Pub Date : 2021-09-24DOI: 10.1109/ICSES52305.2021.9633779
Meet Singh Chauhan, R. Mishra, Manish I. Patel
Human voice is considered one of the most important features and speech helps humans to communicate with each other. Analysis of speech features is carried out to recognize and separate the target speech. Speech signals are continuous and generally contain overlap regions which make conventional methods like signal based matrices inefficient, thus there is a need to develop an advanced and efficient, architecture that can handle speech recognition and speech separation efficiently. This paper provides a brief view of the work carried out for the speech recognition and separation process with the help of deep learning using mel-frequency cepstral coefficients as a parameter. The speech recognition model is implemented using MFCC-DNN based approach and the speech separation model is based on DNN architecture. Various methods were used like MFCC extraction, DNN tuning, etc. to get better performance and higher accuracy than conventional methods like single channel speech separation, HMM etc.
{"title":"Speech Recognition and Separation System using Deep Learning","authors":"Meet Singh Chauhan, R. Mishra, Manish I. Patel","doi":"10.1109/ICSES52305.2021.9633779","DOIUrl":"https://doi.org/10.1109/ICSES52305.2021.9633779","url":null,"abstract":"Human voice is considered one of the most important features and speech helps humans to communicate with each other. Analysis of speech features is carried out to recognize and separate the target speech. Speech signals are continuous and generally contain overlap regions which make conventional methods like signal based matrices inefficient, thus there is a need to develop an advanced and efficient, architecture that can handle speech recognition and speech separation efficiently. This paper provides a brief view of the work carried out for the speech recognition and separation process with the help of deep learning using mel-frequency cepstral coefficients as a parameter. The speech recognition model is implemented using MFCC-DNN based approach and the speech separation model is based on DNN architecture. Various methods were used like MFCC extraction, DNN tuning, etc. to get better performance and higher accuracy than conventional methods like single channel speech separation, HMM etc.","PeriodicalId":6777,"journal":{"name":"2021 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES)","volume":"1 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89506594","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 : 2021-09-24DOI: 10.1109/ICSES52305.2021.9633888
M. Kavitha, D. Immanuel, C. Rex, V. Meenakshi, M. Pushpavalli, Supriya Singari, Vinoba Baskaran
Solar energy is renewable energy source which can be easily converted into electrical energy. This paper presents the selection of proper PV system, battery and inverter for a particular application for any location based on its climatic condition before implementing in real time. So before implementing the experimental set up, the entire system is simulated using any PV*SOL simulation software and the obtained results used to decide and modify the design of planned system. In this paper, a Grid coupled PV system along with electrical battery, electrical vehicle and consumption load is analyzed. Four PV module is used and each PV module uses different tracking systems. Production forecast with consumption, usage of PV energy, coverage of consumption of electric vehicle, battery, grid and other electrical appliances are analyzed by using PV*Sol.
{"title":"Energy Forecasting of Grid Connected Roof Mounted Solar PV Using PV*SOL","authors":"M. Kavitha, D. Immanuel, C. Rex, V. Meenakshi, M. Pushpavalli, Supriya Singari, Vinoba Baskaran","doi":"10.1109/ICSES52305.2021.9633888","DOIUrl":"https://doi.org/10.1109/ICSES52305.2021.9633888","url":null,"abstract":"Solar energy is renewable energy source which can be easily converted into electrical energy. This paper presents the selection of proper PV system, battery and inverter for a particular application for any location based on its climatic condition before implementing in real time. So before implementing the experimental set up, the entire system is simulated using any PV*SOL simulation software and the obtained results used to decide and modify the design of planned system. In this paper, a Grid coupled PV system along with electrical battery, electrical vehicle and consumption load is analyzed. Four PV module is used and each PV module uses different tracking systems. Production forecast with consumption, usage of PV energy, coverage of consumption of electric vehicle, battery, grid and other electrical appliances are analyzed by using PV*Sol.","PeriodicalId":6777,"journal":{"name":"2021 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES)","volume":"75 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91014951","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 : 2021-09-24DOI: 10.1109/ICSES52305.2021.9633911
Nityam Agarwal, Poorvi Seth, Merin Meleet
Techniques that detect sentence similarity have been a very important domain of research and lately many such techniques have been successfully implemented. With the use of Natural Language Processing (NLP) these techniques have been implemented more efficiently. The concept of semantic analysis is very significant in determining sentence similarity. The model proposed in this paper, deploys a NLP based methodology that works on the Sentence Involving Compositional Knowledge (SICK) dataset. The proposed methodology considers the set of sentencesto be a subset of words and it is split based on the semantic and syntactic structure. A lexical database is used by this model, unlike methods deployed by other models. This is followed by the computation of the word order vector. When this NLP based method is tested on the dataset, the accuracy obtained is 82.7% on the basis of mean absolute error. The obtained results are better than the previously used methods. Also, the proposed method is computationally faster than the existing methods.
{"title":"A New Sentence Similarity Computing Technique Using Order and Semantic Similarity","authors":"Nityam Agarwal, Poorvi Seth, Merin Meleet","doi":"10.1109/ICSES52305.2021.9633911","DOIUrl":"https://doi.org/10.1109/ICSES52305.2021.9633911","url":null,"abstract":"Techniques that detect sentence similarity have been a very important domain of research and lately many such techniques have been successfully implemented. With the use of Natural Language Processing (NLP) these techniques have been implemented more efficiently. The concept of semantic analysis is very significant in determining sentence similarity. The model proposed in this paper, deploys a NLP based methodology that works on the Sentence Involving Compositional Knowledge (SICK) dataset. The proposed methodology considers the set of sentencesto be a subset of words and it is split based on the semantic and syntactic structure. A lexical database is used by this model, unlike methods deployed by other models. This is followed by the computation of the word order vector. When this NLP based method is tested on the dataset, the accuracy obtained is 82.7% on the basis of mean absolute error. The obtained results are better than the previously used methods. Also, the proposed method is computationally faster than the existing methods.","PeriodicalId":6777,"journal":{"name":"2021 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES)","volume":"11 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80766709","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 : 2021-09-24DOI: 10.1109/ICSES52305.2021.9633874
S. Aiswarya, K. Ramesh, B. Prabha, S. Sasikumar, K. Vijayakumar
Fog computing is a distributed system that works flawlessly among the cloud and the devices. It enables realtime processing and small latency. It is a distributed decentralized system that is situated between the cloud and computing devices. We are living in the age of the Internet of Things (IoT) or (IoE) Everything that needs immediate processing with minimum latency and wide distribution with location awareness. The characteristics of fog include Mobility, Heterogeneousness, and Wireless Access capability. These factors show a huge part in the development of a real and well-organized IoT platform. As healthcare becomes more patient-centric, it needs a multi-layer architecture to manage the enormoussize of data that is generated by the system. In this paper, we deliberate the importance and applicability of fog and IoT in healthcareby giving a general architecture. In this approach, the system needs a multi-layer architecture that consists of IoT devices, fog, and Cloud computing to manage the complex data with different attributes like its speed, latency, variety, and accuracy.
{"title":"A time optimization model for the Internet of Things-based Healthcare system using Fog computing","authors":"S. Aiswarya, K. Ramesh, B. Prabha, S. Sasikumar, K. Vijayakumar","doi":"10.1109/ICSES52305.2021.9633874","DOIUrl":"https://doi.org/10.1109/ICSES52305.2021.9633874","url":null,"abstract":"Fog computing is a distributed system that works flawlessly among the cloud and the devices. It enables realtime processing and small latency. It is a distributed decentralized system that is situated between the cloud and computing devices. We are living in the age of the Internet of Things (IoT) or (IoE) Everything that needs immediate processing with minimum latency and wide distribution with location awareness. The characteristics of fog include Mobility, Heterogeneousness, and Wireless Access capability. These factors show a huge part in the development of a real and well-organized IoT platform. As healthcare becomes more patient-centric, it needs a multi-layer architecture to manage the enormoussize of data that is generated by the system. In this paper, we deliberate the importance and applicability of fog and IoT in healthcareby giving a general architecture. In this approach, the system needs a multi-layer architecture that consists of IoT devices, fog, and Cloud computing to manage the complex data with different attributes like its speed, latency, variety, and accuracy.","PeriodicalId":6777,"journal":{"name":"2021 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES)","volume":"49 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79777592","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 : 2021-09-24DOI: 10.1109/ICSES52305.2021.9633965
P. Prajwal, D. Prajwal, D. H. Harish, R. Gajanana, B. Jayasri, S. Lokesh
In the Computer Vision domain, there has been continuous growth and development with main focus so as to facilitate a smooth interaction between Machines and human. Perception, planning and control are the main aspects that make up the Self-driving system. Perception subsystem converts the raw data collected by sensors or other information capturing devices into a model of the environment surrounding us. Planning subsystem analyses this model of the surrounding environment and makes certain purposeful decisions based on the inferences obtained from the analysis. Finally, the Control Subsystem is responsible for execution of the actions or the decisions planned previously. The scope of this project is to study and analyze the problems faced in the Perception subsystem in the domain of detecting objects for autonomous cars. Previously, technologies like Radar, LiDAR, GPS and various other sensors had been employed for Driverless cars for mapping the surroundings of the car. However, in the recent past, some deep neural network (DNN) architectures like YOLO (You Only Look Once), and SSD (Single Shot MultiBox Detector) have been developed which are capable of detecting objects even when live video is considered as the input, thus having potential to be included as a part of the Driverless car systems. Selection of a model having considerable accuracy and producing results at a faster rate is very much essential so as to meet the requirements of object detection in driverless cars. In this project, we have used Caffe, which is developed by Berkeley AI Research and Community contributors as the deep learning framework. Keeping in mind the factors that contribute to the selection of a good model, we have chosen SSD model along-side MobileNet Neural network as the base architecture as it results in both faster rate of result production and has a moderate accuracy.
在计算机视觉领域，一直在不断的成长和发展，其主要重点是促进机器与人之间的顺畅交互。感知、规划和控制是构成自动驾驶系统的主要方面。感知子系统将传感器或其他信息捕获设备收集的原始数据转换为我们周围环境的模型。规划子系统对该模型的周边环境进行分析，并根据分析得出的推论做出有针对性的决策。最后，控制子系统负责执行先前计划的操作或决策。本项目的范围是研究和分析自动驾驶汽车物体检测领域感知子系统所面临的问题。此前，无人驾驶汽车采用雷达、激光雷达、GPS和各种其他传感器等技术来绘制汽车周围的地图。然而，在最近的过去，一些深度神经网络(DNN)架构，如YOLO (You Only Look Once)和SSD (Single Shot MultiBox Detector)已经被开发出来，即使将实时视频视为输入，也能够检测到物体，因此有可能被纳入无人驾驶汽车系统的一部分。为了满足无人驾驶汽车中物体检测的要求，选择一个具有相当精度并以更快的速度产生结果的模型是非常必要的。在这个项目中，我们使用了由伯克利人工智能研究和社区贡献者开发的Caffe作为深度学习框架。考虑到有助于选择一个好的模型的因素，我们选择了SSD模型和MobileNet神经网络作为基础架构，因为它既可以更快地产生结果，又具有中等的准确性。
{"title":"Object Detection in Self Driving Cars Using Deep Learning","authors":"P. Prajwal, D. Prajwal, D. H. Harish, R. Gajanana, B. Jayasri, S. Lokesh","doi":"10.1109/ICSES52305.2021.9633965","DOIUrl":"https://doi.org/10.1109/ICSES52305.2021.9633965","url":null,"abstract":"In the Computer Vision domain, there has been continuous growth and development with main focus so as to facilitate a smooth interaction between Machines and human. Perception, planning and control are the main aspects that make up the Self-driving system. Perception subsystem converts the raw data collected by sensors or other information capturing devices into a model of the environment surrounding us. Planning subsystem analyses this model of the surrounding environment and makes certain purposeful decisions based on the inferences obtained from the analysis. Finally, the Control Subsystem is responsible for execution of the actions or the decisions planned previously. The scope of this project is to study and analyze the problems faced in the Perception subsystem in the domain of detecting objects for autonomous cars. Previously, technologies like Radar, LiDAR, GPS and various other sensors had been employed for Driverless cars for mapping the surroundings of the car. However, in the recent past, some deep neural network (DNN) architectures like YOLO (You Only Look Once), and SSD (Single Shot MultiBox Detector) have been developed which are capable of detecting objects even when live video is considered as the input, thus having potential to be included as a part of the Driverless car systems. Selection of a model having considerable accuracy and producing results at a faster rate is very much essential so as to meet the requirements of object detection in driverless cars. In this project, we have used Caffe, which is developed by Berkeley AI Research and Community contributors as the deep learning framework. Keeping in mind the factors that contribute to the selection of a good model, we have chosen SSD model along-side MobileNet Neural network as the base architecture as it results in both faster rate of result production and has a moderate accuracy.","PeriodicalId":6777,"journal":{"name":"2021 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES)","volume":"9 1","pages":"1-7"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85390383","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 : 2021-09-24DOI: 10.1109/ICSES52305.2021.9633836
A. Rameshbabu, G. Sundarrajan, J. B. Paul Glady
The application of multilevel inverter is increased in industries there are different kinds of topology were implemented. The motive of this research work is to reduce the total harmonic distortion by using a reduced number of components. The topology is consisting of H-bridge cascaded with sub multilevel inverter. In this topology four asymmetrical DC sources are been used and eight power electronic switches are used to obtain thirty-one step. The PV (Photo Voltaic) module can be used for the asymmetric DC source. The implemented multi-level inverter topology can generate all voltage levels (positive, negative and zero). The multicarrier sinusoidal pulse width modulation technique is used generate pulse for each switch to obtain a pure sinusoidal waveform as output with low total harmonic distortion. The simulation results are obtained by using MATLAB Simulink. The experimental outputs are also demonstrated in hardware assemble set.
{"title":"Asymmetrical Cascaded H- Bridge 31 Level Inverter with Low THD for PV Application","authors":"A. Rameshbabu, G. Sundarrajan, J. B. Paul Glady","doi":"10.1109/ICSES52305.2021.9633836","DOIUrl":"https://doi.org/10.1109/ICSES52305.2021.9633836","url":null,"abstract":"The application of multilevel inverter is increased in industries there are different kinds of topology were implemented. The motive of this research work is to reduce the total harmonic distortion by using a reduced number of components. The topology is consisting of H-bridge cascaded with sub multilevel inverter. In this topology four asymmetrical DC sources are been used and eight power electronic switches are used to obtain thirty-one step. The PV (Photo Voltaic) module can be used for the asymmetric DC source. The implemented multi-level inverter topology can generate all voltage levels (positive, negative and zero). The multicarrier sinusoidal pulse width modulation technique is used generate pulse for each switch to obtain a pure sinusoidal waveform as output with low total harmonic distortion. The simulation results are obtained by using MATLAB Simulink. The experimental outputs are also demonstrated in hardware assemble set.","PeriodicalId":6777,"journal":{"name":"2021 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES)","volume":"44 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79835581","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 : 2021-09-24DOI: 10.1109/ICSES52305.2021.9633856
M. Prakash, C. Saravanakumar, S. Lakshmi, J. Rose, B. Praba
Artificial intelligence covers a vast area of the real time domain which supports humans for all activities. Machine learning (ML) techniques learn the data and react based on the properties of these data. The properties are identified by extracting the features from the extracted data. Image and video processing methods are essentials in real time application due the IoT (Internet of Things) devices. The data of these types of data is more complex and also high dimensional in nature. These dimensions are reduced by performing reduction techniques before performing the classification process. The proposed ML model targets the traffic management by automating the traffic light based on the flow in the road. The traffic priority is assigned based on the congestion level on the road. The traffic classification is done by considering different features and infrastructure maintained by the city. Existing system suffers the problem due to the following reasons such as traffic congestion, longer waiting time, improper maintenance of the traffic signal, and high carbon emission and so on. The objective of the proposed model is to reduce the traffic congestion by performing traffic flow conditions and make the people comfortable level during the travel.
{"title":"Automatic Feature Extraction and Traffic Management Using Machine Learning and Open CV Model","authors":"M. Prakash, C. Saravanakumar, S. Lakshmi, J. Rose, B. Praba","doi":"10.1109/ICSES52305.2021.9633856","DOIUrl":"https://doi.org/10.1109/ICSES52305.2021.9633856","url":null,"abstract":"Artificial intelligence covers a vast area of the real time domain which supports humans for all activities. Machine learning (ML) techniques learn the data and react based on the properties of these data. The properties are identified by extracting the features from the extracted data. Image and video processing methods are essentials in real time application due the IoT (Internet of Things) devices. The data of these types of data is more complex and also high dimensional in nature. These dimensions are reduced by performing reduction techniques before performing the classification process. The proposed ML model targets the traffic management by automating the traffic light based on the flow in the road. The traffic priority is assigned based on the congestion level on the road. The traffic classification is done by considering different features and infrastructure maintained by the city. Existing system suffers the problem due to the following reasons such as traffic congestion, longer waiting time, improper maintenance of the traffic signal, and high carbon emission and so on. The objective of the proposed model is to reduce the traffic congestion by performing traffic flow conditions and make the people comfortable level during the travel.","PeriodicalId":6777,"journal":{"name":"2021 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES)","volume":"72 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80328761","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 : 2021-09-24DOI: 10.1109/ICSES52305.2021.9633946
K. Vasanth, GPoralla Pradhyumna, Shivani Peram, P. K. Reddy, Tarun Dandetikar, C. Ravi
A smart wearable hand bag for women safety using physical sensor, shock mechanism, along with recordable camera is proposed. The proposed electronics is placed inside the hand bag with the non-lethal electronic shock protruding outside. A shock mechanism will initially help the women in first level of defense followed by the location of the women via SMS to 3 predefined numbers and police control room. A Recordable camera is in place to record the live video images and these images are stored on to a SD card. A Buzzer will act as a indicator to others that a particular person is disturbing. A force sensor is attached at the back of the bag. A violent touch of the bag will start the shock generator circuit. The proposed electronics will work based on the emotional status of women. Hence misuse of the wearable device is prevented.
{"title":"Wearable Device for Commuting Ladies Using IoT","authors":"K. Vasanth, GPoralla Pradhyumna, Shivani Peram, P. K. Reddy, Tarun Dandetikar, C. Ravi","doi":"10.1109/ICSES52305.2021.9633946","DOIUrl":"https://doi.org/10.1109/ICSES52305.2021.9633946","url":null,"abstract":"A smart wearable hand bag for women safety using physical sensor, shock mechanism, along with recordable camera is proposed. The proposed electronics is placed inside the hand bag with the non-lethal electronic shock protruding outside. A shock mechanism will initially help the women in first level of defense followed by the location of the women via SMS to 3 predefined numbers and police control room. A Recordable camera is in place to record the live video images and these images are stored on to a SD card. A Buzzer will act as a indicator to others that a particular person is disturbing. A force sensor is attached at the back of the bag. A violent touch of the bag will start the shock generator circuit. The proposed electronics will work based on the emotional status of women. Hence misuse of the wearable device is prevented.","PeriodicalId":6777,"journal":{"name":"2021 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES)","volume":"45 1","pages":"1-9"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84675332","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 : 2021-09-24DOI: 10.1109/ICSES52305.2021.9633887
Bhargavinath Dornadula, S. Geetha, A. Phamila, R. Priscilla, K. Vijayakumar
Lung related issues are rapidly increasing day by day as it is very important to identify the disease and get treated earliest possible as lungs are part of very complex system, expanding and relaxing thousands of times each day allow us to breathe by bringing oxygen into our bodies and sending carbon dioxide out. Lung related issues are directly preoperational to breathing problems. X-rays are one of the important ways of identifying the status of lungs. As there are many communicable diseases like Covid-19, the person should be identified early and should be treated to control the spread of virus. Lung Opacity is one of the major problem faced by many people and also a very serious problem if not treated early it will spread entire lungs and which leads to cancer similarly Pneumonia is another disease which is an infection to one's lungs caused by spread of virus. All these diseases directly affect Respiratory system of human. The paper aims to lung diseases classification among Pneumonia, Lung opacity, Normal and Covid-19 using the proposed hybrid model. The Deep Transfer Learning model helps to extract good features which helps for better learning and greater results. The Ensembled model of Deep Transfer Learning is used in this paper, which is a combination of VGG, EfficientNet and DenseNet. Considering the output of image augmentation as input for Ensembled model and classification of lung disease. The accuracy of the proposed hybrid model is very much accurate when compared to individual base models.
{"title":"Forecasting the lung diseases from Rediography scans with hybrid Transfer Learning Techniques","authors":"Bhargavinath Dornadula, S. Geetha, A. Phamila, R. Priscilla, K. Vijayakumar","doi":"10.1109/ICSES52305.2021.9633887","DOIUrl":"https://doi.org/10.1109/ICSES52305.2021.9633887","url":null,"abstract":"Lung related issues are rapidly increasing day by day as it is very important to identify the disease and get treated earliest possible as lungs are part of very complex system, expanding and relaxing thousands of times each day allow us to breathe by bringing oxygen into our bodies and sending carbon dioxide out. Lung related issues are directly preoperational to breathing problems. X-rays are one of the important ways of identifying the status of lungs. As there are many communicable diseases like Covid-19, the person should be identified early and should be treated to control the spread of virus. Lung Opacity is one of the major problem faced by many people and also a very serious problem if not treated early it will spread entire lungs and which leads to cancer similarly Pneumonia is another disease which is an infection to one's lungs caused by spread of virus. All these diseases directly affect Respiratory system of human. The paper aims to lung diseases classification among Pneumonia, Lung opacity, Normal and Covid-19 using the proposed hybrid model. The Deep Transfer Learning model helps to extract good features which helps for better learning and greater results. The Ensembled model of Deep Transfer Learning is used in this paper, which is a combination of VGG, EfficientNet and DenseNet. Considering the output of image augmentation as input for Ensembled model and classification of lung disease. The accuracy of the proposed hybrid model is very much accurate when compared to individual base models.","PeriodicalId":6777,"journal":{"name":"2021 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES)","volume":"19 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78501380","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 : 2021-09-24DOI: 10.1109/ICSES52305.2021.9633975
M. Spuritha, Cheruku Sai Kashyap, Tejas Rakesh Nambiar, D. Kiran, N. Rao, G. Reddy
Retailers have been experiencing a drop in their sales due to the rise of E-commerce facilities. This poses a problem where the retail stores need to efficiently manage and price their products to increase their sales. Hence the need for efficient sales prediction and dynamic pricing arises. A forecasting model which can effectively predict the sales of a retail store will help retailers compete in the market. With this intent, the paper proposes a model based on XGBoost whose learners are fitted to the store- product subsets with optimum parameters to increase the overall performance of sales prediction. The proposed model predicted sales for 10 stores with 50 products, with average MAPE, RMSE and R2 values of 11.98 %, 6.63 and 0.76 respectively. In addition, dynamic pricing is applied to the forecasted results which specifies the optimum price of a product based on its demand.
{"title":"Quotidian Sales Forecasting using Machine Learning","authors":"M. Spuritha, Cheruku Sai Kashyap, Tejas Rakesh Nambiar, D. Kiran, N. Rao, G. Reddy","doi":"10.1109/ICSES52305.2021.9633975","DOIUrl":"https://doi.org/10.1109/ICSES52305.2021.9633975","url":null,"abstract":"Retailers have been experiencing a drop in their sales due to the rise of E-commerce facilities. This poses a problem where the retail stores need to efficiently manage and price their products to increase their sales. Hence the need for efficient sales prediction and dynamic pricing arises. A forecasting model which can effectively predict the sales of a retail store will help retailers compete in the market. With this intent, the paper proposes a model based on XGBoost whose learners are fitted to the store- product subsets with optimum parameters to increase the overall performance of sales prediction. The proposed model predicted sales for 10 stores with 50 products, with average MAPE, RMSE and R2 values of 11.98 %, 6.63 and 0.76 respectively. In addition, dynamic pricing is applied to the forecasted results which specifies the optimum price of a product based on its demand.","PeriodicalId":6777,"journal":{"name":"2021 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES)","volume":"2 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2021-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77686833","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：