{"title":"Blind deconvolution using maximum a posteriori estimates with dictionary learning","authors":"V. Maik, Seonhee Park, J. Paik","doi":"10.1109/ICCE-ASIA.2016.7804725","DOIUrl":"https://doi.org/10.1109/ICCE-ASIA.2016.7804725","url":null,"abstract":"Blind deconvolution aims to obtain the original sharp image from the observed blurred image due to various distortion factors such as noise, out-of-focus, camera shake, etc. The solution to this imaging inverse problem is severely ill-posed and various heuristics in the form of some prior is required to approximate the solution. In this paper, we propose an novel deblurring algorithm using maximum a posteriori (MAP) estimation combined with sparse priors from a previously trained dictionary along with edge prior. The proposed numerical optimization methods can produce results, far better when compared to similar existing state-of-the-art methods.","PeriodicalId":229557,"journal":{"name":"2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130438415","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}
Sangheon Kwon, Kyungmin Lee, Yoonsoo Kim, Kyungah Kim, Changmin Lee, W. Ro
{"title":"Measuring error-tolerance in SRAM architecture on hardware accelerated neural network","authors":"Sangheon Kwon, Kyungmin Lee, Yoonsoo Kim, Kyungah Kim, Changmin Lee, W. Ro","doi":"10.1109/ICCE-ASIA.2016.7804818","DOIUrl":"https://doi.org/10.1109/ICCE-ASIA.2016.7804818","url":null,"abstract":"Hardware accelerators for convolutional neural network (CNN) accompany a large amount of SRAM in order to reduce the number of expensive off-chip DRAM accesses. This design trend gives implications to architects: the SRAM area will dominate the entire chip area for the future CNN accelerators. Since the probability of soft errors such as energetic particle strikes goes as the density of SRAM, errors on memory sub-system will become a major concern as process technology scales. In this paper, we investigate the necessity of a fault-tolerant memory system, against such soft errors, in hardware accelerated neural network. We found that convolutional layers have different error tolerance from each other. The error tolerance of a layer tends to get worse as it goes on the output layer.","PeriodicalId":229557,"journal":{"name":"2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116237993","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}
ByeongGi Jang, Sungjae Oh, Young-Jun Park, Kangyoon Lee
{"title":"A high efficiency active rectifier with zero current sensing for loosely-coupled wireless power transfer systems","authors":"ByeongGi Jang, Sungjae Oh, Young-Jun Park, Kangyoon Lee","doi":"10.1109/ICCE-ASIA.2016.7804735","DOIUrl":"https://doi.org/10.1109/ICCE-ASIA.2016.7804735","url":null,"abstract":"This paper presents a full-wave active rectifier with zero current sensing technique for wireless power transfer system. In order to improve the Power Conversion Efficiency (PCE), the four diodes of a passive diode rectifier are replaced by N-type LDMOS to reduce voltage drop and enhance high voltage protection characteristic. Moreover, Zero Current Sensing (ZCS) is proposed to detect reverse leakage current so that gate control signal becomes off before reverse leakage current flows and PCE can be maximized. The proposed rectifier is fabricated using 0.18μm CMOS process and occupies 4 mm2 of area. The maximum simulated PCE of 96.8% at 150-kHz.","PeriodicalId":229557,"journal":{"name":"2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115299685","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}
Dattathreya, Heegwang Kim, Jinho Park, Hasil Park, J. Paik
{"title":"Fire flame detection based on color model and motion estimation","authors":"Dattathreya, Heegwang Kim, Jinho Park, Hasil Park, J. Paik","doi":"10.1109/ICCE-ASIA.2016.7804813","DOIUrl":"https://doi.org/10.1109/ICCE-ASIA.2016.7804813","url":null,"abstract":"Automated fire flame detection in the surveillance system has drawn significant attention. This paper presents a novel fire flame detection algorithm using motion estimation and color information. The key idea is using irregular motion vectors of the fire flame. This paper presents a novel fire flame detection algorithm using motion estimation and color information. The proposed algorithm consists of these following steps; (1) extraction of the fire candidate region in the HSI color space, (2) motion estimation of the candidate region using an optical flow method, (3) quantization of the motion vectors into 8 directions and (4) detection of the region where the motion is generated in various directions as the fire area.","PeriodicalId":229557,"journal":{"name":"2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125089746","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}
Sang-Seol Lee, Sung-Joon Jang, Jungho Kim, Byeong-Ho Choi
{"title":"A hardware architecture of face detection for human-robot interaction and its implementation","authors":"Sang-Seol Lee, Sung-Joon Jang, Jungho Kim, Byeong-Ho Choi","doi":"10.1109/ICCE-ASIA.2016.7804752","DOIUrl":"https://doi.org/10.1109/ICCE-ASIA.2016.7804752","url":null,"abstract":"This paper presents hardware architecture with low-complexity face detection (FD) and parallel processing of local binary pattern (LBP) generation and adaptive boosting (AdaBoost) algorithm using Haar features for the intelligent service robot system. We designed a fully pipelined architecture implemented with the design techniques, such as variable image scaling and parallel processing multiple classifiers without integral image generation, on the FPGA platform. The proposed architecture enables a real-time FD processing for a VGA video at 30 frames per second.","PeriodicalId":229557,"journal":{"name":"2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128074601","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}
{"title":"An exact VNE algorithm based on optimization theory","authors":"Haotong Cao, Longxiang Yang, Jinbo Chen","doi":"10.1109/ICCE-ASIA.2016.7804808","DOIUrl":"https://doi.org/10.1109/ICCE-ASIA.2016.7804808","url":null,"abstract":"The main challenge in network visualization is known as so called virtual network embedding(VNE). Most proposed VNE algorithms are heuristic so far in the literature. This paper presents an exact algorithm based on the Integer Linear Programming(ILP) of optimization theory. The exact algorithm VNE-ILP is able to simultaneously map virtual nodes and virtual links of VNRs. The constraint of virtual link delay, not included in former VNE study, is realized in the VNE-ILP. The goals of VNE-ILP are to consume less substrate resources and achieve less average virtual link delay. Numerical trials are made and simulation results reveal that the VNE-ILP outperforms the state of art heuristics.","PeriodicalId":229557,"journal":{"name":"2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133922076","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}
Jin-Kyu Kim, Miyoung Lee, Ju-Yeob Kim, Byungjo Kim, Joo-Hyun Lee
{"title":"An efficient pruning and weight sharing method for neural network","authors":"Jin-Kyu Kim, Miyoung Lee, Ju-Yeob Kim, Byungjo Kim, Joo-Hyun Lee","doi":"10.1109/ICCE-ASIA.2016.7804738","DOIUrl":"https://doi.org/10.1109/ICCE-ASIA.2016.7804738","url":null,"abstract":"This paper presents a compression method to reduce the number of parameters in convolutional neural networks (CNNs). Although neural networks have an excellent recognition performance in computer vision application, there is a need for a large memory for storing amount of parameters and also necessary in a high-speed computational block. Therefore we propose two the compression schemes (pruning, weight sharing) in LeNet network model using MNIST dataset. The proposed schemes reduced the number of parameters of LeNet from 430,500 to 32 excluding index buffer size.","PeriodicalId":229557,"journal":{"name":"2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132127078","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}
{"title":"Study on the effect of frame size and color histogram bins on the shot boundary detection performance","authors":"Soyoung Park, Jeong-Woo Son, Sun-Joong Kim","doi":"10.1109/ICCE-ASIA.2016.7804726","DOIUrl":"https://doi.org/10.1109/ICCE-ASIA.2016.7804726","url":null,"abstract":"Effect of frame size and color histogram bins on shot boundary detection performance in terms of precision and recall is experimentally analyzed in this paper. An HS histogram-based shot boundary detector is used for Korean soap operas with eight kinds of frame sizes an bins, respectively, applying various thresholds. The biggest image size and bins are more than one hundred higher than the smallest ones. The experimental results show that the image size and bins have no statistically significant effect on the shot boundary detection performance.","PeriodicalId":229557,"journal":{"name":"2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133806515","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}
Gosala Kulupana, Dumidu S. Talagala, H. K. Arachchi, Anil Fernando
{"title":"Energy efficient resource allocation for cloud-based interactive TV applications - auxiliary information","authors":"Gosala Kulupana, Dumidu S. Talagala, H. K. Arachchi, Anil Fernando","doi":"10.1109/ICCE-ASIA.2016.7878706","DOIUrl":"https://doi.org/10.1109/ICCE-ASIA.2016.7878706","url":null,"abstract":"Three main approaches exist to solve the optimization problem described in the paper; linear programming methods, greedy resource allocation methods and heuristic methods.","PeriodicalId":229557,"journal":{"name":"2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia)","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115561000","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}
{"title":"Tracking object considering energy balance on wireless sensor networks","authors":"T. N. Manh, P. N. Huu, C. Le","doi":"10.1109/ICCE-ASIA.2016.7804792","DOIUrl":"https://doi.org/10.1109/ICCE-ASIA.2016.7804792","url":null,"abstract":"In this paper, we propose the method to perform tracking object applications which is based on image/video to be fitted to deploy on wireless sensor networks (WSNs). The basic idea of the proposed method is to design the new network architecture which uses three types of node namely border, camera, and relay ones. Changing of node's behavior not only reduces energy consumption but also increases the energy balance of WSNs.","PeriodicalId":229557,"journal":{"name":"2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114350864","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}