Pub Date : 2023-02-02DOI: 10.1109/ICAIS56108.2023.10073701
Arpana Prasad, V. Asha, A. P. Nirmala, Madhushree S., Mrinal Kumar, S. Sreeja
This study explores the use of machine learning approaches for addiction prediction. Addiction is a major public health problem, and there is a need for reliable methods of predicting which individuals are at risk for developing substance use disorders. Machine learning has emerged as a powerful tool for predictive modelling, and has been applied successfully to a variety of tasks in the field of medicine. A proposed Machine Learning model for addiction prediction from an ongoing study is presented in this paper.
{"title":"Addictive Disorder Susceptibility Prediction Using Machine Learning Approaches","authors":"Arpana Prasad, V. Asha, A. P. Nirmala, Madhushree S., Mrinal Kumar, S. Sreeja","doi":"10.1109/ICAIS56108.2023.10073701","DOIUrl":"https://doi.org/10.1109/ICAIS56108.2023.10073701","url":null,"abstract":"This study explores the use of machine learning approaches for addiction prediction. Addiction is a major public health problem, and there is a need for reliable methods of predicting which individuals are at risk for developing substance use disorders. Machine learning has emerged as a powerful tool for predictive modelling, and has been applied successfully to a variety of tasks in the field of medicine. A proposed Machine Learning model for addiction prediction from an ongoing study is presented in this paper.","PeriodicalId":164345,"journal":{"name":"2023 Third International Conference on Artificial Intelligence and Smart Energy (ICAIS)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123159029","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 : 2023-02-02DOI: 10.1109/ICAIS56108.2023.10073684
U. Lathamaheswari, J. Jebathangam
The widespread presence of a wide variety of diseases during paddy farming is one of the most significant elements that annually contributes to enormous economic losses. These losses occur as a direct result of the widespread prevalence of these diseases. In this paper, a deep learning algorithm using Deep Belief Network (DBN) and a meta-heuristic optimization using Butterfly optimization algorithm (BOA) is used to classify the images to detect the diseases in a Plant Leaf. The steps of classification involve three different process that includes pre-processing, feature extraction and classification. The simulation is conducted in python to test the efficacy of the classifier. The result of simulation shows that the proposed method has obtained higher classification rate than the existing machine learning classifiers. .
{"title":"A Novel Deep Belief Network with Butterfly Optimization Algorithm for the Classification of Paddy Leaf Disease Detection","authors":"U. Lathamaheswari, J. Jebathangam","doi":"10.1109/ICAIS56108.2023.10073684","DOIUrl":"https://doi.org/10.1109/ICAIS56108.2023.10073684","url":null,"abstract":"The widespread presence of a wide variety of diseases during paddy farming is one of the most significant elements that annually contributes to enormous economic losses. These losses occur as a direct result of the widespread prevalence of these diseases. In this paper, a deep learning algorithm using Deep Belief Network (DBN) and a meta-heuristic optimization using Butterfly optimization algorithm (BOA) is used to classify the images to detect the diseases in a Plant Leaf. The steps of classification involve three different process that includes pre-processing, feature extraction and classification. The simulation is conducted in python to test the efficacy of the classifier. The result of simulation shows that the proposed method has obtained higher classification rate than the existing machine learning classifiers. .","PeriodicalId":164345,"journal":{"name":"2023 Third International Conference on Artificial Intelligence and Smart Energy (ICAIS)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127265088","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 : 2023-02-02DOI: 10.1109/ICAIS56108.2023.10073772
N. Nimagalu, N. M. Reddy, V. H. Reddy, K. Deepa, V. Sailaja
Global warming and air pollution are the two main effects of rising carbon dioxide emissions. Additionally, the quick depletion of the world's petroleum reserves continues to harm both the environment and people. principally, industrial, building, electrical, and transportation are the major polluting sectors leading to ozone layer damage. Using non-renewable sources such as fossil fuels like coal, oil, and natural gases are burned increases CO2 emission and utilization. Greenhouse gases effect, CO2 emissions, and critical changes in climatic conditions increase on a day-to-day basis. For environmental sustainability, the current focus has been on reducing CO2 emissions and mitigating the causes of such emissions. There has not been much attention given to the environmental rebound effect (ERE) approach which concentrates on efficiency enhancements and indicators that go beyond energy to multiple environmental concerns. There are many techniques and methods to overcome carbon dioxide emissions. This paper reviews the causes and remedies (photovoltaic, precooling, usage of renewable sources, and sustainable energy technologies) for mitigating CO2 emissions in various sectors.
{"title":"Smart Grid based Mitigation of Carbon Dioxide Emissions in Various Sectors -A Survey","authors":"N. Nimagalu, N. M. Reddy, V. H. Reddy, K. Deepa, V. Sailaja","doi":"10.1109/ICAIS56108.2023.10073772","DOIUrl":"https://doi.org/10.1109/ICAIS56108.2023.10073772","url":null,"abstract":"Global warming and air pollution are the two main effects of rising carbon dioxide emissions. Additionally, the quick depletion of the world's petroleum reserves continues to harm both the environment and people. principally, industrial, building, electrical, and transportation are the major polluting sectors leading to ozone layer damage. Using non-renewable sources such as fossil fuels like coal, oil, and natural gases are burned increases CO2 emission and utilization. Greenhouse gases effect, CO2 emissions, and critical changes in climatic conditions increase on a day-to-day basis. For environmental sustainability, the current focus has been on reducing CO2 emissions and mitigating the causes of such emissions. There has not been much attention given to the environmental rebound effect (ERE) approach which concentrates on efficiency enhancements and indicators that go beyond energy to multiple environmental concerns. There are many techniques and methods to overcome carbon dioxide emissions. This paper reviews the causes and remedies (photovoltaic, precooling, usage of renewable sources, and sustainable energy technologies) for mitigating CO2 emissions in various sectors.","PeriodicalId":164345,"journal":{"name":"2023 Third International Conference on Artificial Intelligence and Smart Energy (ICAIS)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125378295","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 : 2023-02-02DOI: 10.1109/ICAIS56108.2023.10073814
Saravanan Alagarsamy, Prudhivi Deepak, Lavanya M, T. G. Reddy, M. Kedareswari, A. Senthil Kumar
The Smart Vision Application's premise is that there are numerous rising new technologies that are excelling in their fields. The following are a few of the technologies or models that are now in use: estimation of human pose, steering angle capture, lane detection, and object detection. These are all the various approaches and superb models created with Open Pose and other tools. Since each of these systems has unique characteristics, it is vital to separately construct each one before comprehending how it works. Because there is no trial version available for consumers to use to learn how the model works, these models must be constructed using codes creating a web application that will enable students to learn about and experience how each model functions by using the camera on their device. For business professionals who can use their own models to run, deploy, and test, not simply for users. Every module on the list has some connection to autonomous navigation. These systems have been combined into a single Web application so that students may easily experiment with them and see how they work in real-time. As a result, this platform presents excellent opportunities for students and enthusiastic learners to interact with the live demo and understand how each model functions. It is believed that the Web application will serve as an excellent tool for students to experiment with and gain a feel for the operation of the aforementioned computer vision models.
{"title":"Smart Vision Software Application using Machine Learning","authors":"Saravanan Alagarsamy, Prudhivi Deepak, Lavanya M, T. G. Reddy, M. Kedareswari, A. Senthil Kumar","doi":"10.1109/ICAIS56108.2023.10073814","DOIUrl":"https://doi.org/10.1109/ICAIS56108.2023.10073814","url":null,"abstract":"The Smart Vision Application's premise is that there are numerous rising new technologies that are excelling in their fields. The following are a few of the technologies or models that are now in use: estimation of human pose, steering angle capture, lane detection, and object detection. These are all the various approaches and superb models created with Open Pose and other tools. Since each of these systems has unique characteristics, it is vital to separately construct each one before comprehending how it works. Because there is no trial version available for consumers to use to learn how the model works, these models must be constructed using codes creating a web application that will enable students to learn about and experience how each model functions by using the camera on their device. For business professionals who can use their own models to run, deploy, and test, not simply for users. Every module on the list has some connection to autonomous navigation. These systems have been combined into a single Web application so that students may easily experiment with them and see how they work in real-time. As a result, this platform presents excellent opportunities for students and enthusiastic learners to interact with the live demo and understand how each model functions. It is believed that the Web application will serve as an excellent tool for students to experiment with and gain a feel for the operation of the aforementioned computer vision models.","PeriodicalId":164345,"journal":{"name":"2023 Third International Conference on Artificial Intelligence and Smart Energy (ICAIS)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125405415","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 : 2023-02-02DOI: 10.1109/ICAIS56108.2023.10073920
S. Kumaran, Arunachalam S, Surendar V, S. T
In the present technologically advanced society, new innovations are being developed constantly. An ever-expanding network that constantly works to exchange and acquire information on the newest trends in the industrial IoT Platform is the source of these new developments in IoT communications-based projects. Home automation is a way to remotely or automatically control household equipment from the tip of your finger. The user will then be able to afford solutions, improve energy conservation, and make the best use of energy. The development of IoT home automation has been greatly aided by the detection of fire in this next technology. A sudden destructive event like fire has the potential to quickly spread, resulting in significant losses of both societal goods and human lives. Preventive actions are essential necessary since, in the event of a fire, prevention is always preferable to cure. This necessitates the need to develop a fire safety equipment in both home and workplace. More attention has been given to an IoT-based automatic smoke detection system to detect smoke in a room and even keep track of it. Additionally, it enables us to notify users and the Fire and Rescue Department when a gas sensor detects a particular amount of smoke. These smoke detectors can emit an audible and visual signal locally in a home smoke detector or smoke alarm, or they send a signal to a fire alarm control panel as part of a building's central fire alarm system. Utilizing an automatic smoke detection system. The Internet of Things (IoT) is used in this automatic smoke detection system to operate all the devices, and a Wi-Fi shield serves as a bridge to connect the devices to the network so that the data from the smoke sensor can be read. The smoke situation in a home that the user can access via the Favoriot platform is continuously monitored by this system.
{"title":"IoT based Smoke Detection with Air Temperature and Air Humidity; High Accuracy with Machine Learning","authors":"S. Kumaran, Arunachalam S, Surendar V, S. T","doi":"10.1109/ICAIS56108.2023.10073920","DOIUrl":"https://doi.org/10.1109/ICAIS56108.2023.10073920","url":null,"abstract":"In the present technologically advanced society, new innovations are being developed constantly. An ever-expanding network that constantly works to exchange and acquire information on the newest trends in the industrial IoT Platform is the source of these new developments in IoT communications-based projects. Home automation is a way to remotely or automatically control household equipment from the tip of your finger. The user will then be able to afford solutions, improve energy conservation, and make the best use of energy. The development of IoT home automation has been greatly aided by the detection of fire in this next technology. A sudden destructive event like fire has the potential to quickly spread, resulting in significant losses of both societal goods and human lives. Preventive actions are essential necessary since, in the event of a fire, prevention is always preferable to cure. This necessitates the need to develop a fire safety equipment in both home and workplace. More attention has been given to an IoT-based automatic smoke detection system to detect smoke in a room and even keep track of it. Additionally, it enables us to notify users and the Fire and Rescue Department when a gas sensor detects a particular amount of smoke. These smoke detectors can emit an audible and visual signal locally in a home smoke detector or smoke alarm, or they send a signal to a fire alarm control panel as part of a building's central fire alarm system. Utilizing an automatic smoke detection system. The Internet of Things (IoT) is used in this automatic smoke detection system to operate all the devices, and a Wi-Fi shield serves as a bridge to connect the devices to the network so that the data from the smoke sensor can be read. The smoke situation in a home that the user can access via the Favoriot platform is continuously monitored by this system.","PeriodicalId":164345,"journal":{"name":"2023 Third International Conference on Artificial Intelligence and Smart Energy (ICAIS)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125419528","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 : 2023-02-02DOI: 10.1109/ICAIS56108.2023.10073769
Manasi Swain, A. R. Manyatha, Amulya S Dinesh, Gambhire Swati Sampatrao, Mihir Soni
Career, school, work, new adventures of life, those are often given priority. Eating healthy becomes the next important concern, so going to a fast-food joint and preparing instant food may solve the problems related to food at the moment, but eventually it deteriorates health, either through weight fluctuation, energy loss, or both. To help overcome this, our proposed model aims to create a recipe recommendation system based on ingredient recognition. It helps explore new recipes in the kitchen for beginners, busy parents, foodies, and pro chefs alike. Our system helps users decide what they can cook with the available resources by making use of images of ingredients. YOLOv5 has been employed to detect ingredients. This enables multiple object detection in real-time. An API call is done to calculate the calorie based on the amount of each ingredient. Recipe retrieval is done considering the ingredients detected, calorie count, various cuisines, and diet types. Users now have an idea of what they can cook, according to the recipes retrieved along with the nutritional value of each recipe. Based on the chosen recipe, similar recipes will be recommended by content-based recommendation system using K-Means Clustering. It helps improve user experience by saving time and energy in finding recipes for daily routines. By retrieving the appropriate recipes based on the items that are accessible and providing precise recipe suggestions, this system simplifies people’s life.
{"title":"Ingredients to Recipe: A YOLO-based Object Detector and Recommendation System via Clustering Approach","authors":"Manasi Swain, A. R. Manyatha, Amulya S Dinesh, Gambhire Swati Sampatrao, Mihir Soni","doi":"10.1109/ICAIS56108.2023.10073769","DOIUrl":"https://doi.org/10.1109/ICAIS56108.2023.10073769","url":null,"abstract":"Career, school, work, new adventures of life, those are often given priority. Eating healthy becomes the next important concern, so going to a fast-food joint and preparing instant food may solve the problems related to food at the moment, but eventually it deteriorates health, either through weight fluctuation, energy loss, or both. To help overcome this, our proposed model aims to create a recipe recommendation system based on ingredient recognition. It helps explore new recipes in the kitchen for beginners, busy parents, foodies, and pro chefs alike. Our system helps users decide what they can cook with the available resources by making use of images of ingredients. YOLOv5 has been employed to detect ingredients. This enables multiple object detection in real-time. An API call is done to calculate the calorie based on the amount of each ingredient. Recipe retrieval is done considering the ingredients detected, calorie count, various cuisines, and diet types. Users now have an idea of what they can cook, according to the recipes retrieved along with the nutritional value of each recipe. Based on the chosen recipe, similar recipes will be recommended by content-based recommendation system using K-Means Clustering. It helps improve user experience by saving time and energy in finding recipes for daily routines. By retrieving the appropriate recipes based on the items that are accessible and providing precise recipe suggestions, this system simplifies people’s life.","PeriodicalId":164345,"journal":{"name":"2023 Third International Conference on Artificial Intelligence and Smart Energy (ICAIS)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125559416","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 : 2023-02-02DOI: 10.1109/ICAIS56108.2023.10073873
M. Varadharajan, S. Balaji, V. Ezhilarasan, A. Gowthaman
Consistently, normal and human-instigated catastrophes result in infrastructural hurt, monetary costs, emergencies, wounds, and passings. Worldwide environmental change conjointly fortifies the harming force of catastrophic events. during this unique circumstance, net of Things (IoT) based generally calamity discovery and reaction frameworks are wanted to deal with debacles and crises by up catastrophe location. Thusly, IoT gadgets are acclimated to gather data and working with to recognize contrasting sorts of normal and synthetic debacles. This study presents an overall framework with an assortment of sensors sight strange things. The significant qualification between this strategy and existing frameworks is the decentralized and customized cautioning framework. Here, the general information from the disaster recognized space can be obtained and with that information the people present in that space will be monitored and a caution notification regarding the calamity before evidence gets critical. This will be used as an early warning system in the event of the most unexpected events.
{"title":"Internet of Things based Natural Disaster Detection and Personalized Notification System","authors":"M. Varadharajan, S. Balaji, V. Ezhilarasan, A. Gowthaman","doi":"10.1109/ICAIS56108.2023.10073873","DOIUrl":"https://doi.org/10.1109/ICAIS56108.2023.10073873","url":null,"abstract":"Consistently, normal and human-instigated catastrophes result in infrastructural hurt, monetary costs, emergencies, wounds, and passings. Worldwide environmental change conjointly fortifies the harming force of catastrophic events. during this unique circumstance, net of Things (IoT) based generally calamity discovery and reaction frameworks are wanted to deal with debacles and crises by up catastrophe location. Thusly, IoT gadgets are acclimated to gather data and working with to recognize contrasting sorts of normal and synthetic debacles. This study presents an overall framework with an assortment of sensors sight strange things. The significant qualification between this strategy and existing frameworks is the decentralized and customized cautioning framework. Here, the general information from the disaster recognized space can be obtained and with that information the people present in that space will be monitored and a caution notification regarding the calamity before evidence gets critical. This will be used as an early warning system in the event of the most unexpected events.","PeriodicalId":164345,"journal":{"name":"2023 Third International Conference on Artificial Intelligence and Smart Energy (ICAIS)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114939362","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 : 2023-02-02DOI: 10.1109/ICAIS56108.2023.10073881
Akansha Mittal, Anurag Goel
A community is referred to as a set of nodes in a network that has a high degree of connectivity with each other and a low degree of connectivity with other nodes in the same network. Community Detection is a renowned research problem for the past many years. The applications of Community Detection is spread across several domains like social networks, transportation networks, genetic networks, citation networks, web networks etc. In this work, several unsupervised learning techniques namely Louvain Algorithm, K-means clustering Algorithm and Gaussian Mixture Model have been examined to identify communities in social networks. The results demonstrated that the Louvain Algorithm outperforms the other two unsupervised learning techniques.
{"title":"Community Detection using Unsupervised Learning Approach","authors":"Akansha Mittal, Anurag Goel","doi":"10.1109/ICAIS56108.2023.10073881","DOIUrl":"https://doi.org/10.1109/ICAIS56108.2023.10073881","url":null,"abstract":"A community is referred to as a set of nodes in a network that has a high degree of connectivity with each other and a low degree of connectivity with other nodes in the same network. Community Detection is a renowned research problem for the past many years. The applications of Community Detection is spread across several domains like social networks, transportation networks, genetic networks, citation networks, web networks etc. In this work, several unsupervised learning techniques namely Louvain Algorithm, K-means clustering Algorithm and Gaussian Mixture Model have been examined to identify communities in social networks. The results demonstrated that the Louvain Algorithm outperforms the other two unsupervised learning techniques.","PeriodicalId":164345,"journal":{"name":"2023 Third International Conference on Artificial Intelligence and Smart Energy (ICAIS)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115083397","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 : 2023-02-02DOI: 10.1109/ICAIS56108.2023.10073838
M. Indirani, Cuddapah Anitha, Sohan Goswami, K. Baranitharan, S. Govindaraju, M. R.
Salient detection is an active and critical area that is designed within the detection of items of a video recording, nonetheless, it attracts elevated interest among scientists. With rising powerful video clip information, the overall performance of saliency item detection techniques is degrading with typical item detection techniques. The problems lie with blurry moving goals, super-fast motion of items as well as dynamic background or background occlusion alteration on foreground areas within the video clip frames. This kind of obstacle leads to bad saliency detection. This paper models a full mastering design to deal with the difficulties, and that works on an advanced framework by merging the thought of Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) with firefly Optimization technique for video clip saliency detection. Good utilization of the firefly algorithm together with CRNN is completed for the removal of characteristics by the video clips for item recognition. The primary objective of this newspaper is to present an effective hyperparameter choice framework for Convolution Recurrent Neural Networks (CRNNs) that employ one of the more popular swarm intelligence methods, the firefly algorithm. The suggested technique goals at creating a spatiotemporal design that exploits temporal, local, and spatial restriction cues to attain worldwide SEO. The process of locating the salient items in deep benchmark powerful video recording datasets will be completed by recording the temporal, local, and spatial restriction characteristics with all the CRNN. The CRNN is examined on benchmark datasets from typical video clip salient item detection techniques within the terminology of accuracy and load of Computation. The tests show that the proposed design accomplishes enhanced overall performance compared to some other existing versions which prove to significantly satisfy all the traditional object detection models.
{"title":"Detection of Salient Objects in a Video using a Hybrid Neural Network Model","authors":"M. Indirani, Cuddapah Anitha, Sohan Goswami, K. Baranitharan, S. Govindaraju, M. R.","doi":"10.1109/ICAIS56108.2023.10073838","DOIUrl":"https://doi.org/10.1109/ICAIS56108.2023.10073838","url":null,"abstract":"Salient detection is an active and critical area that is designed within the detection of items of a video recording, nonetheless, it attracts elevated interest among scientists. With rising powerful video clip information, the overall performance of saliency item detection techniques is degrading with typical item detection techniques. The problems lie with blurry moving goals, super-fast motion of items as well as dynamic background or background occlusion alteration on foreground areas within the video clip frames. This kind of obstacle leads to bad saliency detection. This paper models a full mastering design to deal with the difficulties, and that works on an advanced framework by merging the thought of Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) with firefly Optimization technique for video clip saliency detection. Good utilization of the firefly algorithm together with CRNN is completed for the removal of characteristics by the video clips for item recognition. The primary objective of this newspaper is to present an effective hyperparameter choice framework for Convolution Recurrent Neural Networks (CRNNs) that employ one of the more popular swarm intelligence methods, the firefly algorithm. The suggested technique goals at creating a spatiotemporal design that exploits temporal, local, and spatial restriction cues to attain worldwide SEO. The process of locating the salient items in deep benchmark powerful video recording datasets will be completed by recording the temporal, local, and spatial restriction characteristics with all the CRNN. The CRNN is examined on benchmark datasets from typical video clip salient item detection techniques within the terminology of accuracy and load of Computation. The tests show that the proposed design accomplishes enhanced overall performance compared to some other existing versions which prove to significantly satisfy all the traditional object detection models.","PeriodicalId":164345,"journal":{"name":"2023 Third International Conference on Artificial Intelligence and Smart Energy (ICAIS)","volume":"466 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115323519","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}
Electric vehicles (EVs) are becoming more popular due to their many desirable characteristics, such as their ability to store energy in batteries and their small carbon impact. Electric vehicles represent a revolution in both the transportation and electrical sectors, and by uniting the two, they have the ability to improve both. This relationship needs the implementation of effective Power Factor Correction (PFC) systems for charging EV batteries, which minimises the supply front-inherent end's Power Quality (PQ) concerns. This study uses a Bridgeless Landsman converter for PFC, since it is efficient and can detect changes in the link voltage. The usage of an ANN-based PI controller facilitates prediction and classification with regards to reaction time. This is accomplished by connecting the hysteresis controller to a PWM generator, which then determines the correct switching frequency for the converter in steady state. The suggested strategy aids in effective minimising of harmonics with heightened efficiency.
{"title":"ANN based Bridgeless Landsman Converter Design for Electric Vehicle Power Factor Correction","authors":"Suresh Vendoti, Rangala Manikanta Swamy, Tibirisetti Sai Saran Jyothi, Bochu Varun","doi":"10.1109/ICAIS56108.2023.10073855","DOIUrl":"https://doi.org/10.1109/ICAIS56108.2023.10073855","url":null,"abstract":"Electric vehicles (EVs) are becoming more popular due to their many desirable characteristics, such as their ability to store energy in batteries and their small carbon impact. Electric vehicles represent a revolution in both the transportation and electrical sectors, and by uniting the two, they have the ability to improve both. This relationship needs the implementation of effective Power Factor Correction (PFC) systems for charging EV batteries, which minimises the supply front-inherent end's Power Quality (PQ) concerns. This study uses a Bridgeless Landsman converter for PFC, since it is efficient and can detect changes in the link voltage. The usage of an ANN-based PI controller facilitates prediction and classification with regards to reaction time. This is accomplished by connecting the hysteresis controller to a PWM generator, which then determines the correct switching frequency for the converter in steady state. The suggested strategy aids in effective minimising of harmonics with heightened efficiency.","PeriodicalId":164345,"journal":{"name":"2023 Third International Conference on Artificial Intelligence and Smart Energy (ICAIS)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122060395","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：