Pub Date : 2018-06-01DOI: 10.1109/ICIVC.2018.8492848
Li Hui, Zou Chengjun, Luo Min
As important elements in the natural landscape, flowers feature numerous varieties, diverse morphs, complex structures, rich textures and strong characteristics; these features, in conjunction with influence of details such as illumination and wind, make their 3D visualization rather challenging. The visualization of flowering plant morphs, growth and development has been among the hottest and most difficult research points in the field of computer image and graphics. In this paper, literature reviews are made on the research contents, modeling methods, dynamic simulation and industry status etc of 3D visualization of flower morphs; analyses and comparisons are conducted on the basic principles, key techniques, production methods and advantages and disadvantages of some typical methods; in the end, prospect is made on the future development trend in this field.
{"title":"Research Progress on 3D Visualization of Flowers in China","authors":"Li Hui, Zou Chengjun, Luo Min","doi":"10.1109/ICIVC.2018.8492848","DOIUrl":"https://doi.org/10.1109/ICIVC.2018.8492848","url":null,"abstract":"As important elements in the natural landscape, flowers feature numerous varieties, diverse morphs, complex structures, rich textures and strong characteristics; these features, in conjunction with influence of details such as illumination and wind, make their 3D visualization rather challenging. The visualization of flowering plant morphs, growth and development has been among the hottest and most difficult research points in the field of computer image and graphics. In this paper, literature reviews are made on the research contents, modeling methods, dynamic simulation and industry status etc of 3D visualization of flower morphs; analyses and comparisons are conducted on the basic principles, key techniques, production methods and advantages and disadvantages of some typical methods; in the end, prospect is made on the future development trend in this field.","PeriodicalId":173981,"journal":{"name":"2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123216410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Noise removal is a fundamental problem in image processing. Among many approaches, the total variation has attracted great attention because of its nice mathematical interpretation. Traditional total variation explores the gradient information of the vertical and the horizontal directions. Thus, the number of directions can be increased to further improve denoising performance. The resulting challenge is higher computation since multiple constraints are introduced in denoising model. This work first transforms the quaternion total variation constraints problem in the spatial domain into a problem in the frequency domain by using the fast Fourier transform and the convolution theorem. Then, it incorporates the alternating direction method of multipliers (ADMM) to enable fast image denoising. This fast computation is verified by the comparisons with other total variation based methods including state-of-the-art methods.
{"title":"Four-Directional Total Variation Denoising Using Fast Fourier Transform and ADMM","authors":"Zhuyuan Cheng, Yuqun Chen, Lingzhi Wang, Fan Lin, Haiguang Wang, Yingpin Chen","doi":"10.1109/ICIVC.2018.8492869","DOIUrl":"https://doi.org/10.1109/ICIVC.2018.8492869","url":null,"abstract":"Noise removal is a fundamental problem in image processing. Among many approaches, the total variation has attracted great attention because of its nice mathematical interpretation. Traditional total variation explores the gradient information of the vertical and the horizontal directions. Thus, the number of directions can be increased to further improve denoising performance. The resulting challenge is higher computation since multiple constraints are introduced in denoising model. This work first transforms the quaternion total variation constraints problem in the spatial domain into a problem in the frequency domain by using the fast Fourier transform and the convolution theorem. Then, it incorporates the alternating direction method of multipliers (ADMM) to enable fast image denoising. This fast computation is verified by the comparisons with other total variation based methods including state-of-the-art methods.","PeriodicalId":173981,"journal":{"name":"2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123230903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/ICIVC.2018.8492726
Xinwei Qi, Ligang Miao
Template matching is a process of finding template image location from a known image, which is one of the main research contents in machine vision. For the multi-scale and rotated image template matching, most of template matching algorithms usually form a templates collection with different scaling ratios templates, and then the templates in the collection are matched separately. The algorithm will greatly increase the calculation burden of template matching, and the matching efficiency will be greatly reduced. This paper proposes an algorithm for multi-scale and rotated image template matching. The algorithm first computes the ring projection vector of the template, and then, the ring projection of the scaled template can be obtained by ring projection vector conversion. The normalized cross correlation is used to calculate the similarity between the new ring projection vector and the ring projection vector of each point of the scene image. In the end the similarities determine the optimal matching position and scale ratio. Experimental results show that the proposed algorithm can accurately find the correct matching position for multi-scale and rotated image template matching.
{"title":"A Template Matching Method for Multi-Scale and Rotated Images Using Ring Projection Vector Conversion","authors":"Xinwei Qi, Ligang Miao","doi":"10.1109/ICIVC.2018.8492726","DOIUrl":"https://doi.org/10.1109/ICIVC.2018.8492726","url":null,"abstract":"Template matching is a process of finding template image location from a known image, which is one of the main research contents in machine vision. For the multi-scale and rotated image template matching, most of template matching algorithms usually form a templates collection with different scaling ratios templates, and then the templates in the collection are matched separately. The algorithm will greatly increase the calculation burden of template matching, and the matching efficiency will be greatly reduced. This paper proposes an algorithm for multi-scale and rotated image template matching. The algorithm first computes the ring projection vector of the template, and then, the ring projection of the scaled template can be obtained by ring projection vector conversion. The normalized cross correlation is used to calculate the similarity between the new ring projection vector and the ring projection vector of each point of the scene image. In the end the similarities determine the optimal matching position and scale ratio. Experimental results show that the proposed algorithm can accurately find the correct matching position for multi-scale and rotated image template matching.","PeriodicalId":173981,"journal":{"name":"2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC)","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123559030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/ICIVC.2018.8492841
Cheng Chen, Funchun Yang, F. Wang, Liang Deng, Dan Zhao
Heterogeneous architectures are widely adopted in high performance computing. The MIC (Many Integrated Cores) processor, unveiled by Intel, has been widely used and draws attention by simplifying heterogeneous programming and improving performance. In this paper, we first introduce the architecture and executing models of MIC. Then we discuss programming and performance optimization on MIC system. Finally, some open issues and future directions in the heterogeneous system are discussed.
{"title":"Review of Programming and Performance Optimization on CPU-MIC Heterogeneous System","authors":"Cheng Chen, Funchun Yang, F. Wang, Liang Deng, Dan Zhao","doi":"10.1109/ICIVC.2018.8492841","DOIUrl":"https://doi.org/10.1109/ICIVC.2018.8492841","url":null,"abstract":"Heterogeneous architectures are widely adopted in high performance computing. The MIC (Many Integrated Cores) processor, unveiled by Intel, has been widely used and draws attention by simplifying heterogeneous programming and improving performance. In this paper, we first introduce the architecture and executing models of MIC. Then we discuss programming and performance optimization on MIC system. Finally, some open issues and future directions in the heterogeneous system are discussed.","PeriodicalId":173981,"journal":{"name":"2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123640015","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/ICIVC.2018.8492810
Meiping Song, Dongqing Cui, Chunyan Yu, Jubai An, Chein-I. Chang, Meping Song
Aiming at detecting cracks in photovoltaic images, a crack detection algorithm of photovoltaic images based on multi-scale pyramid and improved region growing is implemented in this paper. Firstly, in order to suppress noise from the crack area, the image is subjected to a filtering process. Then, the multi-scale pyramid is used to extract the fracture characteristics of photovoltaic images on different scales. There are obvious noise disturbances in the extracted cracks that do not conform to the characteristics of the cracks, which can be removed through an optimization process. Finally, this paper focuses on an improved directional region growing algorithm to complement the detected cracks. For comparison, the wavelet modulus maximum method is tested too. The results show that the proposed method has a better performance on noise suppression, suspect crack removing, and crack integrality.
{"title":"Crack Detection Algorithm for Photovoltaic Image Based on Multi-Scale Pyramid and Improved Region Growing","authors":"Meiping Song, Dongqing Cui, Chunyan Yu, Jubai An, Chein-I. Chang, Meping Song","doi":"10.1109/ICIVC.2018.8492810","DOIUrl":"https://doi.org/10.1109/ICIVC.2018.8492810","url":null,"abstract":"Aiming at detecting cracks in photovoltaic images, a crack detection algorithm of photovoltaic images based on multi-scale pyramid and improved region growing is implemented in this paper. Firstly, in order to suppress noise from the crack area, the image is subjected to a filtering process. Then, the multi-scale pyramid is used to extract the fracture characteristics of photovoltaic images on different scales. There are obvious noise disturbances in the extracted cracks that do not conform to the characteristics of the cracks, which can be removed through an optimization process. Finally, this paper focuses on an improved directional region growing algorithm to complement the detected cracks. For comparison, the wavelet modulus maximum method is tested too. The results show that the proposed method has a better performance on noise suppression, suspect crack removing, and crack integrality.","PeriodicalId":173981,"journal":{"name":"2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122123827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/ICIVC.2018.8492776
Nasim Abbas, Fengqi Yu
Wireless multimedia sensor networks (WMSNs) applications produce large amount of data, which require high transmission rates. An efficient and seamless delivery of multimedia services in WMSN s is still a challenging task. This article presents the design and implementation of a video surveillance system for linear wireless multimedia sensor networks. We adopt a procedure in which every node has unique IP address and can establish a route from itself to sink through multi-hop communication. We present a dynamic queue scheduler which filters the packets according to packet priority. The core component of this system is an open source hardware platform, which is based on Raspberry Pi sensor nodes. Our network is extensively evaluated on 7 Raspberry Pi sensor nodes. We present results of 7 -node real-world deployment in video surveillance application and show that it works well in long-term deployments.
{"title":"Design and Implementation of a Video Surveillance System for Linear Wireless Multimedia Sensor Networks","authors":"Nasim Abbas, Fengqi Yu","doi":"10.1109/ICIVC.2018.8492776","DOIUrl":"https://doi.org/10.1109/ICIVC.2018.8492776","url":null,"abstract":"Wireless multimedia sensor networks (WMSNs) applications produce large amount of data, which require high transmission rates. An efficient and seamless delivery of multimedia services in WMSN s is still a challenging task. This article presents the design and implementation of a video surveillance system for linear wireless multimedia sensor networks. We adopt a procedure in which every node has unique IP address and can establish a route from itself to sink through multi-hop communication. We present a dynamic queue scheduler which filters the packets according to packet priority. The core component of this system is an open source hardware platform, which is based on Raspberry Pi sensor nodes. Our network is extensively evaluated on 7 Raspberry Pi sensor nodes. We present results of 7 -node real-world deployment in video surveillance application and show that it works well in long-term deployments.","PeriodicalId":173981,"journal":{"name":"2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129848409","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/ICIVC.2018.8492899
Chuangang Wang, Fuqiang Li, Yanfeng Li, Houjin Chen, Xuyang Cao
Gear is an important component in railway. The defect status of the gear affects railway riding quality and safety. In this paper, an automatic and quantitative defect status detecting method for external gear is proposed. First, a two-stage scheme is proposed for the segmentation of the meshing region in the gear tooth. Then adaptive thresholding and shape analysis are combined to detect the surface defects. The proposed method is tested on 140 gear tooth images. The area overlap of the meshing region is 0.87. The defect detection method has better performance than some related approaches.
{"title":"A Defect Status Detecting Method for External Gear in Railway","authors":"Chuangang Wang, Fuqiang Li, Yanfeng Li, Houjin Chen, Xuyang Cao","doi":"10.1109/ICIVC.2018.8492899","DOIUrl":"https://doi.org/10.1109/ICIVC.2018.8492899","url":null,"abstract":"Gear is an important component in railway. The defect status of the gear affects railway riding quality and safety. In this paper, an automatic and quantitative defect status detecting method for external gear is proposed. First, a two-stage scheme is proposed for the segmentation of the meshing region in the gear tooth. Then adaptive thresholding and shape analysis are combined to detect the surface defects. The proposed method is tested on 140 gear tooth images. The area overlap of the meshing region is 0.87. The defect detection method has better performance than some related approaches.","PeriodicalId":173981,"journal":{"name":"2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC)","volume":"58 10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128256732","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/icivc.2018.8492740
{"title":"ICIVC 2018","authors":"","doi":"10.1109/icivc.2018.8492740","DOIUrl":"https://doi.org/10.1109/icivc.2018.8492740","url":null,"abstract":"","PeriodicalId":173981,"journal":{"name":"2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127140657","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/ICIVC.2018.8492795
Xiaoguo Wang, Hao Wu, Zhichao Yi
Internet finance is developing rapidly. As online payments such as Alipay and WeChat Pay become more and more popular, cases of fraud associated with are also rising. In this paper, we describe the entire process of fraud detection using Hidden Markov model (HMM). We use the k-means algorithm to symbolize the transaction amount and frequency sequence of a bank account. This sequence is used to build and test the model. An HMM is initially trained with the normal behavior of an account. If an incoming credit card transaction is not accepted by the trained HMM with sufficiently high probability, it is considered to be fraudulent. We illustrate the feasibility of the model through simulation experiments and verify the validity of the model with real-world bank transaction data. Especially, in the case of enough historical transactions, this method performs well for low, medium frequency and amount of user groups.
{"title":"Research on Bank Anti-Fraud Model Based on K-Means and Hidden Markov Model","authors":"Xiaoguo Wang, Hao Wu, Zhichao Yi","doi":"10.1109/ICIVC.2018.8492795","DOIUrl":"https://doi.org/10.1109/ICIVC.2018.8492795","url":null,"abstract":"Internet finance is developing rapidly. As online payments such as Alipay and WeChat Pay become more and more popular, cases of fraud associated with are also rising. In this paper, we describe the entire process of fraud detection using Hidden Markov model (HMM). We use the k-means algorithm to symbolize the transaction amount and frequency sequence of a bank account. This sequence is used to build and test the model. An HMM is initially trained with the normal behavior of an account. If an incoming credit card transaction is not accepted by the trained HMM with sufficiently high probability, it is considered to be fraudulent. We illustrate the feasibility of the model through simulation experiments and verify the validity of the model with real-world bank transaction data. Especially, in the case of enough historical transactions, this method performs well for low, medium frequency and amount of user groups.","PeriodicalId":173981,"journal":{"name":"2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122374110","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/ICIVC.2018.8492754
Lin Li, Shengbing Zhang, Juan Wu
In order to explore the efficient design and implementation of activation function in deep learning processor, this paper presents an efficient five-stage pipelined hardware architecture for activation function based on the piecewise linear interpolation, and a novel neuron data-LUT address mapping algorithm. Compared with the previous designs based on serial calculation, the proposed hardware architecture can achieve at least 3 times of acceleration. Four commonly used activation functions are designed based on the proposed hardware architecture, which is implemented on the XC6VLX240T of Xilinx. The LeNet-5 and AlexNet are selected as benchmarks to test the inference accuracy of different activation functions with different piecewise numbers on the MNIST and CIFAR-10 test sets in the deep learning processor prototype system. The experiment results show that the proposed hardware architecture can effectively accomplish the relevant calculation of activation functions in the deep learning processor and the accuracy loss is negligible. The proposed hardware architecture is adaptable for numerous activation functions, which can be widely used in the design of other deep learning processors.
{"title":"An Efficient Hardware Architecture for Activation Function in Deep Learning Processor","authors":"Lin Li, Shengbing Zhang, Juan Wu","doi":"10.1109/ICIVC.2018.8492754","DOIUrl":"https://doi.org/10.1109/ICIVC.2018.8492754","url":null,"abstract":"In order to explore the efficient design and implementation of activation function in deep learning processor, this paper presents an efficient five-stage pipelined hardware architecture for activation function based on the piecewise linear interpolation, and a novel neuron data-LUT address mapping algorithm. Compared with the previous designs based on serial calculation, the proposed hardware architecture can achieve at least 3 times of acceleration. Four commonly used activation functions are designed based on the proposed hardware architecture, which is implemented on the XC6VLX240T of Xilinx. The LeNet-5 and AlexNet are selected as benchmarks to test the inference accuracy of different activation functions with different piecewise numbers on the MNIST and CIFAR-10 test sets in the deep learning processor prototype system. The experiment results show that the proposed hardware architecture can effectively accomplish the relevant calculation of activation functions in the deep learning processor and the accuracy loss is negligible. The proposed hardware architecture is adaptable for numerous activation functions, which can be widely used in the design of other deep learning processors.","PeriodicalId":173981,"journal":{"name":"2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125673734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：