Pub Date : 1900-01-01DOI: 10.23919/elinfocom.2018.8330707
Seok Lee, Youngmo Jung, H. Moon, C. Kang, Chulki Kim, Taikjin Lee, Sang Kyung Kim, Won Kyo Jeong, Jeonghun Shin, Chang Ho Cho, D. Woo
We present a heterosensor array that exploits the collective recognition ability of a set of different gas sensors to identify individual chemical vapors. In contrast to the conventional sensing approach based on the "lock-and-key" design, the heterosensor array rely on the response patterns produced over the collection of different sensors. Diverse sensors with different operating principles are employed to provide as much chemical diversity as possible. We demonstrate that this platform provides a highly discriminative tool with single element precision.
{"title":"Single element differentiation of chemical vapors using an array","authors":"Seok Lee, Youngmo Jung, H. Moon, C. Kang, Chulki Kim, Taikjin Lee, Sang Kyung Kim, Won Kyo Jeong, Jeonghun Shin, Chang Ho Cho, D. Woo","doi":"10.23919/elinfocom.2018.8330707","DOIUrl":"https://doi.org/10.23919/elinfocom.2018.8330707","url":null,"abstract":"We present a heterosensor array that exploits the collective recognition ability of a set of different gas sensors to identify individual chemical vapors. In contrast to the conventional sensing approach based on the \"lock-and-key\" design, the heterosensor array rely on the response patterns produced over the collection of different sensors. Diverse sensors with different operating principles are employed to provide as much chemical diversity as possible. We demonstrate that this platform provides a highly discriminative tool with single element precision.","PeriodicalId":413646,"journal":{"name":"2018 International Conference on Electronics, Information, and Communication (ICEIC)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116052063","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 : 1900-01-01DOI: 10.23919/ELINFOCOM.2018.8330562
Kangwon Lee, Soonuk Seol
Existing researches use CoAP for Internet of Things application, but the process of expanding the service scale is complicated or the researches are not suitable for the national network service. To solve this problem in this paper, we propose a method to apply CoAP for application to control device remotely. Through this, It possible to exchange data using CoAP without setting up a network in a nationwide network or large network. We apply CoAP to the proposed web-based remote control platform and check performance of the platform.
{"title":"Applying CoAP for real-time device control over public networks","authors":"Kangwon Lee, Soonuk Seol","doi":"10.23919/ELINFOCOM.2018.8330562","DOIUrl":"https://doi.org/10.23919/ELINFOCOM.2018.8330562","url":null,"abstract":"Existing researches use CoAP for Internet of Things application, but the process of expanding the service scale is complicated or the researches are not suitable for the national network service. To solve this problem in this paper, we propose a method to apply CoAP for application to control device remotely. Through this, It possible to exchange data using CoAP without setting up a network in a nationwide network or large network. We apply CoAP to the proposed web-based remote control platform and check performance of the platform.","PeriodicalId":413646,"journal":{"name":"2018 International Conference on Electronics, Information, and Communication (ICEIC)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116744286","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 : 1900-01-01DOI: 10.23919/ELINFOCOM.2018.8330691
Jisu Kim, Wonchang Lee, Ho-Soon Yang, Yunwoo Lee
Aspherical mirrors have higher performance and lighter weight than spherical mirrors. However, it is not easy to process, measure, and evaluate their shapes. In particular, large aperture aspherical mirrors used in satellites require high precision and take a long time to process. The conventional process is machined using a computer numerical control with a gantry structure. However, there is a disadvantage that it is difficult to process complex shapes due to lack of degrees of freedom. In order to overcome these problems, we developed a shape processing system using an industrial robot with 6 degrees of freedom. The system consists of toolpath generation program, real-time robot monitoring, and control program. We have verified the performance of the developed system through simulation software provided by the robot manufacturer as well as actual robot operation.
{"title":"Real-time monitoring and control system of an industrial robot with 6 degrees of freedom for grinding and polishing of aspherical mirror","authors":"Jisu Kim, Wonchang Lee, Ho-Soon Yang, Yunwoo Lee","doi":"10.23919/ELINFOCOM.2018.8330691","DOIUrl":"https://doi.org/10.23919/ELINFOCOM.2018.8330691","url":null,"abstract":"Aspherical mirrors have higher performance and lighter weight than spherical mirrors. However, it is not easy to process, measure, and evaluate their shapes. In particular, large aperture aspherical mirrors used in satellites require high precision and take a long time to process. The conventional process is machined using a computer numerical control with a gantry structure. However, there is a disadvantage that it is difficult to process complex shapes due to lack of degrees of freedom. In order to overcome these problems, we developed a shape processing system using an industrial robot with 6 degrees of freedom. The system consists of toolpath generation program, real-time robot monitoring, and control program. We have verified the performance of the developed system through simulation software provided by the robot manufacturer as well as actual robot operation.","PeriodicalId":413646,"journal":{"name":"2018 International Conference on Electronics, Information, and Communication (ICEIC)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125087428","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 : 1900-01-01DOI: 10.23919/ELINFOCOM.2018.8330694
Jong Hyeong Kim, J. Jang, K. Jang
In this paper, we propose superresolution convolutional neural network with color adjustment layer to obtain high quality images. We added a color adjustment layer to the last convolutional neural network layer to correct the color error.
{"title":"A color adjustment convolutional neural network for image superresolution","authors":"Jong Hyeong Kim, J. Jang, K. Jang","doi":"10.23919/ELINFOCOM.2018.8330694","DOIUrl":"https://doi.org/10.23919/ELINFOCOM.2018.8330694","url":null,"abstract":"In this paper, we propose superresolution convolutional neural network with color adjustment layer to obtain high quality images. We added a color adjustment layer to the last convolutional neural network layer to correct the color error.","PeriodicalId":413646,"journal":{"name":"2018 International Conference on Electronics, Information, and Communication (ICEIC)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123850998","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 : 1900-01-01DOI: 10.23919/ELINFOCOM.2018.8330645
Shuai Li, Yukui Luo, K. Sun, K. Choi
One of the major challenges in these days is "How can we implement up-to-date object detection algorithm in the heterogeneous system?" As in 2012 Visual Object Classes Challenge (VOC)[1] have achieved a very satisfied performance of deep learning neural network (DNN) algorithm, but it depends on CUDA [2] GPU framework and can only be applied on NVIDIA accelerators. We prefer to use a more generic acceleration framework, OpenCL [3] is a golden key to achieve the requirement. Instead of CUDA for NVIDIA GPU only, OpenCL can be applied to the heterogeneous system including CPU, GPU, DSP, FPGA, etc. Heterogeneous systems are more flexible, some of them are designed for portable devices, and some are designed for low power parallel computation. These special devices play a very important role in modern life. In this paper, we present OpenCL based heterogeneous system implementation and apply DNN framework in two typical heterogeneous systems: portable system and FPGA system. Our work shows following contributions: (1) We implement a generic OpenCL based DNN object recognition framework which can executed on general GPUs (AMD, NVIDIA, etc). (2) We implement our framework on embedded system Odroid XU4 [4] by using multiple GPUs and increase 25.8% processing time. (3) We implement our framework on FPGA system and reduce the power consumption by 84.3% compared with TitanXGPU.
{"title":"Heterogeneous system implementation of deep learning neural network for object detection in OpenCL framework","authors":"Shuai Li, Yukui Luo, K. Sun, K. Choi","doi":"10.23919/ELINFOCOM.2018.8330645","DOIUrl":"https://doi.org/10.23919/ELINFOCOM.2018.8330645","url":null,"abstract":"One of the major challenges in these days is \"How can we implement up-to-date object detection algorithm in the heterogeneous system?\" As in 2012 Visual Object Classes Challenge (VOC)[1] have achieved a very satisfied performance of deep learning neural network (DNN) algorithm, but it depends on CUDA [2] GPU framework and can only be applied on NVIDIA accelerators. We prefer to use a more generic acceleration framework, OpenCL [3] is a golden key to achieve the requirement. Instead of CUDA for NVIDIA GPU only, OpenCL can be applied to the heterogeneous system including CPU, GPU, DSP, FPGA, etc. Heterogeneous systems are more flexible, some of them are designed for portable devices, and some are designed for low power parallel computation. These special devices play a very important role in modern life. In this paper, we present OpenCL based heterogeneous system implementation and apply DNN framework in two typical heterogeneous systems: portable system and FPGA system. Our work shows following contributions: (1) We implement a generic OpenCL based DNN object recognition framework which can executed on general GPUs (AMD, NVIDIA, etc). (2) We implement our framework on embedded system Odroid XU4 [4] by using multiple GPUs and increase 25.8% processing time. (3) We implement our framework on FPGA system and reduce the power consumption by 84.3% compared with TitanXGPU.","PeriodicalId":413646,"journal":{"name":"2018 International Conference on Electronics, Information, and Communication (ICEIC)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126994726","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 : 1900-01-01DOI: 10.23919/ELINFOCOM.2018.8330544
Seungmin Lee, Dat Ngo, B. Kang
We propose an algorithm which blends two different exposure time images to get WDR image. Previous WDR algorithm method is used for the low — luminance single image. In this paper, blending method is applied to two input images with different transfer function in previous WDR algorithm, and WDR algorithm method is applied. As a confirmation method, we used sigma value which is used as a numerical value of image quality. The sigma value indicates the saturation degree of the WDR image or the number of zero value pixels. Simulation result has shown that our algorithm outperformed previous method in terms of image visual quality.
{"title":"Blending for wide dynamic range image using different exposure image","authors":"Seungmin Lee, Dat Ngo, B. Kang","doi":"10.23919/ELINFOCOM.2018.8330544","DOIUrl":"https://doi.org/10.23919/ELINFOCOM.2018.8330544","url":null,"abstract":"We propose an algorithm which blends two different exposure time images to get WDR image. Previous WDR algorithm method is used for the low — luminance single image. In this paper, blending method is applied to two input images with different transfer function in previous WDR algorithm, and WDR algorithm method is applied. As a confirmation method, we used sigma value which is used as a numerical value of image quality. The sigma value indicates the saturation degree of the WDR image or the number of zero value pixels. Simulation result has shown that our algorithm outperformed previous method in terms of image visual quality.","PeriodicalId":413646,"journal":{"name":"2018 International Conference on Electronics, Information, and Communication (ICEIC)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133228587","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 : 1900-01-01DOI: 10.23919/ELINFOCOM.2018.8330626
Sanghun Yun, C. Son, Sang-ho Lee, Won-Seok Kang
In this paper, we present Hidden Markov Models (HMM) approach for forecasting the changes of heart rate. Heart rate is an important indicator of the state of our body. Forecasting changes of heart rate is equivalent to forecasting changes of the body state. We use numerous HMM models that is trained by datasets clustered on similarity basis. We find the optimal models with best probabilities in various learned HMM models and use this model to predict next heart rate variability. The heart rate data are collected by Fitbit-HR from 190 healthy persons. The prediction performance was accuracy = 91.87% and recall = 91.67%.
{"title":"Forecasting of heart rate variability using wrist-worn heart rate monitor based on hidden Markov model","authors":"Sanghun Yun, C. Son, Sang-ho Lee, Won-Seok Kang","doi":"10.23919/ELINFOCOM.2018.8330626","DOIUrl":"https://doi.org/10.23919/ELINFOCOM.2018.8330626","url":null,"abstract":"In this paper, we present Hidden Markov Models (HMM) approach for forecasting the changes of heart rate. Heart rate is an important indicator of the state of our body. Forecasting changes of heart rate is equivalent to forecasting changes of the body state. We use numerous HMM models that is trained by datasets clustered on similarity basis. We find the optimal models with best probabilities in various learned HMM models and use this model to predict next heart rate variability. The heart rate data are collected by Fitbit-HR from 190 healthy persons. The prediction performance was accuracy = 91.87% and recall = 91.67%.","PeriodicalId":413646,"journal":{"name":"2018 International Conference on Electronics, Information, and Communication (ICEIC)","volume":"480 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133622914","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 : 1900-01-01DOI: 10.23919/ELINFOCOM.2018.8330710
Kyu-hyun Nam, Won-jae Jung, N. Hong, Jin-sup Kim, Jun-Seok Park
This paper present a 5.9 GHz dedicated short range communication (DSRC) transmitter for wireless vehicular communication that satisfies specification of IEEE 802.11p standard mask c. The proposed transmitter consists of programmable gain amplifier (PGA), up-conversion mixer, drive amplifier, and synthesizer. Passive mixer, push-pull local oscillator (LO) buffer, and intermodulation distortion (IMD) canceller are applied to achieve high linearity and low power consumption. The transmitter is designed and fabricated on CMOS 0.18 μm process. The total power consumption is 63 mA at 1.8 V. A phase noise of synthesizer is −109.4 dBc/Hz at 1MHz offset. The transmitter total noise figure (NF) and adjacent channel leakage ratio (ACLR) are 14.2 dB and − 53.6 dBc, respectively.
{"title":"A 5.9 GHz DSRC transmitter IC for vehicle wireless communication system","authors":"Kyu-hyun Nam, Won-jae Jung, N. Hong, Jin-sup Kim, Jun-Seok Park","doi":"10.23919/ELINFOCOM.2018.8330710","DOIUrl":"https://doi.org/10.23919/ELINFOCOM.2018.8330710","url":null,"abstract":"This paper present a 5.9 GHz dedicated short range communication (DSRC) transmitter for wireless vehicular communication that satisfies specification of IEEE 802.11p standard mask c. The proposed transmitter consists of programmable gain amplifier (PGA), up-conversion mixer, drive amplifier, and synthesizer. Passive mixer, push-pull local oscillator (LO) buffer, and intermodulation distortion (IMD) canceller are applied to achieve high linearity and low power consumption. The transmitter is designed and fabricated on CMOS 0.18 μm process. The total power consumption is 63 mA at 1.8 V. A phase noise of synthesizer is −109.4 dBc/Hz at 1MHz offset. The transmitter total noise figure (NF) and adjacent channel leakage ratio (ACLR) are 14.2 dB and − 53.6 dBc, respectively.","PeriodicalId":413646,"journal":{"name":"2018 International Conference on Electronics, Information, and Communication (ICEIC)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115024616","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 : 1900-01-01DOI: 10.23919/ELINFOCOM.2018.8330592
Kiyeon Kim, Seonhee Park, Soohwan Yu, J. Paik
This paper presents a back-light image enhancement method using modified histogram as an optimal intensity transfer function based on clipped histogram equalization. The proposed method separates and clips the input histogram according to the clipping rate computed by the mean brightness of an input image. Next, the clipped histogram bins are adaptively redistributed according to the shape of histogram in the bright region. Since the bright region of modified histogram has a uniform shape, the proposed method makes the cumulative distribution function be a linear function in the bright region of the histogram. The experimental results show that the proposed method enhances the contrast of dark region without saturation and over-enhancement in the bright region.
{"title":"Bright region preserving back-light image enhancement using clipped histogram equalization","authors":"Kiyeon Kim, Seonhee Park, Soohwan Yu, J. Paik","doi":"10.23919/ELINFOCOM.2018.8330592","DOIUrl":"https://doi.org/10.23919/ELINFOCOM.2018.8330592","url":null,"abstract":"This paper presents a back-light image enhancement method using modified histogram as an optimal intensity transfer function based on clipped histogram equalization. The proposed method separates and clips the input histogram according to the clipping rate computed by the mean brightness of an input image. Next, the clipped histogram bins are adaptively redistributed according to the shape of histogram in the bright region. Since the bright region of modified histogram has a uniform shape, the proposed method makes the cumulative distribution function be a linear function in the bright region of the histogram. The experimental results show that the proposed method enhances the contrast of dark region without saturation and over-enhancement in the bright region.","PeriodicalId":413646,"journal":{"name":"2018 International Conference on Electronics, Information, and Communication (ICEIC)","volume":"270 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116211942","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 : 1900-01-01DOI: 10.23919/ELINFOCOM.2018.8330551
Youn-Sik Son, S. Heo
In automotive radar systems resolution of each target in multi-target environment is necessary. A conventional FMCW radar which transmits and receives triangular waves, does not effectively remove the ghost-targets which arise in the process of Doppler-range localization process. Using a sophisticated waveform design can help removing the ghost-targets, however, there remains many of them. In this paper, we propose a novel ghost-target removal algorithm using the inter-relationship between the upbeat and downbeat frequency after the Fast Fourier Transform(FFT).
{"title":"A novel multi-target detection algorithm for automotive FMCW radar","authors":"Youn-Sik Son, S. Heo","doi":"10.23919/ELINFOCOM.2018.8330551","DOIUrl":"https://doi.org/10.23919/ELINFOCOM.2018.8330551","url":null,"abstract":"In automotive radar systems resolution of each target in multi-target environment is necessary. A conventional FMCW radar which transmits and receives triangular waves, does not effectively remove the ghost-targets which arise in the process of Doppler-range localization process. Using a sophisticated waveform design can help removing the ghost-targets, however, there remains many of them. In this paper, we propose a novel ghost-target removal algorithm using the inter-relationship between the upbeat and downbeat frequency after the Fast Fourier Transform(FFT).","PeriodicalId":413646,"journal":{"name":"2018 International Conference on Electronics, Information, and Communication (ICEIC)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123573622","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：