2,PENG Jinye 3,HADID Abdenour 1,4 2017, 38(5): 729-735.  DOI: 10.11996/JG.j.2095-302X.2017050729 Abstract ( 194 )   PDF (2115KB) ( 394 )   Face-based kinship verification is a challenging and new problem in recent years, which aims to automatically predict whether two persons have a biological kin relation or not by examining their facial attributes. Based on the fact that most of the existing methods focus on gray-scale facial images, thus discarding the chrominance information, this paper investigates the role of color in the verification of kinship relationships from facial images. In this context, we analyze the effect of chrominance information on kinship verification and compare the performance of kinship verification from color images and gray images. Furthermore, our experiments involve in different color spaces and the combination of weighted features from two color spaces. Experiment results on KinFaceW and TSKinFace indicate that color images can achieve better performance than gray scale features and HSV color space performs best. Moreover, this paper also discusses a feature extraction by fusing with different color spaces. Fusing HSV with YCbCr color spaces can further improve the performance. Related Articles | Metrics

Select

An Interpret of Active Contour Models Based on Mechanics Theory LI Liming, TANG Kelun, WEN Huabin 2017, 38(5): 736-744.  DOI: 10.11996/JG.j.2095-302X.2017050736 Abstract ( 89 )   PDF (420KB) ( 177 )   Snake models compromises low-level knowledge of images and the high-level knowledge of images. Snake models have been widely used in target tracking, edge detection, image segment and reverse engineering. It is considered as an active research field. Clear mechanic theory is included in Traditional snake models and some improved active contour models. The mechanic theory of traditional snake models and the B-snake models which is based on the theory of B-spline is presented firstly. Because of some defects in traditional snake model, some improved active contour models also are discussed from the mechanics view. In additional, the convergence process of the snake models and the active contour with inertia force are interpreted by Lagrange dynamics. In the end, the summary and the expectation are proposed. Related Articles | Metrics

Select

Single Image Defogging Algorithm Based on High-Order Markov Random Fields LENG Jianwei 1,LI Peng 2 2017, 38(5): 745-753.  DOI: 10.11996/JG.j.2095-302X.2017050745 Abstract ( 105 )   PDF (1592KB) ( 434 )   Aim to based on the dark channel prior image defogging algorithm lack of prior information and such as the large white object region failure problems, put forward a single image defogging algorithm based on high-order Markov random fields (MRF). According to the imaging characteristics of the fog haze image into the bright region and dark region. Then according to the different areas of improvement and the attenuation of the dark channel, and through the dark channel prior to obtain the transmission map. Finally use high-order MRF to optimize transmission diagram, solve the problem of insufficient information of their experience. The experimental results show that the proposed algorithm can effectively improve the partial distortion caused by the prior failure of the dark channel, and the restored image has higher resolution and more detail. Related Articles | Metrics

Select

Convolutional Neural Network-Based Chinese Ink-Painting Artistic Style Extraction WANG Chenchen 1,WANG Yelin 2,GE Zhongqin 1,CHU Kaiyue 3,CAI Jing 3,JIN Jianhua 3,CHEN Ying 1,GE Yun 1 2017, 38(5): 754-759.  DOI: 10.11996/JG.j.2095-302X.2017050754 Abstract ( 127 )   PDF (3317KB) ( 474 )   This paper discusses the process of Chinese ink-painting style learning using convolution neural network. Firstly, the frame structure of VGG19 neural network model is analyzed, and the process of using VGG19 model to separate and recombine the content and style of artistic images. Secondly, based on the theory, according to the actual characteristics of Chinese ink painting, the appropriate choice of the convoluted layer to process the content image is found and proved by experimental results. The optimal combination of convoluted layer to extract the style from Chinese ink painting is also found by experiment, and the criteria for visual evaluation of image quality are proposed. Finally, by adjusting the proportion coefficient of the content image and the style image, the expected combined image can be obtained, which verifies the feasibility of the theory and puts forward a new method for Chinese ink-painting style extraction. Related Articles | Metrics

Select

Research on Automated Generation of Annotations for Construction Drawings Based on IFC Standard HUANG Jingfei, LAI Huahui, DENG Xueyuan 2017, 38(5): 760-766.  DOI: 10.11996/JG.j.2095-302X.2017050760 Abstract ( 192 )   PDF (1288KB) ( 363 )   As a new concept and technology, building information modeling (BIM) gradually changes the work habits of design industry. However, a working pattern of ‘modeling after drafting’ which means the modeling process lags behind the drafting process is widely applied for the time-being, and it ultimately fails to improve the work quality and efficiency. This paper conducts a research about generating 2D drawings based on IFC standard. Through analyzing the technical features of 2D drawings, this paper summarizes the IFC expressions of 3D models and 2D annotations, suggests an automated method to generate IFC-based annotations with consideration of practical demands of drafting, and provides references for the development of localized BIM collaborative design platform. Related Articles | Metrics

Select

The Deep Insight Into the Engineering Graphic Courses of IUPUI ZHU Keqian 1,ACHESON Douglas 2,YUAN Huixin 1,LIU Shanshu 1,DING Jianning 1,3,YUAN Ningyi 3,4 2017, 38(5): 767-771.  DOI: 10.11996/JG.j.2095-302X.2017050767 Abstract ( 89 )   PDF (3242KB) ( 213 )   The current teaching situations of the engineering graphics courses in Purdue University Indianapolis (IUPUI) are introduced in detail in this paper. Some characteristics of engineering graphics education in U.S. universities are discussed, as well as some efforts in students’ quality and ability. Besides, a couple of useful references to us are illustrated. Related Articles | Metrics

Select

Teaching Research and Practice of Computer Experiment on CAD of Engineering Drawing Course Based on the “Internet Plus” MU Haozhi 1,WANG Xiaofei 1,CAI Bing 2,NIU Xinghua 1 2017, 38(5): 772-778.  DOI: 10.11996/JG.j.2095-302X.2017050772 Abstract ( 88 )   PDF (1488KB) ( 283 )   Based on the application of “Internet plus” technology, the teaching reformation of computer experiment on CAD of Engineering Drawing Course by using information and communication technology and the internet platform to carry out teaching activities is put forward. According to the teaching theory and cognitive law, teaching module of online course platform is constructed. The online course, teaching materials and physical experiment are combined organically with the help of planar bar code. Instant online help can be offered by scanning the planar bar code in the teaching material. The practice shows that the reformation stimulates the enthusiasm of students’ autonomous learning and improves the efficiency of both teaching and learning. Related Articles | Metrics

Select

The Construction of Student-Centered Engineering Drawing SPOC Platform LOU Hui 2017, 38(5): 779-784.  DOI: 10.11996/JG.j.2095-302X.2017050779 Abstract ( 78 )   PDF (1125KB) ( 223 )   Based on many years of network aided teaching practice, the platform of integrated campus network education is set up and used for blended learning of engineering drawing, which combines offline and online. In the construction and use of the platform, it is based on the students-centered concept, focuses on students and their needs, and also regards students as the main players of the education reform. The platform provides resources for students’ autonomous learning, designs learning activities which attract continuous students to participate in, and also provides incentives to spur students to learn online. The statistical data shows that the platform can promote the online learning enthusiasm of students and enhance their learning achievement. The students’ feedback indicates that the platform can be accepted by students for their requirement can be met. Related Articles | Metrics

Select

Research on Hybrid Engineering Drawing Curriculum SHI Yanhong, LIU Huazhou 2017, 38(5): 785-788.  DOI: 10.11996/JG.j.2095-302X.2017050785 Abstract ( 90 )   PDF (261KB) ( 219 )   This passage analyses the features of engineering drawing curriculum which introduces 3D modeling and 2D compuater aided drawing, compares pros and cons of three teaching models adopted by domestic universities, and puts forward principles and implementation scheme of “hybrid” engineering drawing curriculum. This passage puts emphasis on points of penetration and instructional design of integration of 3D modeling and engineering drawing from plane figures, elementary units, assemblies, parts drawing and assembly drawing. Teaching materials and arragements are proposed, taking the 120-hour course as an example. Related Articles | Metrics

Select

Virtual Running Exercise System Based on Kinect HUANG Dongjin 1,2,YAO Yuanqiu 1,DING Youdong 1,2,WEN Tang 3 2017, 38(5): 789-795.  DOI: 10.11996/JG.j.2095-302X.2017050789 Abstract ( 82 )   PDF (6976KB) ( 162 )   Fitness running is a common exercise in sports. Due to the improper running form, it is not only hard to reach the ideal fitness result, but also easy to harm the body seriously, especially the knee. To address the problems of lacking scientific guidance and monitoring in current running systems, this paper proposed a virtual running exercise system based on Kinect. In the action recognition, the dimension of running data collected from Kinect is reduced according to the angle characteristics and the relative position of human joints, and running action units are divided from motion features set. Then all kinds of running actions are recognized by matching with curve characteristics of action units and classification models. In addition, scientific guidance about their running poses is given to users while exercise. In the motion modeling, we analyze the relationship between the heart rate and running intensity, and build the fitness running model and the individualized scientific running exercise scheme, making the heart rate always in the safe and effective range. The experiment results show that our system is practical and feasible to provide a scientific method for safe and effective fitness running to users. Related Articles | Metrics