Pub Date : 2022-12-01DOI: 10.1109/ICECA55336.2022.10009107
S. L. Kumar, M. Swathy, T. Arunkumar, M. Maniventhan, S. Vigneshwaran
Electromyograph (EMG)-based prosthetic devices use a classifier to identify the myoelectric signals obtained from the amputee's limb. The classifier's output then controls the movement of the prosthetic device. The spectrum of the EMG signal is known to dynamic change with muscle length. These changes are reflected in the classifier's output, and so they are an important component in constructing a strong classification system to determine the user's intent. This work presents a method for dynamically changing the sample frequency of the EMG based on the spectrum of the input myoelectric signal. The shift in sampling frequency allows for correction of muscle length variations and the resulting resilience in signal classification. The approach was successfully validated and implemented on simulated EMG. The results indicate that variations in classifier output caused by changes in muscle length may be compensated for by changes in the spectrum.
{"title":"Analysis of dynamic change in sampling rate of EMG signal for designing prosthesis control","authors":"S. L. Kumar, M. Swathy, T. Arunkumar, M. Maniventhan, S. Vigneshwaran","doi":"10.1109/ICECA55336.2022.10009107","DOIUrl":"https://doi.org/10.1109/ICECA55336.2022.10009107","url":null,"abstract":"Electromyograph (EMG)-based prosthetic devices use a classifier to identify the myoelectric signals obtained from the amputee's limb. The classifier's output then controls the movement of the prosthetic device. The spectrum of the EMG signal is known to dynamic change with muscle length. These changes are reflected in the classifier's output, and so they are an important component in constructing a strong classification system to determine the user's intent. This work presents a method for dynamically changing the sample frequency of the EMG based on the spectrum of the input myoelectric signal. The shift in sampling frequency allows for correction of muscle length variations and the resulting resilience in signal classification. The approach was successfully validated and implemented on simulated EMG. The results indicate that variations in classifier output caused by changes in muscle length may be compensated for by changes in the spectrum.","PeriodicalId":356949,"journal":{"name":"2022 6th International Conference on Electronics, Communication and Aerospace Technology","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127261998","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 : 2022-12-01DOI: 10.1109/ICECA55336.2022.10009351
Vallepu Rambabu, K. Malathi, R. Mahaveerakannan
To compare novel LR with the SVM technique to estimate the precision of phishing websites. Materials and Methods: The SVM method's algorithm for supervised learning (N = 20) is compared to the Logistic Regression algorithm's supervised learning algorithm (N = 20). To achieve great precision, the G power value is set to 0.8. Machine Learning is used in the framework. Compared to the SVM approach, LR has more precision (92.00%). (90.26%). With a confidence value of 95%, the impartial T-Test was run (p =.375), indicating the importance score that is statistically insignificant (p>0.05). Conclusion: The LR approach appeared to detect phishing websites with greater accuracy than the SVM technique.
{"title":"An Innovative Method to Predict the Accuracy of Phishing Websites by Comparing Logistic Regression Algorithm with Support Vector Machine Algorithm","authors":"Vallepu Rambabu, K. Malathi, R. Mahaveerakannan","doi":"10.1109/ICECA55336.2022.10009351","DOIUrl":"https://doi.org/10.1109/ICECA55336.2022.10009351","url":null,"abstract":"To compare novel LR with the SVM technique to estimate the precision of phishing websites. Materials and Methods: The SVM method's algorithm for supervised learning (N = 20) is compared to the Logistic Regression algorithm's supervised learning algorithm (N = 20). To achieve great precision, the G power value is set to 0.8. Machine Learning is used in the framework. Compared to the SVM approach, LR has more precision (92.00%). (90.26%). With a confidence value of 95%, the impartial T-Test was run (p =.375), indicating the importance score that is statistically insignificant (p>0.05). Conclusion: The LR approach appeared to detect phishing websites with greater accuracy than the SVM technique.","PeriodicalId":356949,"journal":{"name":"2022 6th International Conference on Electronics, Communication and Aerospace Technology","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127472677","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 : 2022-12-01DOI: 10.1109/ICECA55336.2022.10009487
Chetan Pandey, Sachin Sharma, Priya Matta
Data are now processed relatively in an efficient manner due to the development of machine learning techniques. Such strategies for knowledge extraction are frequently employed in a variety of contexts, including business, social media, voting, wagering, forecasting, and more. Healthcare in Body Sensor Network is one of these key fields where modelling and data analysis are extensively used. The data that is captured and processed in this network is used to track a person's everyday activities, check that the data is accurate, determine when a medical emergency is required, and more. There are sufficient studies based on such analysis; some offered their own methodology while others employed pre-defined techniques such as Machine Learning, Neural Networks, Deep Learning, and more. In order to analysis the sensor data, various methodologies that have been stated in some selected research articles are compared in this document. Both the analysis methods and the study's findings are very diverse and have many unique characteristics. The comparison study provides a comprehensible demonstration of these methods and features.
{"title":"Data Analysis and Modeling of Body Sensor Network in Healthcare Application","authors":"Chetan Pandey, Sachin Sharma, Priya Matta","doi":"10.1109/ICECA55336.2022.10009487","DOIUrl":"https://doi.org/10.1109/ICECA55336.2022.10009487","url":null,"abstract":"Data are now processed relatively in an efficient manner due to the development of machine learning techniques. Such strategies for knowledge extraction are frequently employed in a variety of contexts, including business, social media, voting, wagering, forecasting, and more. Healthcare in Body Sensor Network is one of these key fields where modelling and data analysis are extensively used. The data that is captured and processed in this network is used to track a person's everyday activities, check that the data is accurate, determine when a medical emergency is required, and more. There are sufficient studies based on such analysis; some offered their own methodology while others employed pre-defined techniques such as Machine Learning, Neural Networks, Deep Learning, and more. In order to analysis the sensor data, various methodologies that have been stated in some selected research articles are compared in this document. Both the analysis methods and the study's findings are very diverse and have many unique characteristics. The comparison study provides a comprehensible demonstration of these methods and features.","PeriodicalId":356949,"journal":{"name":"2022 6th International Conference on Electronics, Communication and Aerospace Technology","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129986502","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 : 2022-12-01DOI: 10.1109/ICECA55336.2022.10009197
C. Srinivasan, P. Sridhar, V. Hari Priya, S. Swathi
Image processing is an important requirement in IoT applications such as robotics, augmented reality, computer vision, Industry 4.0 etc. The capabilities of IoT devices for image processing are limited to sensing the environment, processing and communicating the results. Tiny machine learning (TinyML) is a new paradigm that takes advantage of the IoT to deploy deep learning models to perform complex tasks in resource constrained embedded devices. Image classification is an important task in IoT to interpret images of a particular scene or class. Currently, this task is performed in embedded devices using Binarized Neural Networks (BNNs), which can be converted to a set of weights using a one-hot encoding process. These networks integrated with hardware accelerators can be trained to perform image processing tasks in real-time. This paper proposes a BNN for image classification based on residual learning paradigm, called Tiny-BNN which exploits the skip connections to reduce information loss, and improve the training time and accuracy. Experimental results show that the model achieves a classification accuracy of 90.1 % and 91.6% on the on CIFAR-10 and MNIST datasets respectively.
{"title":"A TinyML based Residual Binarized Neural Network for real-time Image Classification","authors":"C. Srinivasan, P. Sridhar, V. Hari Priya, S. Swathi","doi":"10.1109/ICECA55336.2022.10009197","DOIUrl":"https://doi.org/10.1109/ICECA55336.2022.10009197","url":null,"abstract":"Image processing is an important requirement in IoT applications such as robotics, augmented reality, computer vision, Industry 4.0 etc. The capabilities of IoT devices for image processing are limited to sensing the environment, processing and communicating the results. Tiny machine learning (TinyML) is a new paradigm that takes advantage of the IoT to deploy deep learning models to perform complex tasks in resource constrained embedded devices. Image classification is an important task in IoT to interpret images of a particular scene or class. Currently, this task is performed in embedded devices using Binarized Neural Networks (BNNs), which can be converted to a set of weights using a one-hot encoding process. These networks integrated with hardware accelerators can be trained to perform image processing tasks in real-time. This paper proposes a BNN for image classification based on residual learning paradigm, called Tiny-BNN which exploits the skip connections to reduce information loss, and improve the training time and accuracy. Experimental results show that the model achieves a classification accuracy of 90.1 % and 91.6% on the on CIFAR-10 and MNIST datasets respectively.","PeriodicalId":356949,"journal":{"name":"2022 6th International Conference on Electronics, Communication and Aerospace Technology","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128898731","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 : 2022-12-01DOI: 10.1109/ICECA55336.2022.10009292
V. Prasad, P. Venkateswarlu, S. Raju, N. K. Darwante
Predicting and scheduling energy use in Smart Buildings (SB) is essential for implementing Energy-Efficient Management Systems. Managed Smart Grid technology is a critical component for the system's capacity and cost variances to be in real-time. Various methods and models are used to anticipate and schedule energy. This study has analyzed various models before utilizing the machine learning techniques. Here, a combination of ANNs and GANs are used. To test the proposed model, a real-time SB testbed is used. CompactRIO is used here to train and evaluate the proposed model by using the real-time data collected from a PV solar system and S B electrical appliances for ANN implementation. As a blueprint for researchers interested in deploying real-world S B testbeds and investigating machine learning as a possible arena for energy consumption prediction and scheduling, the proposed model has been developed, despite its moderate accuracy and dataset.
{"title":"ANN Modelling based on Machine Learning Approach to Accomplish Energy Source","authors":"V. Prasad, P. Venkateswarlu, S. Raju, N. K. Darwante","doi":"10.1109/ICECA55336.2022.10009292","DOIUrl":"https://doi.org/10.1109/ICECA55336.2022.10009292","url":null,"abstract":"Predicting and scheduling energy use in Smart Buildings (SB) is essential for implementing Energy-Efficient Management Systems. Managed Smart Grid technology is a critical component for the system's capacity and cost variances to be in real-time. Various methods and models are used to anticipate and schedule energy. This study has analyzed various models before utilizing the machine learning techniques. Here, a combination of ANNs and GANs are used. To test the proposed model, a real-time SB testbed is used. CompactRIO is used here to train and evaluate the proposed model by using the real-time data collected from a PV solar system and S B electrical appliances for ANN implementation. As a blueprint for researchers interested in deploying real-world S B testbeds and investigating machine learning as a possible arena for energy consumption prediction and scheduling, the proposed model has been developed, despite its moderate accuracy and dataset.","PeriodicalId":356949,"journal":{"name":"2022 6th International Conference on Electronics, Communication and Aerospace Technology","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132401332","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 : 2022-12-01DOI: 10.1109/ICECA55336.2022.10009157
Martin Victor K, Immanuel Johnraja Jebadurai, Getzi Jeba Leelipushpam Paulraj, Jebaveerasingh Jebadurai
Internet of Things connects objects seamlessly for various applications viz., smart healthcare, industries, farming and many more. In an Internet of Things environment, various standards and protocols have been used to connect applications. Routing protocol for low power and lossy network is one such protocol used to connect devices for data transmission. As these protocols have been used, it is essential to preserve the security and privacy of the users. This paper proposes a secure routing protocol for low power and lossy network using an optimized k means clustering technique. Initially, every node calculates the sequence number variance, route presence ratio and transited routing messages for itself. Then optimized k means clustering technique has been used to cluster the nodes into normal and malicious. The nodes designated as abnormal are eliminated from the network. The proposed technique is simulated and performance analysis is carried out on performance metrics viz., packet delivery ratio, false positive rate and falsenegative rate.
{"title":"Mitigating Sinkhole attack in RPL based Internet of Things Environment using Optimized K means Clustering technique","authors":"Martin Victor K, Immanuel Johnraja Jebadurai, Getzi Jeba Leelipushpam Paulraj, Jebaveerasingh Jebadurai","doi":"10.1109/ICECA55336.2022.10009157","DOIUrl":"https://doi.org/10.1109/ICECA55336.2022.10009157","url":null,"abstract":"Internet of Things connects objects seamlessly for various applications viz., smart healthcare, industries, farming and many more. In an Internet of Things environment, various standards and protocols have been used to connect applications. Routing protocol for low power and lossy network is one such protocol used to connect devices for data transmission. As these protocols have been used, it is essential to preserve the security and privacy of the users. This paper proposes a secure routing protocol for low power and lossy network using an optimized k means clustering technique. Initially, every node calculates the sequence number variance, route presence ratio and transited routing messages for itself. Then optimized k means clustering technique has been used to cluster the nodes into normal and malicious. The nodes designated as abnormal are eliminated from the network. The proposed technique is simulated and performance analysis is carried out on performance metrics viz., packet delivery ratio, false positive rate and falsenegative rate.","PeriodicalId":356949,"journal":{"name":"2022 6th International Conference on Electronics, Communication and Aerospace Technology","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130267800","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 : 2022-12-01DOI: 10.1109/ICECA55336.2022.10009509
R. Elankavi, D. Dinakaran, R. Chetty, M. M. Ramya, J. Jose
Pipelines are crucial in transportation, requiring regular maintenance to function correctly. Robots perform maintenance works, and each robot moves through the pipes using a different mechanism. The In-Pipe Inspection robots are categorized based on their locomotion types. The mechanisms utilized by wheeled wall-pressed type In-Pipe Inspection Robots are divided into dependent and independent mechanisms. The present wheeled IPIR using the dependent mechanism finds difficulty in passing through vertical pipe with obstacles. In this study, one robot from each mechanism is chosen, and the mobility of that robot inside a vertical pipe with an obstacle is examined through simulation using the MSC ADAMS software and observed experimentally. The results show that the robot wheels used in the independent mechanism always make contact with the pipeline's inner surface compared to the dependent mechanism. This result indicates that the robot with the independent mechanism is more suitable for In-Pipe inspection.
{"title":"Optimized Design and Performance Comparison of Wheeled Type In-Pipe Inspection Robot","authors":"R. Elankavi, D. Dinakaran, R. Chetty, M. M. Ramya, J. Jose","doi":"10.1109/ICECA55336.2022.10009509","DOIUrl":"https://doi.org/10.1109/ICECA55336.2022.10009509","url":null,"abstract":"Pipelines are crucial in transportation, requiring regular maintenance to function correctly. Robots perform maintenance works, and each robot moves through the pipes using a different mechanism. The In-Pipe Inspection robots are categorized based on their locomotion types. The mechanisms utilized by wheeled wall-pressed type In-Pipe Inspection Robots are divided into dependent and independent mechanisms. The present wheeled IPIR using the dependent mechanism finds difficulty in passing through vertical pipe with obstacles. In this study, one robot from each mechanism is chosen, and the mobility of that robot inside a vertical pipe with an obstacle is examined through simulation using the MSC ADAMS software and observed experimentally. The results show that the robot wheels used in the independent mechanism always make contact with the pipeline's inner surface compared to the dependent mechanism. This result indicates that the robot with the independent mechanism is more suitable for In-Pipe inspection.","PeriodicalId":356949,"journal":{"name":"2022 6th International Conference on Electronics, Communication and Aerospace Technology","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125316924","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 : 2022-12-01DOI: 10.1109/ICECA55336.2022.10009224
S. Balasubramaniyan, R. Vidyalakshmi, T. Srinivas, K. Sekar, Atul Sarojwal, Satyendra Vishwakarma
Society is more reliant on traditional energy sources, and the percentage of power use is rising every day. Continuation of this trend could lead to the demise of traditional energy sources within a few years. So, now would be the moment to use renewable energies. The most environmentally friendly and long-term renewable energy is the solar energy. It is possible to capture the energy of the sun by employing a solar panel. Many factors influence the rate at which a solar panel generates energy, like the irradiance of sunlight and the temperature of the material. Hence more sunshine the solar panel receives, the more power it generates. Because a sun's location in the sky varies throughout the day, a fixed solar panel can't detect the greatest amount of sunlight throughout the daylight hours. The solar panel must have an automatic tracking mechanism to guarantee that it receives the maximum amount of sunlight. This paper aims to design an autonomous solar tracking system and make the solar panel rotate depending on the sunlight direction. As an advanced controller for the tracking system, the Adaptive Neuro-Fuzzy Inference System (ANFIS) is selected. The solar panel is tilted to the desired angle by this controller. The conventional Proportional Integral Derivative controller is also employed and compared with the ANFIS controller. The controller's performance is evaluated by comparing the controller's time characteristics and errors. In both set point tracking and disturbance rejection, the simulated results suggest that the ANFIS controller is the optimum choice for a solar tracking system.
{"title":"Advanced Controller for Single Axis Solar Tracking System","authors":"S. Balasubramaniyan, R. Vidyalakshmi, T. Srinivas, K. Sekar, Atul Sarojwal, Satyendra Vishwakarma","doi":"10.1109/ICECA55336.2022.10009224","DOIUrl":"https://doi.org/10.1109/ICECA55336.2022.10009224","url":null,"abstract":"Society is more reliant on traditional energy sources, and the percentage of power use is rising every day. Continuation of this trend could lead to the demise of traditional energy sources within a few years. So, now would be the moment to use renewable energies. The most environmentally friendly and long-term renewable energy is the solar energy. It is possible to capture the energy of the sun by employing a solar panel. Many factors influence the rate at which a solar panel generates energy, like the irradiance of sunlight and the temperature of the material. Hence more sunshine the solar panel receives, the more power it generates. Because a sun's location in the sky varies throughout the day, a fixed solar panel can't detect the greatest amount of sunlight throughout the daylight hours. The solar panel must have an automatic tracking mechanism to guarantee that it receives the maximum amount of sunlight. This paper aims to design an autonomous solar tracking system and make the solar panel rotate depending on the sunlight direction. As an advanced controller for the tracking system, the Adaptive Neuro-Fuzzy Inference System (ANFIS) is selected. The solar panel is tilted to the desired angle by this controller. The conventional Proportional Integral Derivative controller is also employed and compared with the ANFIS controller. The controller's performance is evaluated by comparing the controller's time characteristics and errors. In both set point tracking and disturbance rejection, the simulated results suggest that the ANFIS controller is the optimum choice for a solar tracking system.","PeriodicalId":356949,"journal":{"name":"2022 6th International Conference on Electronics, Communication and Aerospace Technology","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126229392","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 : 2022-12-01DOI: 10.1109/ICECA55336.2022.10009296
Priya K V, Dinesh Peter J
Water is the vital component in a human body. Healthy body should be hydrated enough for the proper functioning of various organs. Water has major roles in the physical functioning of a human. Water plays a major role in the metabolic activities in the body. The various nutrients formed in the human body are transferred to different organs through water. Intake of water and water discharge should be controlled to maintain the water balance. Requirements of water in the human body may not meet through the food items or beverages and also not possible to get from metabolic activities. Sometimes the present dehydration level may be life threatening. There should be a proper mechanism to calculate the severity level of dehydration. If the severity of dehydration could be calculated, it is possible to take proper remedies. Dehydration may lead to different chronic diseases like kidney failure, coma, heart related illness, electrolyte abnormalities etc. The intake of plain water is required to maintain the water balance in the human body for better health. It is inevitable to meet the daily water requirements as the deficiency of water in human being may lead to various chronic diseases. Deep learning methods can be used to develop a predictive model for the early diagnosis of chronic diseases with a proper dataset which indudes not only the test results but also the hydration level in human body.
{"title":"Analysis of Hydration Level Estimation Strategies using Deep Learning","authors":"Priya K V, Dinesh Peter J","doi":"10.1109/ICECA55336.2022.10009296","DOIUrl":"https://doi.org/10.1109/ICECA55336.2022.10009296","url":null,"abstract":"Water is the vital component in a human body. Healthy body should be hydrated enough for the proper functioning of various organs. Water has major roles in the physical functioning of a human. Water plays a major role in the metabolic activities in the body. The various nutrients formed in the human body are transferred to different organs through water. Intake of water and water discharge should be controlled to maintain the water balance. Requirements of water in the human body may not meet through the food items or beverages and also not possible to get from metabolic activities. Sometimes the present dehydration level may be life threatening. There should be a proper mechanism to calculate the severity level of dehydration. If the severity of dehydration could be calculated, it is possible to take proper remedies. Dehydration may lead to different chronic diseases like kidney failure, coma, heart related illness, electrolyte abnormalities etc. The intake of plain water is required to maintain the water balance in the human body for better health. It is inevitable to meet the daily water requirements as the deficiency of water in human being may lead to various chronic diseases. Deep learning methods can be used to develop a predictive model for the early diagnosis of chronic diseases with a proper dataset which indudes not only the test results but also the hydration level in human body.","PeriodicalId":356949,"journal":{"name":"2022 6th International Conference on Electronics, Communication and Aerospace Technology","volume":"245 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123019725","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}
Traffic sign classification automatically detects roadside traffic signs, such as speed limit signs, yield signs, etc. Automatically recognizing traffic signs enables the development of “smarter automobiles.” Self-driving automobiles require traffic sign recognition to interpret and comprehend the roadway accurately. Similarly, “driver alert” systems within cars must understand the surrounding roadway to assist and protect drivers. Our automation would assist drivers in detecting and identifying traffic signs without distracting them from the road. With convolution neural networks, the signboards can be accurately classified. The precision can be improved by adding more layers. The GTSRB dataset is utilized here for training and testing; by fine-tuning the parameters, the 43 types of traffic signs are categorized accurately, and the detection speed also increases.
{"title":"SMS: SIGNS MAY SAVE – Traffic Sign Recognition and Detection using CNN","authors":"Praveen Tumuluru, Lakshmi Burra, N. Sunanda, Shaik Sharez Hussain, Dudipalli Madhu, Hasthi Venkat Varma","doi":"10.1109/ICECA55336.2022.10009638","DOIUrl":"https://doi.org/10.1109/ICECA55336.2022.10009638","url":null,"abstract":"Traffic sign classification automatically detects roadside traffic signs, such as speed limit signs, yield signs, etc. Automatically recognizing traffic signs enables the development of “smarter automobiles.” Self-driving automobiles require traffic sign recognition to interpret and comprehend the roadway accurately. Similarly, “driver alert” systems within cars must understand the surrounding roadway to assist and protect drivers. Our automation would assist drivers in detecting and identifying traffic signs without distracting them from the road. With convolution neural networks, the signboards can be accurately classified. The precision can be improved by adding more layers. The GTSRB dataset is utilized here for training and testing; by fine-tuning the parameters, the 43 types of traffic signs are categorized accurately, and the detection speed also increases.","PeriodicalId":356949,"journal":{"name":"2022 6th International Conference on Electronics, Communication and Aerospace Technology","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123033041","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：