Pub Date : 2019-04-01DOI: 10.1109/ICOEI.2019.8862698
Ulagamuthalvi., J.B. Janet Felicita, D. Abinaya
Image processing and computer vision have gained an enormous advance in the field of machine learning techniques. Some of the major research areas within machine learning are Object detection and Scene Recognition. Though there are numerous existing works related to the specified fields object detection still encounters numerous challenges when it comes to implementing in the real-time scenario. The problem occurs in the detection due to various objects present in the background. Object detection mechanism detects a specified object when a particular scene is given. Classifiers like SVM and Neural Networks are used to train the classifier in such a way they are able to detect an object when a new image is given. In this paper, we have proposed a model which detects texts from an image. Bounding boxes are used to detect the texts and localize it. The neural network is used to train the model where numerous images having texts are given as the training set. The performance evaluation is done on the model and it is observed that it detects the texts when a new image is given. Object detection is a fundamental problem in computer vision, which aims to detect general objects in images.
{"title":"An Efficient Object Detection Model Using Convolution Neural Networks","authors":"Ulagamuthalvi., J.B. Janet Felicita, D. Abinaya","doi":"10.1109/ICOEI.2019.8862698","DOIUrl":"https://doi.org/10.1109/ICOEI.2019.8862698","url":null,"abstract":"Image processing and computer vision have gained an enormous advance in the field of machine learning techniques. Some of the major research areas within machine learning are Object detection and Scene Recognition. Though there are numerous existing works related to the specified fields object detection still encounters numerous challenges when it comes to implementing in the real-time scenario. The problem occurs in the detection due to various objects present in the background. Object detection mechanism detects a specified object when a particular scene is given. Classifiers like SVM and Neural Networks are used to train the classifier in such a way they are able to detect an object when a new image is given. In this paper, we have proposed a model which detects texts from an image. Bounding boxes are used to detect the texts and localize it. The neural network is used to train the model where numerous images having texts are given as the training set. The performance evaluation is done on the model and it is observed that it detects the texts when a new image is given. Object detection is a fundamental problem in computer vision, which aims to detect general objects in images.","PeriodicalId":212501,"journal":{"name":"2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134352060","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 : 2019-04-01DOI: 10.1109/ICOEI.2019.8862603
Madhumitha Nara, B. Mukesh, Preethi Padala, Bharath A. Kinnal
Deep Learning (DL) applications have skyrocketed in recent years and are being applied in various domains. There has been a tremendous surge in the development of DL frameworks to make implementation easier. In this paper, we aim to make a comparative study of GPU-accelerated deep learning software frameworks such as Torch and TenserFlow (with Keras API). We attempt to benchmark the performance of these frameworks by implementing three different neural networks, each designed for a popular Computer Vision problem (MNIST, CIFAR10, Fashion MNIST). We performed this experiment on both CPU and GPU(Nvidia GeForce GTX 960M) settings. The performance metrics used here include evaluation time, training time, and accuracy. This paper aims to act as a guide to selecting the most suitable framework for a particular problem. The special interest of the paper is to evaluate the performance lost due to the utility of an API like Keras and a comparative study of the performance over a user-defined neural network and a standard network. Our interest also lies in their performance when subjected to networks of different sizes.
{"title":"Performance Evaluation of Deep Learning frameworks on Computer Vision problems","authors":"Madhumitha Nara, B. Mukesh, Preethi Padala, Bharath A. Kinnal","doi":"10.1109/ICOEI.2019.8862603","DOIUrl":"https://doi.org/10.1109/ICOEI.2019.8862603","url":null,"abstract":"Deep Learning (DL) applications have skyrocketed in recent years and are being applied in various domains. There has been a tremendous surge in the development of DL frameworks to make implementation easier. In this paper, we aim to make a comparative study of GPU-accelerated deep learning software frameworks such as Torch and TenserFlow (with Keras API). We attempt to benchmark the performance of these frameworks by implementing three different neural networks, each designed for a popular Computer Vision problem (MNIST, CIFAR10, Fashion MNIST). We performed this experiment on both CPU and GPU(Nvidia GeForce GTX 960M) settings. The performance metrics used here include evaluation time, training time, and accuracy. This paper aims to act as a guide to selecting the most suitable framework for a particular problem. The special interest of the paper is to evaluate the performance lost due to the utility of an API like Keras and a comparative study of the performance over a user-defined neural network and a standard network. Our interest also lies in their performance when subjected to networks of different sizes.","PeriodicalId":212501,"journal":{"name":"2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132595406","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 : 2019-04-01DOI: 10.1109/ICOEI.2019.8862742
E. Brumancia, R. Gomathi, D.Ayyappa Naidu, D. Srikanth
Diverse thriving applications in Vehicular Ad hoc Networks (VANETs) are based on offer. Engineering a pass on tradition that satisfies VANET applications' necessities is fundamental. In this paper, we propose a strong and mind boggling multi-weave pass on arranging custom for VANETs. The proposed tradition gives the strict consistent quality in various flood hour gridlock conditions. This tradition other than performs low overhead by systems for decreasing rebroadcast abundance in a high-thickness build condition. We furthermore propose a redesigned multipoint hand-off (MPR) decision watch that thinks about vehicles' movability and after that utilization it for exchange center point decision. We show the execution examination of the proposed custom by age with ns-2 in different conditions, and give the preoccupation results appearing of the proposed tradition isolated and other VANET present designs.
{"title":"Contention Based Forwarding Message in Vanet After an Emergency Event","authors":"E. Brumancia, R. Gomathi, D.Ayyappa Naidu, D. Srikanth","doi":"10.1109/ICOEI.2019.8862742","DOIUrl":"https://doi.org/10.1109/ICOEI.2019.8862742","url":null,"abstract":"Diverse thriving applications in Vehicular Ad hoc Networks (VANETs) are based on offer. Engineering a pass on tradition that satisfies VANET applications' necessities is fundamental. In this paper, we propose a strong and mind boggling multi-weave pass on arranging custom for VANETs. The proposed tradition gives the strict consistent quality in various flood hour gridlock conditions. This tradition other than performs low overhead by systems for decreasing rebroadcast abundance in a high-thickness build condition. We furthermore propose a redesigned multipoint hand-off (MPR) decision watch that thinks about vehicles' movability and after that utilization it for exchange center point decision. We show the execution examination of the proposed custom by age with ns-2 in different conditions, and give the preoccupation results appearing of the proposed tradition isolated and other VANET present designs.","PeriodicalId":212501,"journal":{"name":"2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127660947","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 : 2019-04-01DOI: 10.1109/ICOEI.2019.8862789
V. Veena, E. Prabhu, N. Mohan
Design for testing implies on adding an extra hardware to circuit under test so that the difficulty in testing the circuit becomes easy and large number of faults can be detected to increase the test coverage of the circuit. In a circuit there might be a large number of faults and some fault in the circuit will have high controllability as well as observability those faults will be very difficult to detect. A new approach has been introduced which involves the insertion of observation points at most suitable location to capture the most difficult to observe faults which facilitate structural testing for both on-chip and off-chip for better fault coverage. Insertion of the observation point into the internal part of the circuit enables direct observation of the internal part of the circuit. The observation points are inserted at those locations where observability is high and the occurrence of fault at those location makes the faults hard to propagate to the output.
{"title":"Improved Test Coverage by Observation Point Insertion for Fault Coverage Analysis","authors":"V. Veena, E. Prabhu, N. Mohan","doi":"10.1109/ICOEI.2019.8862789","DOIUrl":"https://doi.org/10.1109/ICOEI.2019.8862789","url":null,"abstract":"Design for testing implies on adding an extra hardware to circuit under test so that the difficulty in testing the circuit becomes easy and large number of faults can be detected to increase the test coverage of the circuit. In a circuit there might be a large number of faults and some fault in the circuit will have high controllability as well as observability those faults will be very difficult to detect. A new approach has been introduced which involves the insertion of observation points at most suitable location to capture the most difficult to observe faults which facilitate structural testing for both on-chip and off-chip for better fault coverage. Insertion of the observation point into the internal part of the circuit enables direct observation of the internal part of the circuit. The observation points are inserted at those locations where observability is high and the occurrence of fault at those location makes the faults hard to propagate to the output.","PeriodicalId":212501,"journal":{"name":"2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124359409","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 : 2019-04-01DOI: 10.1109/ICOEI.2019.8862780
T. Kavitha, S. Silas
Wireless Sensor Networks plays a vital in our day to day life in various applications such as healthcare, smart home, smart classroom, and in remote military surveillance. Some of the challenges that our Wireless Sensor Networks faces to combat with the energy saving issues are clustering, choosing the cluster head, data aggregation, routing techniques and protocols. This aims at reviewing various techniques and protocols that can be applied in WSN for efficient usage of the energy constraint and for extending the lifespan of the network in various aspects of routing and its protocols. This paper summarizes the advantage, disadvantages and the open research issues and challenges in combating the energy usage efficiently.
{"title":"A Survey on Energy Harvesting Routing Protocol for WSN","authors":"T. Kavitha, S. Silas","doi":"10.1109/ICOEI.2019.8862780","DOIUrl":"https://doi.org/10.1109/ICOEI.2019.8862780","url":null,"abstract":"Wireless Sensor Networks plays a vital in our day to day life in various applications such as healthcare, smart home, smart classroom, and in remote military surveillance. Some of the challenges that our Wireless Sensor Networks faces to combat with the energy saving issues are clustering, choosing the cluster head, data aggregation, routing techniques and protocols. This aims at reviewing various techniques and protocols that can be applied in WSN for efficient usage of the energy constraint and for extending the lifespan of the network in various aspects of routing and its protocols. This paper summarizes the advantage, disadvantages and the open research issues and challenges in combating the energy usage efficiently.","PeriodicalId":212501,"journal":{"name":"2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116369856","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 : 2019-04-01DOI: 10.1109/ICOEI.2019.8862571
Sheetal Shivagunde, M. Biswas
Nowadays Super Resolution (SR) is a trending term in many fields of image processing area where resolution describes the degree of details an image holds like, pixel count, sharpness and clarity. In many applications captured images are of low quality known as low resolution (LR) image, these images hold very less details. Therefore there is a need to convert such LR images to High resolution (HR) images which can be obtained by various SR methods like, interpolation methods, frequency domain based methods, reconstruction based methods and learning based methods. HR images obtained from interpolation methods contain reconstruction artifacts like edge blurs, edge halos and ringing effects, whereas learning based methods produce good results but predict unknown HR pixel values based only on input LR/HR image pairs. Thus, to overcome above mentioned disadvantages we proposed modified interpolation method for obtaining HR image, which predicts unknown HR pixel values from interpolated LR patch and corresponding HR patch using multilayer perceptron (MLP) and discrete wavelet transform (DWT). Experimental results subjectively and objectively show that for considered test images corresponding HR images obtained by our proposed method are of better quality than classical bicubic, Chopade et al. and Man et al. method.
{"title":"Single Image Super-Resolution Based on Modified Interpolation Method Using MLP and DWT","authors":"Sheetal Shivagunde, M. Biswas","doi":"10.1109/ICOEI.2019.8862571","DOIUrl":"https://doi.org/10.1109/ICOEI.2019.8862571","url":null,"abstract":"Nowadays Super Resolution (SR) is a trending term in many fields of image processing area where resolution describes the degree of details an image holds like, pixel count, sharpness and clarity. In many applications captured images are of low quality known as low resolution (LR) image, these images hold very less details. Therefore there is a need to convert such LR images to High resolution (HR) images which can be obtained by various SR methods like, interpolation methods, frequency domain based methods, reconstruction based methods and learning based methods. HR images obtained from interpolation methods contain reconstruction artifacts like edge blurs, edge halos and ringing effects, whereas learning based methods produce good results but predict unknown HR pixel values based only on input LR/HR image pairs. Thus, to overcome above mentioned disadvantages we proposed modified interpolation method for obtaining HR image, which predicts unknown HR pixel values from interpolated LR patch and corresponding HR patch using multilayer perceptron (MLP) and discrete wavelet transform (DWT). Experimental results subjectively and objectively show that for considered test images corresponding HR images obtained by our proposed method are of better quality than classical bicubic, Chopade et al. and Man et al. method.","PeriodicalId":212501,"journal":{"name":"2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123531140","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 : 2019-04-01DOI: 10.1109/ICOEI.2019.8862722
Jones S.B Ribin, N. Kumar
Cloud Computing has emerged into an inevitable platform for computing services by effectively implementing Service Oriented Architecture (SOA) and Virtualization. However it is still vulnerable to traditional security threats and offers scope for innovative security attacks such as EDoS [1]. While it offers platform to generate innumerable Virtual components from a single physical component, it inadvertently provides wide spectrum of possibilities for distributed attacks. Moreover such attacks have adapted to cloud platform and have exploited various inherent vulnerabilities. In an unprecedented manner, they became unpredictable, evasive and challenging to Cloud Security measures. Therefore various versions of DDoS attack that targets the cloud platform have been extensively researched and narrated. The Cloud Security faces unprecedented challenges such as the Single-point-of-Failure occurs when a Cloud Supervisory Component or hypervisor fails due to a security breach. Moreover Cloud requirements often require being liberal to meet the Clients needs. This does not help the CSP to adapt traditional stringent security measures in Cloud System the reasons have been discussed in details.
{"title":"Precursory study on varieties of DDoS attacks and its implications in Cloud Systems","authors":"Jones S.B Ribin, N. Kumar","doi":"10.1109/ICOEI.2019.8862722","DOIUrl":"https://doi.org/10.1109/ICOEI.2019.8862722","url":null,"abstract":"Cloud Computing has emerged into an inevitable platform for computing services by effectively implementing Service Oriented Architecture (SOA) and Virtualization. However it is still vulnerable to traditional security threats and offers scope for innovative security attacks such as EDoS [1]. While it offers platform to generate innumerable Virtual components from a single physical component, it inadvertently provides wide spectrum of possibilities for distributed attacks. Moreover such attacks have adapted to cloud platform and have exploited various inherent vulnerabilities. In an unprecedented manner, they became unpredictable, evasive and challenging to Cloud Security measures. Therefore various versions of DDoS attack that targets the cloud platform have been extensively researched and narrated. The Cloud Security faces unprecedented challenges such as the Single-point-of-Failure occurs when a Cloud Supervisory Component or hypervisor fails due to a security breach. Moreover Cloud requirements often require being liberal to meet the Clients needs. This does not help the CSP to adapt traditional stringent security measures in Cloud System the reasons have been discussed in details.","PeriodicalId":212501,"journal":{"name":"2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121865829","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 : 2019-04-01DOI: 10.1109/ICOEI.2019.8862575
G. Madhupriya, Narayanan M Guru, S. Praveen, B. Nivetha
The proposed work is based on Deep learning technique which is a deep neural network and probabilistic neural network to detect unwanted masses in the brain. Our work is personalized for both high and low-level grades. Tumors can appear in anyplace of the brain and its natures like shape, contrast, and size have always been an uncertain one, which means that there is no standard fact about tumor structure. The rate at which people woe from brain tumor becomes increasing nowadays. These reasons stimulate us to provide an intelligent solution which uses deep learning technique to segment abnormal tissues in the brain. It can help to find out whether the tumor is in the brain or not. With the help of these MRI images, segmentation can be performed and the segmented images can be compared with the normal brain tissues also with the tumor cells. The results are provided (whether the brain contains a tumor or not) based on the comparison. In this paper, the segmentation is done using a convolution neural network and Probabilistic neural network. Here, the comparison sketch of various models is done. Based on that, we discovered an architecture which is based on Convolutional Neural Networks (CNN) with both $3^{ast} 3$ and $7^{ast} 7$ in an overlapped manner, and build a cascaded architecture, so that we can able to segment a tumor accurately in an effective manner, since we use Image dataset Brats13. Similarly, we use a probabilistic neural network for detecting tumors and compare the result of both of them. We proposed a unique CNN and PNN architectures which are different from those conventional models used in image processing and computer vision techniques. Our model deals with both local and global features.
{"title":"Brain Tumor Segmentation with Deep Learning Technique","authors":"G. Madhupriya, Narayanan M Guru, S. Praveen, B. Nivetha","doi":"10.1109/ICOEI.2019.8862575","DOIUrl":"https://doi.org/10.1109/ICOEI.2019.8862575","url":null,"abstract":"The proposed work is based on Deep learning technique which is a deep neural network and probabilistic neural network to detect unwanted masses in the brain. Our work is personalized for both high and low-level grades. Tumors can appear in anyplace of the brain and its natures like shape, contrast, and size have always been an uncertain one, which means that there is no standard fact about tumor structure. The rate at which people woe from brain tumor becomes increasing nowadays. These reasons stimulate us to provide an intelligent solution which uses deep learning technique to segment abnormal tissues in the brain. It can help to find out whether the tumor is in the brain or not. With the help of these MRI images, segmentation can be performed and the segmented images can be compared with the normal brain tissues also with the tumor cells. The results are provided (whether the brain contains a tumor or not) based on the comparison. In this paper, the segmentation is done using a convolution neural network and Probabilistic neural network. Here, the comparison sketch of various models is done. Based on that, we discovered an architecture which is based on Convolutional Neural Networks (CNN) with both $3^{ast} 3$ and $7^{ast} 7$ in an overlapped manner, and build a cascaded architecture, so that we can able to segment a tumor accurately in an effective manner, since we use Image dataset Brats13. Similarly, we use a probabilistic neural network for detecting tumors and compare the result of both of them. We proposed a unique CNN and PNN architectures which are different from those conventional models used in image processing and computer vision techniques. Our model deals with both local and global features.","PeriodicalId":212501,"journal":{"name":"2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125838637","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 : 2019-04-01DOI: 10.1109/ICOEI.2019.8862641
T. Anandhi, Kumaar V.S. Kishore, Ganesh S. Maha, R. Gomathi
We present a highly efficient method for vehicle parking management in large scale parking lots based on IoT. The proposed system can be installed in a Raspberry pi module which will help us in solving the current problems in vehicle parking management. The system consists of an on-site deployment of an IoT module that is used to monitor and signalize the state of availability of each single parking slot. This can be done on a mobile phone using android application. This application can be managed by an administrator at the malls. This project primarily makes finding of parking systems more efficient and less time-consuming, and also eliminates the need of staff to be employed for such purpose of these places, thus increasing profit margins. The system helps a user know the availability of parking slots on a real time basis. Android mobile application is used to display the total parking slots in the parking lot, number of parking slots available, occupied parking slots and the reserved parking slots. To book the parking slot at that time your information are encoded to the QR code. This makes our parking system more secured. Once you park the vehicle in the slot, the timer will start to monitor the parking time. If time exceeds alert message is sent to the user's mobile.
{"title":"A Sustainable Vehicle Parking using IoT","authors":"T. Anandhi, Kumaar V.S. Kishore, Ganesh S. Maha, R. Gomathi","doi":"10.1109/ICOEI.2019.8862641","DOIUrl":"https://doi.org/10.1109/ICOEI.2019.8862641","url":null,"abstract":"We present a highly efficient method for vehicle parking management in large scale parking lots based on IoT. The proposed system can be installed in a Raspberry pi module which will help us in solving the current problems in vehicle parking management. The system consists of an on-site deployment of an IoT module that is used to monitor and signalize the state of availability of each single parking slot. This can be done on a mobile phone using android application. This application can be managed by an administrator at the malls. This project primarily makes finding of parking systems more efficient and less time-consuming, and also eliminates the need of staff to be employed for such purpose of these places, thus increasing profit margins. The system helps a user know the availability of parking slots on a real time basis. Android mobile application is used to display the total parking slots in the parking lot, number of parking slots available, occupied parking slots and the reserved parking slots. To book the parking slot at that time your information are encoded to the QR code. This makes our parking system more secured. Once you park the vehicle in the slot, the timer will start to monitor the parking time. If time exceeds alert message is sent to the user's mobile.","PeriodicalId":212501,"journal":{"name":"2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124602746","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 : 2019-04-01DOI: 10.1109/ICOEI.2019.8862730
K. Yazhini, D. Loganathan
Acquisition of knowledge and actionable insights from complex, high-dimensional and nonhomogeneous healthcare data still remains a major difficulty in the evolving health care applications. Different data types have been emerged in the advanced healthcare research area such as maintaining patient's records, imaging, sensors data and content that are not simple, nonhomogeneous, badly annotated and normally not structured well. Conventional data mining and machine learning methods has been executing feature engineering to attain efficient and highly robust features from the data, and then constructs a model to predict or cluster data. Several difficulties exist in the situation of complex information and insufficient domain information. The recent advancements in the Deep Learning (DL) models offer novel and efficient end to end frameworks for health care data. In this study, we attempt to survey the recently presented DL models in the advanced medicinal filed in various aspects.
{"title":"A State of Art Approaches on Deep Learning Models in Healthcare: An Application Perspective","authors":"K. Yazhini, D. Loganathan","doi":"10.1109/ICOEI.2019.8862730","DOIUrl":"https://doi.org/10.1109/ICOEI.2019.8862730","url":null,"abstract":"Acquisition of knowledge and actionable insights from complex, high-dimensional and nonhomogeneous healthcare data still remains a major difficulty in the evolving health care applications. Different data types have been emerged in the advanced healthcare research area such as maintaining patient's records, imaging, sensors data and content that are not simple, nonhomogeneous, badly annotated and normally not structured well. Conventional data mining and machine learning methods has been executing feature engineering to attain efficient and highly robust features from the data, and then constructs a model to predict or cluster data. Several difficulties exist in the situation of complex information and insufficient domain information. The recent advancements in the Deep Learning (DL) models offer novel and efficient end to end frameworks for health care data. In this study, we attempt to survey the recently presented DL models in the advanced medicinal filed in various aspects.","PeriodicalId":212501,"journal":{"name":"2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129788231","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：