Pub Date : 2019-06-01DOI: 10.1109/ECAI46879.2019.9041956
Marian-Bogdan Bodea, P. Anghelescu, Bogdan-Mihai G. Gavriloaia, N. Vizireanu
Given that the society we live in is continuously moving and expanding and that most people have access to the Internet and thus to information, more and more people rely on safety systems such as “Home Automation”. In this paper we will describe the design and implementation of such a system. Several types of sensors, a controller, a wi-fi connection and a NodeMCU module have been used in the design of the system. All the sensors are connected to a controller and the internet connection is enabled by interconnection between the module NodeMCU and the Arduino Mega 2560 controller. The paper sets out experimental results stemming from various working scenarios.
鉴于我们所生活的社会不断发展和扩大，大多数人都可以访问互联网，从而获得信息，越来越多的人依赖于安全系统，如“家庭自动化”。在本文中我们将描述这样一个系统的设计和实现。在系统的设计中使用了几种类型的传感器、控制器、wi-fi连接和NodeMCU模块。所有传感器都连接到控制器上，通过模块NodeMCU和Arduino Mega 2560控制器之间的互连实现互联网连接。本文列出了各种工作场景下的实验结果。
{"title":"Design and Implementation of a Home Automation System","authors":"Marian-Bogdan Bodea, P. Anghelescu, Bogdan-Mihai G. Gavriloaia, N. Vizireanu","doi":"10.1109/ECAI46879.2019.9041956","DOIUrl":"https://doi.org/10.1109/ECAI46879.2019.9041956","url":null,"abstract":"Given that the society we live in is continuously moving and expanding and that most people have access to the Internet and thus to information, more and more people rely on safety systems such as “Home Automation”. In this paper we will describe the design and implementation of such a system. Several types of sensors, a controller, a wi-fi connection and a NodeMCU module have been used in the design of the system. All the sensors are connected to a controller and the internet connection is enabled by interconnection between the module NodeMCU and the Arduino Mega 2560 controller. The paper sets out experimental results stemming from various working scenarios.","PeriodicalId":285780,"journal":{"name":"2019 11th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114751093","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 : 2019-06-01DOI: 10.1109/ECAI46879.2019.9041987
C. Mocanu, G. Suciu
Developing and implementing a software algorithm for detecting, recognizing and interpreting hand gestures. The project involves an automatic realtime image analysis system that recognizes several predefined movements. The analysis system first involves locating and separating the areas of the image associated with the hand, followed by the motion classification (recognition). These movements will be transposed into commands to an integrated device.
{"title":"Automatic recognition of hand gestures","authors":"C. Mocanu, G. Suciu","doi":"10.1109/ECAI46879.2019.9041987","DOIUrl":"https://doi.org/10.1109/ECAI46879.2019.9041987","url":null,"abstract":"Developing and implementing a software algorithm for detecting, recognizing and interpreting hand gestures. The project involves an automatic realtime image analysis system that recognizes several predefined movements. The analysis system first involves locating and separating the areas of the image associated with the hand, followed by the motion classification (recognition). These movements will be transposed into commands to an integrated device.","PeriodicalId":285780,"journal":{"name":"2019 11th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128563311","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 : 2019-06-01DOI: 10.1109/ECAI46879.2019.9041993
Alexandru Mărchidan
This paper focuses mainly on the dangers of new communication technologies for the intellectual and cultural development of young people. Although these technologies also have some advantages for information and communication, however, the disadvantages are much greater. As such, we will analyze their effects structurally (neuro-physically), psychosocially and morally, and in the end we will try to find out what can be done to limit these negative effects.
{"title":"More technologized is not more educated","authors":"Alexandru Mărchidan","doi":"10.1109/ECAI46879.2019.9041993","DOIUrl":"https://doi.org/10.1109/ECAI46879.2019.9041993","url":null,"abstract":"This paper focuses mainly on the dangers of new communication technologies for the intellectual and cultural development of young people. Although these technologies also have some advantages for information and communication, however, the disadvantages are much greater. As such, we will analyze their effects structurally (neuro-physically), psychosocially and morally, and in the end we will try to find out what can be done to limit these negative effects.","PeriodicalId":285780,"journal":{"name":"2019 11th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116155922","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 : 2019-06-01DOI: 10.1109/ECAI46879.2019.9042152
H. Ilgin, A. Akbulut, Aykut Kalaycioglu
In this study, we present an image and video compression method, which is efficient at low bit rates and includes local cosine transform (LCT) and significance map coding. Low bit rate image or video coding causes distortion and artifacts such as blocking effect, which is induced by block-based discrete cosine transform (DCT). On the other hand, another block-based method, LCT, which implements DCT, decreases blocking artifact, because it folds contiguous blocks before applying DCT. Significance map coding (SMC) achieves better image/video quality and decreases artifacts when compared to the conventional DCT coding approach. We apply SMC in conjunction with LCT for further improvement as shown by the experiments.
{"title":"Image and Video Compression Artifact Reduction at Low Bit Rates Using Local Cosine Transform in Conjunction with Significance Map Coding","authors":"H. Ilgin, A. Akbulut, Aykut Kalaycioglu","doi":"10.1109/ECAI46879.2019.9042152","DOIUrl":"https://doi.org/10.1109/ECAI46879.2019.9042152","url":null,"abstract":"In this study, we present an image and video compression method, which is efficient at low bit rates and includes local cosine transform (LCT) and significance map coding. Low bit rate image or video coding causes distortion and artifacts such as blocking effect, which is induced by block-based discrete cosine transform (DCT). On the other hand, another block-based method, LCT, which implements DCT, decreases blocking artifact, because it folds contiguous blocks before applying DCT. Significance map coding (SMC) achieves better image/video quality and decreases artifacts when compared to the conventional DCT coding approach. We apply SMC in conjunction with LCT for further improvement as shown by the experiments.","PeriodicalId":285780,"journal":{"name":"2019 11th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116405007","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 : 2019-06-01DOI: 10.1109/ECAI46879.2019.9042002
Anamaria-Loredana Chiva, G. Suciu
The increasing number of cars brings upon itself the need of rapidly check the lawfulness of the drivers. When travelling, there are several legal obligations a driver have to fulfill according to the law. In Romania, the authorities enforce the presence of legal documents such vignette, RCA (civil liability) and PTI/ITP (periodical technical inspection) when being subject to a routine control. This article aims to present a solution to ease the process of verification through a software solution which identifies the number-plate and then verifies in a unified database the availability of the above-mentioned legal documents. This solution is implemented in Matlab, using correlation factor between the identified number-plate and a set of characters (the alphabet) to determine the letters from the plate number. The program is connected to a database in which are saved the details (vignette, RCA, PTI) of the identified number, this being used by the authorities to verify the validity of the drivers' documents.
{"title":"Number-plate recognition and examination of requirements for public Romanian roads usage","authors":"Anamaria-Loredana Chiva, G. Suciu","doi":"10.1109/ECAI46879.2019.9042002","DOIUrl":"https://doi.org/10.1109/ECAI46879.2019.9042002","url":null,"abstract":"The increasing number of cars brings upon itself the need of rapidly check the lawfulness of the drivers. When travelling, there are several legal obligations a driver have to fulfill according to the law. In Romania, the authorities enforce the presence of legal documents such vignette, RCA (civil liability) and PTI/ITP (periodical technical inspection) when being subject to a routine control. This article aims to present a solution to ease the process of verification through a software solution which identifies the number-plate and then verifies in a unified database the availability of the above-mentioned legal documents. This solution is implemented in Matlab, using correlation factor between the identified number-plate and a set of characters (the alphabet) to determine the letters from the plate number. The program is connected to a database in which are saved the details (vignette, RCA, PTI) of the identified number, this being used by the authorities to verify the validity of the drivers' documents.","PeriodicalId":285780,"journal":{"name":"2019 11th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122126758","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 : 2019-06-01DOI: 10.1109/ECAI46879.2019.9042077
V. Gnatyuk, S. Zavalishin, X. Petrova, G. Odinokikh, A. Fartukov, Alexey Danilevich, V. Eremeev, Ju-Hee Yoo, Kwanghyun Lee, Heejun Lee, Daekyu Shin, I. Solomatin
In this paper, we propose a novel algorithm for automatic camera parameter adjustment, which is exploited for getting the correct image exposure required for iris recognition. We use two-step processing, where the first step adjusts the camera parameters on the basis of a single shot, and the second step applies precise iterative adjustment. In order to get the correct iris exposure, we use a weighted mask, which is constructed offline using a set of face images. In contrast to the existing algorithms, our method does not need to be calibrated for a particular camera sensor. We show that the proposed method significantly decreases false rejection rate caused by incorrect image exposure and reduces recognition time.
{"title":"Fast Automatic Exposure Adjustment Method for Iris Recognition System","authors":"V. Gnatyuk, S. Zavalishin, X. Petrova, G. Odinokikh, A. Fartukov, Alexey Danilevich, V. Eremeev, Ju-Hee Yoo, Kwanghyun Lee, Heejun Lee, Daekyu Shin, I. Solomatin","doi":"10.1109/ECAI46879.2019.9042077","DOIUrl":"https://doi.org/10.1109/ECAI46879.2019.9042077","url":null,"abstract":"In this paper, we propose a novel algorithm for automatic camera parameter adjustment, which is exploited for getting the correct image exposure required for iris recognition. We use two-step processing, where the first step adjusts the camera parameters on the basis of a single shot, and the second step applies precise iterative adjustment. In order to get the correct iris exposure, we use a weighted mask, which is constructed offline using a set of face images. In contrast to the existing algorithms, our method does not need to be calibrated for a particular camera sensor. We show that the proposed method significantly decreases false rejection rate caused by incorrect image exposure and reduces recognition time.","PeriodicalId":285780,"journal":{"name":"2019 11th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125064634","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 : 2019-06-01DOI: 10.1109/ECAI46879.2019.9042157
M. Ianculescu, Lidia Bajenaru, Ion Alexandru Marinescu, C. Dobre
Digital technology landscape is continuously improving, dragging along both the transformation of public services and new demands of citizens. Emerging new technologies like Artificial Intelligence, Machine Learning, Deep Learning or Internet of Things provide tremendous means to implement intelligent solutions for reshaping digital public services. This paper aims to disclose the most important features of several intelligent technologies and of these types of public services that can be integrated for providing new capabilities. An AI-based architecture for supporting digital public services in the smart transportation sector is presented in order to demonstrate the highlighted ideas and concepts.
{"title":"Intelligent solutions - based framework for digital public services. A case study for smart transportation","authors":"M. Ianculescu, Lidia Bajenaru, Ion Alexandru Marinescu, C. Dobre","doi":"10.1109/ECAI46879.2019.9042157","DOIUrl":"https://doi.org/10.1109/ECAI46879.2019.9042157","url":null,"abstract":"Digital technology landscape is continuously improving, dragging along both the transformation of public services and new demands of citizens. Emerging new technologies like Artificial Intelligence, Machine Learning, Deep Learning or Internet of Things provide tremendous means to implement intelligent solutions for reshaping digital public services. This paper aims to disclose the most important features of several intelligent technologies and of these types of public services that can be integrated for providing new capabilities. An AI-based architecture for supporting digital public services in the smart transportation sector is presented in order to demonstrate the highlighted ideas and concepts.","PeriodicalId":285780,"journal":{"name":"2019 11th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130212559","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 : 2019-06-01DOI: 10.1109/ECAI46879.2019.9041960
Hadeer Ahmed, G. Alkady, Hassan H. Halawa, R. Daoud, H. Amer, I. Adly, T. Refaat, Manar N. Shaker
Networked Control Systems (NCSs) form an interesting architecture for industrial applications. They are composed of smart sensors, actuators and controllers. Also, FPGAs have recently been progressively used in industry. One of the main advantages of FPGAs is the feature of Dynamic Partial Reconfiguration (DPR) which is used to repair some failures without interrupting system operation. This paper combined both NCS and FPGAs in a new architecture that has several advantages. First, it includes an image storing capability which is very important in certain applications. Second, the architecture has several fault-tolerant features in order to prolong system lifetime. Third, even in the case of failure, raw video is never sent over the network. A case study is presented to quantify the increase in reliability of the presented system. Markov models were developed and the SHARPE package was used to compare the reliability of the proposed system to that of a comparable system in the literature; the results indicate that the proposed system is more reliable.
{"title":"Reliable FPGA-based Camera Sensor for NCS","authors":"Hadeer Ahmed, G. Alkady, Hassan H. Halawa, R. Daoud, H. Amer, I. Adly, T. Refaat, Manar N. Shaker","doi":"10.1109/ECAI46879.2019.9041960","DOIUrl":"https://doi.org/10.1109/ECAI46879.2019.9041960","url":null,"abstract":"Networked Control Systems (NCSs) form an interesting architecture for industrial applications. They are composed of smart sensors, actuators and controllers. Also, FPGAs have recently been progressively used in industry. One of the main advantages of FPGAs is the feature of Dynamic Partial Reconfiguration (DPR) which is used to repair some failures without interrupting system operation. This paper combined both NCS and FPGAs in a new architecture that has several advantages. First, it includes an image storing capability which is very important in certain applications. Second, the architecture has several fault-tolerant features in order to prolong system lifetime. Third, even in the case of failure, raw video is never sent over the network. A case study is presented to quantify the increase in reliability of the presented system. Markov models were developed and the SHARPE package was used to compare the reliability of the proposed system to that of a comparable system in the literature; the results indicate that the proposed system is more reliable.","PeriodicalId":285780,"journal":{"name":"2019 11th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126441788","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 : 2019-06-01DOI: 10.1109/ECAI46879.2019.9041962
Qixin Xie, Chao Li, Boyu Diao, Zhulin An, Yongjun Xu
Deep neural networks have made remarkable achievements in many tasks. However, we are not able to deploy such successful but heavy models on mobile devices directly due to the limited power and computing capacity. Thus, an obvious solution to tackle this problem is to compress neural network by pruning useless weights in neural networks. The point is how to remove these redundancies while maintain the performance of neural networks. In this work, we propose a novel neural network pruning method: guiding the weights of a neural network to be sparse by introducing LO regularization during the training stage, which can effectively resist the damage on the performance while pruning as well as dramatically reduce the time overhead of retraining stage. Experiment results on MNIST with LeNet and CIFAR-10 with VGG-16 demonstrate the effectiveness of this method to the classic method.
{"title":"L0 Regularization based Fine-grained Neural Network Pruning Method","authors":"Qixin Xie, Chao Li, Boyu Diao, Zhulin An, Yongjun Xu","doi":"10.1109/ECAI46879.2019.9041962","DOIUrl":"https://doi.org/10.1109/ECAI46879.2019.9041962","url":null,"abstract":"Deep neural networks have made remarkable achievements in many tasks. However, we are not able to deploy such successful but heavy models on mobile devices directly due to the limited power and computing capacity. Thus, an obvious solution to tackle this problem is to compress neural network by pruning useless weights in neural networks. The point is how to remove these redundancies while maintain the performance of neural networks. In this work, we propose a novel neural network pruning method: guiding the weights of a neural network to be sparse by introducing LO regularization during the training stage, which can effectively resist the damage on the performance while pruning as well as dramatically reduce the time overhead of retraining stage. Experiment results on MNIST with LeNet and CIFAR-10 with VGG-16 demonstrate the effectiveness of this method to the classic method.","PeriodicalId":285780,"journal":{"name":"2019 11th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)","volume":"135 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127594320","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 : 2019-06-01DOI: 10.1109/ECAI46879.2019.9042037
Cihan Aydin, I. Sefa
Latest generation FPGAs determine the future usage of FIR filters. Their DSP blocks are able to implement fixed-point data types for efficient computations. The systolic multiply-accumulate architecture is utilized for various order and parallel multiple channel to efficiently handle resource and timing considerations. Implementing various order filter taps, resource and latency of the particular architecture of Xilinx Artix-7 (XC7A100T-1CSG324C) series with the clock frequency of 100 MHz and 12 bit input and 12 bit output is observed. The proposed design also shows that this design is suitable for multichannel parallel implementation such as power electronics applications in smart grids.
{"title":"FPGA Implementation of Multichannel FIR Filters","authors":"Cihan Aydin, I. Sefa","doi":"10.1109/ECAI46879.2019.9042037","DOIUrl":"https://doi.org/10.1109/ECAI46879.2019.9042037","url":null,"abstract":"Latest generation FPGAs determine the future usage of FIR filters. Their DSP blocks are able to implement fixed-point data types for efficient computations. The systolic multiply-accumulate architecture is utilized for various order and parallel multiple channel to efficiently handle resource and timing considerations. Implementing various order filter taps, resource and latency of the particular architecture of Xilinx Artix-7 (XC7A100T-1CSG324C) series with the clock frequency of 100 MHz and 12 bit input and 12 bit output is observed. The proposed design also shows that this design is suitable for multichannel parallel implementation such as power electronics applications in smart grids.","PeriodicalId":285780,"journal":{"name":"2019 11th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115667053","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：