{"title":"Interfacing Personal Computers","authors":"Zdravko Karakehayov, K. Christensen, O. Winther","doi":"10.1201/9781482276824-6","DOIUrl":"https://doi.org/10.1201/9781482276824-6","url":null,"abstract":"","PeriodicalId":57136,"journal":{"name":"单片机与嵌入式系统应用","volume":"28 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2018-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78802160","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 : 2018-05-07DOI: 10.1109/ICESIT-ICICTES.2018.8442053
Theerapat Lapjaturapit, Kobkrit Viriyayudhakom, T. Theeramunkong
Most existing word segmentation methods output one single segmentation solution. This paper provides an analysis of word segmentation performance when more than one solutions are taken into account. Towards this investigation, a deep neural network with multiple thresholds is applied to generate multiple candidates for segmentation. As a test-bed, the well-known bidirectional long short-term memory (BiLSTM) units are used with eleven contexts in a deep neural network. As performance indices, three measures; recall, precision and f-measure, are plotted with respect to various thresholds for both boundary level and word level evaluation. By a number of experiments, the result shows that the multi-candidate word segmentation can help us increase the recalls while maintaining the precisions.
{"title":"Multi-Candidate Word Segmentation using Bi-directional LSTM Neural Networks","authors":"Theerapat Lapjaturapit, Kobkrit Viriyayudhakom, T. Theeramunkong","doi":"10.1109/ICESIT-ICICTES.2018.8442053","DOIUrl":"https://doi.org/10.1109/ICESIT-ICICTES.2018.8442053","url":null,"abstract":"Most existing word segmentation methods output one single segmentation solution. This paper provides an analysis of word segmentation performance when more than one solutions are taken into account. Towards this investigation, a deep neural network with multiple thresholds is applied to generate multiple candidates for segmentation. As a test-bed, the well-known bidirectional long short-term memory (BiLSTM) units are used with eleven contexts in a deep neural network. As performance indices, three measures; recall, precision and f-measure, are plotted with respect to various thresholds for both boundary level and word level evaluation. By a number of experiments, the result shows that the multi-candidate word segmentation can help us increase the recalls while maintaining the precisions.","PeriodicalId":57136,"journal":{"name":"单片机与嵌入式系统应用","volume":"72 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2018-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83557694","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 : 2018-05-07DOI: 10.1109/ICESIT-ICICTES.2018.8442072
S. Vorapojpisut
This paper addresses the implementation of device driver blocks to be used in model-based design of embedded software. After notations related to hardware and real-world events are defined, two design patterns, namely Frame Buffer and Slot Buffer, are proposed for serial communication tasks in IoT/WSN nodes. To assure model determinism involving hardware operations, the balance equation is discussed and generalized for the time-triggered architecture of Simulink environment.
{"title":"Model-based Design of IoT/WSN Nodes: Device Driver Implementation","authors":"S. Vorapojpisut","doi":"10.1109/ICESIT-ICICTES.2018.8442072","DOIUrl":"https://doi.org/10.1109/ICESIT-ICICTES.2018.8442072","url":null,"abstract":"This paper addresses the implementation of device driver blocks to be used in model-based design of embedded software. After notations related to hardware and real-world events are defined, two design patterns, namely Frame Buffer and Slot Buffer, are proposed for serial communication tasks in IoT/WSN nodes. To assure model determinism involving hardware operations, the balance equation is discussed and generalized for the time-triggered architecture of Simulink environment.","PeriodicalId":57136,"journal":{"name":"单片机与嵌入式系统应用","volume":"5 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2018-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87594206","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}
Clickbait is a widely used writing style of the news headline that has been utilized to gain attentions from the reader for the revenues generated from the clicks. After the readers open the link to read the content of the clickbait headline, it leaves them a disappointment. In this work, the characteristic of Thai clickbait headlines is studied. To compare the effect of the special characters such as ‘!’, ‘?’, and ‘#’ on the classification, two datasets, which contains the special characters and the one without them, are provided. They have been utilized in the experiments using well-known classifiers such as Decision Tree, Support Vector Machine, and Naive Bayes. The result shows that the special characters and the decision tree classifier gives 99. 90% accuracy.
{"title":"Thai Clickbait Headline News Classification and its Characteristic","authors":"Natnicha Wongsap, Tastanya Prapphan, Lisha Lou, S. Kongyoung, Sasiwimol Jumun, Natsuda Kaothanthong","doi":"10.1109/ICESIT-ICICTES.2018.8442064","DOIUrl":"https://doi.org/10.1109/ICESIT-ICICTES.2018.8442064","url":null,"abstract":"Clickbait is a widely used writing style of the news headline that has been utilized to gain attentions from the reader for the revenues generated from the clicks. After the readers open the link to read the content of the clickbait headline, it leaves them a disappointment. In this work, the characteristic of Thai clickbait headlines is studied. To compare the effect of the special characters such as ‘!’, ‘?’, and ‘#’ on the classification, two datasets, which contains the special characters and the one without them, are provided. They have been utilized in the experiments using well-known classifiers such as Decision Tree, Support Vector Machine, and Naive Bayes. The result shows that the special characters and the decision tree classifier gives 99. 90% accuracy.","PeriodicalId":57136,"journal":{"name":"单片机与嵌入式系统应用","volume":"125 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80201127","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 : 2018-05-01DOI: 10.1109/ICESIT-ICICTES.2018.8442052
T. Tantidham, S. Ngamsuriyaros, N. Tungamnuayrith, T. Nildam, K. Banthao, P. Intakot
With increasing cost and energy consumption, many organizations would find a way for monitoring and to save energy. Our work would design and implement a monitoring system to collect energy usage data and use such data to analyze energy consumption patterns. We measure the amount of electricity of AC outlets, lights and air conditions in a building. Our work consists of 3 parts: Power Data Collection, Data Processing and Data Report. Power Data Collection is composed of power meters built on an ESP8266 Wi-Fi platform and PZEM-004T which will be attached to 3 types of devices: AC outlet, air condition and light. The Data Processing part is performed at the Elasticsearch server by gathering the measured power data from multiple power meters via MQTT protocol. The Data Report part uses Grafana to show usage statistics via a web interface. Usually, one room will have number of lights and outlets, and one or two air conditions. We also collect the measured data every 15 minutes. Thus, collected data will become very huge in a short period of time. In the future, we need to do pattern analysis of energy consumption of a building.
{"title":"Energy Consumption Collection Design for Smart Building","authors":"T. Tantidham, S. Ngamsuriyaros, N. Tungamnuayrith, T. Nildam, K. Banthao, P. Intakot","doi":"10.1109/ICESIT-ICICTES.2018.8442052","DOIUrl":"https://doi.org/10.1109/ICESIT-ICICTES.2018.8442052","url":null,"abstract":"With increasing cost and energy consumption, many organizations would find a way for monitoring and to save energy. Our work would design and implement a monitoring system to collect energy usage data and use such data to analyze energy consumption patterns. We measure the amount of electricity of AC outlets, lights and air conditions in a building. Our work consists of 3 parts: Power Data Collection, Data Processing and Data Report. Power Data Collection is composed of power meters built on an ESP8266 Wi-Fi platform and PZEM-004T which will be attached to 3 types of devices: AC outlet, air condition and light. The Data Processing part is performed at the Elasticsearch server by gathering the measured power data from multiple power meters via MQTT protocol. The Data Report part uses Grafana to show usage statistics via a web interface. Usually, one room will have number of lights and outlets, and one or two air conditions. We also collect the measured data every 15 minutes. Thus, collected data will become very huge in a short period of time. In the future, we need to do pattern analysis of energy consumption of a building.","PeriodicalId":57136,"journal":{"name":"单片机与嵌入式系统应用","volume":"79 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82441299","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 : 2018-05-01DOI: 10.1109/ICESIT-ICICTES.2018.8442054
Samol Chreang, P. Kumhom
Ahstract-Light-Emitting Diode (LED) bulbs or lamps are popularly used nowadays because of their less power consumption and long lifespan. However, usages of LED bulbs including the DC and AC types affect power quality of the power system because the needs of DC-to-DC converters and/or inverters. This paper presents a comparison of power quality affected by the DC and AC LED bulb loads supplied by a PV generator (PVG) connected to a PV microgrid. Various loading conditions based on centralized and decentralized drivers are carried out. Also the effect switching frequency for DC-to-DC converter used inside LED drivers is explored. The power quality is measured by means of total distortion percentage in the power line currents on both DC and AC side. The study is based on simulations using the existing simulation models of a PV generator, a converter, an inverter, DC LED bulb, and AC LED bulbs. The results show that DC LED lamp loading has higher effects to the total distortion than AC LED has. Moreover, using the centralized driver has a small advantage over the decentralized case while the distortion increases when the switching frequency is reduced. Although the resulting distortions from the simulations may not be accurate, the comparison results may be used as a guideline for designing a PV microzrid.
{"title":"Power Quality Affected by DC and AC LED Bulb Loading of a PV Generator Connected to a Microgrid","authors":"Samol Chreang, P. Kumhom","doi":"10.1109/ICESIT-ICICTES.2018.8442054","DOIUrl":"https://doi.org/10.1109/ICESIT-ICICTES.2018.8442054","url":null,"abstract":"Ahstract-Light-Emitting Diode (LED) bulbs or lamps are popularly used nowadays because of their less power consumption and long lifespan. However, usages of LED bulbs including the DC and AC types affect power quality of the power system because the needs of DC-to-DC converters and/or inverters. This paper presents a comparison of power quality affected by the DC and AC LED bulb loads supplied by a PV generator (PVG) connected to a PV microgrid. Various loading conditions based on centralized and decentralized drivers are carried out. Also the effect switching frequency for DC-to-DC converter used inside LED drivers is explored. The power quality is measured by means of total distortion percentage in the power line currents on both DC and AC side. The study is based on simulations using the existing simulation models of a PV generator, a converter, an inverter, DC LED bulb, and AC LED bulbs. The results show that DC LED lamp loading has higher effects to the total distortion than AC LED has. Moreover, using the centralized driver has a small advantage over the decentralized case while the distortion increases when the switching frequency is reduced. Although the resulting distortions from the simulations may not be accurate, the comparison results may be used as a guideline for designing a PV microzrid.","PeriodicalId":57136,"journal":{"name":"单片机与嵌入式系统应用","volume":"21 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80545794","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 : 2018-05-01DOI: 10.1109/ICESIT-ICICTES.2018.8442070
K. Athikulwongse, Charuwalee Suwatthikul, Witsarawat Chantaweesomboon
Ultra-wideband (UWB) technology has recently improved the performance of real-time locating systems (RTLS) to centimeter-level accuracy. One of commercially available integrated circuits (IC) for UWB RTLS is from Decawave. Its DW1000 IC can be used in two-way ranging (TWR) method to measure precise time of flight, thus accurate estimation of the distance between an anchor and a tag. A trilateration algorithm can then estimate the position of a tag in two-dimensional (2-D) and three-dimensional (3-D) space from the distance between the tag and three or more anchors. In this work, the impact of altitude of anchors above the tag on the performance of a TWR UWB RTLS was studied. In the experiments, the anchors and tag were first placed on the first floor of a building. Then, some of the anchors were raised or moved to higher floors. The results revealed that the error distances along x-, y-, and z-axis varied differently with increasing altitude of the anchors, thus changing their contribution to the 2-D and 3-D performance. The results indicated that altitude of anchors dramatically affected the performance, especially in 3-D. The results can help suggest how to find optimal positions of anchors for deployment of the RTLS.
{"title":"Impact of Altitude of Anchors on Performance of UWB Real-Time Locating System","authors":"K. Athikulwongse, Charuwalee Suwatthikul, Witsarawat Chantaweesomboon","doi":"10.1109/ICESIT-ICICTES.2018.8442070","DOIUrl":"https://doi.org/10.1109/ICESIT-ICICTES.2018.8442070","url":null,"abstract":"Ultra-wideband (UWB) technology has recently improved the performance of real-time locating systems (RTLS) to centimeter-level accuracy. One of commercially available integrated circuits (IC) for UWB RTLS is from Decawave. Its DW1000 IC can be used in two-way ranging (TWR) method to measure precise time of flight, thus accurate estimation of the distance between an anchor and a tag. A trilateration algorithm can then estimate the position of a tag in two-dimensional (2-D) and three-dimensional (3-D) space from the distance between the tag and three or more anchors. In this work, the impact of altitude of anchors above the tag on the performance of a TWR UWB RTLS was studied. In the experiments, the anchors and tag were first placed on the first floor of a building. Then, some of the anchors were raised or moved to higher floors. The results revealed that the error distances along x-, y-, and z-axis varied differently with increasing altitude of the anchors, thus changing their contribution to the 2-D and 3-D performance. The results indicated that altitude of anchors dramatically affected the performance, especially in 3-D. The results can help suggest how to find optimal positions of anchors for deployment of the RTLS.","PeriodicalId":57136,"journal":{"name":"单片机与嵌入式系统应用","volume":"16 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82524822","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 : 2018-05-01DOI: 10.1109/ICESIT-ICICTES.2018.8442060
W. Khan-ngern, Wiwat Keyoonwong
The project present about Embedded Electronic Differential System(EDs) on Two Brushless DC (BLDC) Motors Drives for Electric Vehicle Steering Control. EDs using for steering control to balance speed between two BLDC motors and enhances efficiency of electric vehicle driving system. Efficiency testing on BLDC motor driving system to determine the efficiency of BLDC motor driving system including inverter and BLDC motor at various load conditions. The maximum efficiency of BLDC motor at 48 volts is 75.71% with 6.52 kg load Embedded system including speed limit to protect BLDC motor from overrated speed and feedback speed, Handwriting coding by Arduino mega 2560. The steering control using digital potentiometer, MCP41010. When the turning angle of electric vehicle is any degrees (right or left turning status), the control system will command. And when BLDC motor near overrated speed and turning status, the overall speed of electric vehicle is reduced because the both BLDC motors speed are limited at 840 rpm. The experimental results give satisfactory performance.
本课题介绍了一种嵌入式电子差动系统(EDs)在两个无刷直流(BLDC)电机驱动上用于电动汽车转向控制。用于转向控制的无刷直流电动机可以平衡两台无刷直流电动机的速度,提高电动汽车驱动系统的效率。对无刷直流电动机驱动系统进行效率测试,确定包括逆变器和无刷直流电动机在内的无刷直流电动机驱动系统在各种负载条件下的效率。嵌入式系统包括限速,以保护BLDC电机不被高估的速度和反馈速度,由Arduino mega 2560手写编码。转向控制采用数字电位器,MCP41010。当电动汽车的转弯角度为任意角度(右转弯或左转弯)时,控制系统发出指令。当无刷直流电动机转速接近过额定转速和转弯状态时,由于双无刷直流电动机转速限制在840转/分,电动汽车的整体速度降低。实验结果令人满意。
{"title":"Embedded Electronic Differential System on Two Brushless DC Motor Drives for Electric Vehicle Steering Control","authors":"W. Khan-ngern, Wiwat Keyoonwong","doi":"10.1109/ICESIT-ICICTES.2018.8442060","DOIUrl":"https://doi.org/10.1109/ICESIT-ICICTES.2018.8442060","url":null,"abstract":"The project present about Embedded Electronic Differential System(EDs) on Two Brushless DC (BLDC) Motors Drives for Electric Vehicle Steering Control. EDs using for steering control to balance speed between two BLDC motors and enhances efficiency of electric vehicle driving system. Efficiency testing on BLDC motor driving system to determine the efficiency of BLDC motor driving system including inverter and BLDC motor at various load conditions. The maximum efficiency of BLDC motor at 48 volts is 75.71% with 6.52 kg load Embedded system including speed limit to protect BLDC motor from overrated speed and feedback speed, Handwriting coding by Arduino mega 2560. The steering control using digital potentiometer, MCP41010. When the turning angle of electric vehicle is any degrees (right or left turning status), the control system will command. And when BLDC motor near overrated speed and turning status, the overall speed of electric vehicle is reduced because the both BLDC motors speed are limited at 840 rpm. The experimental results give satisfactory performance.","PeriodicalId":57136,"journal":{"name":"单片机与嵌入式系统应用","volume":"97 1","pages":"1-4"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75993529","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 : 2018-05-01DOI: 10.1109/ICESIT-ICICTES.2018.8442051
Chadaporn Keatmanee, Siriaksorn Jakborvornphan, Chakrapan Potiwanna, Wimol San-Uml, M. Dailey
Ahstract- Several intelligent transportation system applications, such as automatic steering systems, lane departure warning systems, and driver performance monitoring systems, depend on obtaining accurate estimates of the vehicle's trajectory with respect to the lane the vehicle is driving in. However, lane marking detection and tracking using video or laser sensors tend to be extremely noisy, requiring robust methods for lane marking detection and tracking. Therefore, it has been an active field of research for decades. In this paper, we survey methods for detecting lane marking as well as tracking lane boundaries in real time focusing on vision-based system. We present a generic problem and elaborate the broad ranged proposed method in vision-based system. Finally, we identify these gaps and recommend research direction that may bridge them.
{"title":"Vision-Based Lane Keeping - A Survey","authors":"Chadaporn Keatmanee, Siriaksorn Jakborvornphan, Chakrapan Potiwanna, Wimol San-Uml, M. Dailey","doi":"10.1109/ICESIT-ICICTES.2018.8442051","DOIUrl":"https://doi.org/10.1109/ICESIT-ICICTES.2018.8442051","url":null,"abstract":"Ahstract- Several intelligent transportation system applications, such as automatic steering systems, lane departure warning systems, and driver performance monitoring systems, depend on obtaining accurate estimates of the vehicle's trajectory with respect to the lane the vehicle is driving in. However, lane marking detection and tracking using video or laser sensors tend to be extremely noisy, requiring robust methods for lane marking detection and tracking. Therefore, it has been an active field of research for decades. In this paper, we survey methods for detecting lane marking as well as tracking lane boundaries in real time focusing on vision-based system. We present a generic problem and elaborate the broad ranged proposed method in vision-based system. Finally, we identify these gaps and recommend research direction that may bridge them.","PeriodicalId":57136,"journal":{"name":"单片机与嵌入式系统应用","volume":"149 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86125109","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 : 2018-05-01DOI: 10.1109/ICESIT-ICICTES.2018.8442068
Duangkamol Tedlek, S. Khoomboon, T. Kasetkasem, T. Chanwimaluang, I. Kumazawa
This paper proposes a cloud removal algorithm for satellite images that are obscured by clouds. Our concept is to separate the land cover map into the segments. After that, we localize and indicate the degree of cloud thickness map by using a level set function. As the continuity of the level set function, the gradient approach is deployed by using the calculus of variation. Our example shows that our model can mitigate the effect clouds in images.
{"title":"A Cloud-Contamination Removal Algorithm by Combining Image Segmentation and Level-set-based Approaches for Remote Sensing Images","authors":"Duangkamol Tedlek, S. Khoomboon, T. Kasetkasem, T. Chanwimaluang, I. Kumazawa","doi":"10.1109/ICESIT-ICICTES.2018.8442068","DOIUrl":"https://doi.org/10.1109/ICESIT-ICICTES.2018.8442068","url":null,"abstract":"This paper proposes a cloud removal algorithm for satellite images that are obscured by clouds. Our concept is to separate the land cover map into the segments. After that, we localize and indicate the degree of cloud thickness map by using a level set function. As the continuity of the level set function, the gradient approach is deployed by using the calculus of variation. Our example shows that our model can mitigate the effect clouds in images.","PeriodicalId":57136,"journal":{"name":"单片机与嵌入式系统应用","volume":"107 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90765485","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}