Pub Date : 2020-11-26DOI: 10.1109/ICRCICN50933.2020.9296156
A. Agrawal, Kadamb Agarwal, J. Choudhary, Aradhita Bhattacharya, Srihitha Tangudu, Nishkarsh Makhija, R. B
The rate of road accidents has increased to a large extent over the last few years. This has eventually resulted in a huge loss of lives and property. Hence, the need of the hour has aroused to detect such accident spots as quickly as possible so that proper life-saving measures can be taken and the mishap prone areas can be put on alert. In order to address this problem, we propose a Machine Learning and Deep Learning based model on the concepts of Clustering and Classification that can be used to detect accidents from the traffic surveillance cameras. Firstly all the videos are split up into smaller shots according to scene changes. And then key frames are extracted from each shot based on histogram difference of consecutive frames. Then distance between the vehicles is determined to detect the potential accident. The obtained key frames are passed through a ResNet50 architecture for feature extraction. After obtaining the feature vectors of all videos, K-Means clustering has been applied to obtain Bag of Visual Words(BOVW). Finally, these bag of visual words is sent as input to a Support Vector Machine(SVM) classifier that outputs if a video contained an accident or not. The proposed method has an accuracy of 94.14%.
在过去的几年里，交通事故的发生率在很大程度上增加了。这最终造成了巨大的生命和财产损失。因此，迫切需要尽快发现这些事故地点，以便采取适当的救生措施，并提高事故易发地区的警戒水平。为了解决这个问题，我们提出了一个基于机器学习和深度学习的模型，该模型基于聚类和分类的概念，可用于从交通监控摄像机中检测事故。首先，所有的视频根据场景的变化被分割成更小的镜头。然后根据连续帧的直方图差异从每个镜头中提取关键帧。然后确定车辆之间的距离，以检测潜在的事故。获得的关键帧通过ResNet50架构进行特征提取。在获得所有视频的特征向量后，应用K-Means聚类得到视觉词包(Bag of Visual Words, BOVW)。最后，这些视觉词包作为输入发送给支持向量机(SVM)分类器，该分类器输出视频是否包含事故。该方法的准确率为94.14%。
{"title":"Automatic Traffic Accident Detection System Using ResNet and SVM","authors":"A. Agrawal, Kadamb Agarwal, J. Choudhary, Aradhita Bhattacharya, Srihitha Tangudu, Nishkarsh Makhija, R. B","doi":"10.1109/ICRCICN50933.2020.9296156","DOIUrl":"https://doi.org/10.1109/ICRCICN50933.2020.9296156","url":null,"abstract":"The rate of road accidents has increased to a large extent over the last few years. This has eventually resulted in a huge loss of lives and property. Hence, the need of the hour has aroused to detect such accident spots as quickly as possible so that proper life-saving measures can be taken and the mishap prone areas can be put on alert. In order to address this problem, we propose a Machine Learning and Deep Learning based model on the concepts of Clustering and Classification that can be used to detect accidents from the traffic surveillance cameras. Firstly all the videos are split up into smaller shots according to scene changes. And then key frames are extracted from each shot based on histogram difference of consecutive frames. Then distance between the vehicles is determined to detect the potential accident. The obtained key frames are passed through a ResNet50 architecture for feature extraction. After obtaining the feature vectors of all videos, K-Means clustering has been applied to obtain Bag of Visual Words(BOVW). Finally, these bag of visual words is sent as input to a Support Vector Machine(SVM) classifier that outputs if a video contained an accident or not. The proposed method has an accuracy of 94.14%.","PeriodicalId":138966,"journal":{"name":"2020 Fifth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114162578","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 : 2020-11-26DOI: 10.1109/ICRCICN50933.2020.9295964
Subhadeep Koley, Hiranmoy Roy, S. Dhar, D. Bhattacharjee
The advent of computer-vision based systems has given rise to the need for efficient edge detection algorithms. This paper presents a novel approach called the Local-Friis-Radiation-Magnitude-Ratio (LFRMR) for edge detection. LFRMR incorporates the renowned Friis Equation of antenna radiation and extends it to the grid of image pixels to establish a relation among the pixels residing in a local neighbourhood, to extract accurate illumination-invariant and noise resistant edge maps. Quantitative and qualitative experimental results on BSDS500 dataset depicts that the proposed scheme can extract true edges with utmost precision and recall. Furthermore, the proposed scheme is quite robust against Gaussian channel noise and Salt & Pepper noise. A detailed mathematical investigation has also been carried out to prove that the proposed framework is illumination-invariant and robust in noisy environments. Optimum algorithm parameters are decided via experimental analysis. A comparison with the latest state-of-the-art methods is also presented.
{"title":"Edge Detection based on Local-Friis-Radiation-Magnitude-Ratio (LFRMR)","authors":"Subhadeep Koley, Hiranmoy Roy, S. Dhar, D. Bhattacharjee","doi":"10.1109/ICRCICN50933.2020.9295964","DOIUrl":"https://doi.org/10.1109/ICRCICN50933.2020.9295964","url":null,"abstract":"The advent of computer-vision based systems has given rise to the need for efficient edge detection algorithms. This paper presents a novel approach called the Local-Friis-Radiation-Magnitude-Ratio (LFRMR) for edge detection. LFRMR incorporates the renowned Friis Equation of antenna radiation and extends it to the grid of image pixels to establish a relation among the pixels residing in a local neighbourhood, to extract accurate illumination-invariant and noise resistant edge maps. Quantitative and qualitative experimental results on BSDS500 dataset depicts that the proposed scheme can extract true edges with utmost precision and recall. Furthermore, the proposed scheme is quite robust against Gaussian channel noise and Salt & Pepper noise. A detailed mathematical investigation has also been carried out to prove that the proposed framework is illumination-invariant and robust in noisy environments. Optimum algorithm parameters are decided via experimental analysis. A comparison with the latest state-of-the-art methods is also presented.","PeriodicalId":138966,"journal":{"name":"2020 Fifth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125497726","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 : 2020-11-26DOI: 10.1109/ICRCICN50933.2020.9296159
Pramod Sunagar, A. Kanavalli, S. Shweta
In this digital world, social media has become a communication platform for the entire world. It allows the users to express their views and opinions on various platforms. During this process, both structured and unstructured data is collected in a random manner. The exclusion of categorization causes the user to have difficulty in understanding or accessing information relating to those categories that they like. In the field of social network analysis, the automation procedure for inferring special interests from users is a challenging task. The solution for this is classification of text which inherently classifies with natural language against certain categories on text. Feature Expansion is one of the main aspects of designing an effective machine learning model for classifying texts. This technique has more relevance when unstructured data is in question. In this paper, a comparison study of various methods used for text classification is presented. The methods are broadly categorized into two major types. One is without feature expansion and the other with Hypernym-Hyponym based feature expansions. Different machine learning algorithms under both the categories are mentioned. The datasets, algorithms, results of evaluation of various algorithms are surveyed and tabulated.
{"title":"A Survey Report on Hypernym Techniques for Text Classification","authors":"Pramod Sunagar, A. Kanavalli, S. Shweta","doi":"10.1109/ICRCICN50933.2020.9296159","DOIUrl":"https://doi.org/10.1109/ICRCICN50933.2020.9296159","url":null,"abstract":"In this digital world, social media has become a communication platform for the entire world. It allows the users to express their views and opinions on various platforms. During this process, both structured and unstructured data is collected in a random manner. The exclusion of categorization causes the user to have difficulty in understanding or accessing information relating to those categories that they like. In the field of social network analysis, the automation procedure for inferring special interests from users is a challenging task. The solution for this is classification of text which inherently classifies with natural language against certain categories on text. Feature Expansion is one of the main aspects of designing an effective machine learning model for classifying texts. This technique has more relevance when unstructured data is in question. In this paper, a comparison study of various methods used for text classification is presented. The methods are broadly categorized into two major types. One is without feature expansion and the other with Hypernym-Hyponym based feature expansions. Different machine learning algorithms under both the categories are mentioned. The datasets, algorithms, results of evaluation of various algorithms are surveyed and tabulated.","PeriodicalId":138966,"journal":{"name":"2020 Fifth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129224521","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 : 2020-11-26DOI: 10.1109/icrcicn50933.2020.9296158
{"title":"ICRCICN 2020 TOC","authors":"","doi":"10.1109/icrcicn50933.2020.9296158","DOIUrl":"https://doi.org/10.1109/icrcicn50933.2020.9296158","url":null,"abstract":"","PeriodicalId":138966,"journal":{"name":"2020 Fifth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126339371","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 : 2020-11-26DOI: 10.1109/ICRCICN50933.2020.9296187
E. Ramanujam, T. Chandrakumar, K.T. Thivyadharsine, D. Varsha
More than 77 million people in India are influenced by diabetes mellitus and a significant number of them are under risk with specific complications, for instance cardiovascular failure, stroke, nerve infection, etc., The prevalence ratio of diabetes is high in urban areas due to the migration of rural people and industrialization. While considering diabetes in prosperous urban, it has become a grave anxiety among rural people also. Early diagnosis and proper therapeutic management may reduce the expenditure and mortality rate. however, the cost of early diagnosis and laboratory testing is very high. To provide a user-friendly and cost-effective system, this paper proposes a multilingual decision support system by integrating the best predictive model (among various machine learning algorithms) and clinical decision support system. The proposed system provides a user interface to assess diabetes by themselves or with a nursing assistant available in primary health centre.
{"title":"A Multilingual Decision Support System for early detection of Diabetes using Machine Learning approach: Case study for Rural Indian people","authors":"E. Ramanujam, T. Chandrakumar, K.T. Thivyadharsine, D. Varsha","doi":"10.1109/ICRCICN50933.2020.9296187","DOIUrl":"https://doi.org/10.1109/ICRCICN50933.2020.9296187","url":null,"abstract":"More than 77 million people in India are influenced by diabetes mellitus and a significant number of them are under risk with specific complications, for instance cardiovascular failure, stroke, nerve infection, etc., The prevalence ratio of diabetes is high in urban areas due to the migration of rural people and industrialization. While considering diabetes in prosperous urban, it has become a grave anxiety among rural people also. Early diagnosis and proper therapeutic management may reduce the expenditure and mortality rate. however, the cost of early diagnosis and laboratory testing is very high. To provide a user-friendly and cost-effective system, this paper proposes a multilingual decision support system by integrating the best predictive model (among various machine learning algorithms) and clinical decision support system. The proposed system provides a user interface to assess diabetes by themselves or with a nursing assistant available in primary health centre.","PeriodicalId":138966,"journal":{"name":"2020 Fifth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115049734","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 : 2020-11-26DOI: 10.1109/ICRCICN50933.2020.9296192
K. Bhargav, S. Ambika, S. Deepak, S. Sudha
We propose a method to reconstruct natural grayscale images and handwritten characters from Functional Magnetic Resonance Imaging (fMRI) data and achieve a high degree of similarity to the original stimuli images. The approach utilizes a pre-trained Deep Convolutional Generative Adversarial Network (DCGAN) to reconstruct images and provide visual confirmations regarding the resemblance between the reconstructed and original images. A linear regressor is used to elicit information from the fMRI data and estimate a latent space representation for the formerly trained generative model. A composite loss function combining the Perceptual and Multi-Scale Structural Similarity Index (MS-SSIM) losses is used to train the regressor. The advantages of both functions are evident with the Perceptual loss capturing semantic information and the MS-SSIM loss carrying information about objects in a scene. With this loss function, we were able to reconstruct human objects in the stimuli to a degree of accuracy. The reconstructions obtained were then validated using the Scale Invariant Feature Transform (SIFT) method to elucidate the number of features matched between the original and recreated images. The SSIM scores for the reconstructed images are observed to be higher than state-of-the-art methods. Parallels are drawn between the distortions produced in images submerged underwater and those in the reconstructed images using the Contrast Limited Adaptive Histogram Equalization (CLAHE), an image enhancement technique. A sharp increase in the number of SIFT features matched, is observed with the application of CLAHE on the reconstructed images.
{"title":"Imagenation - A DCGAN based method for Image Reconstruction from fMRI","authors":"K. Bhargav, S. Ambika, S. Deepak, S. Sudha","doi":"10.1109/ICRCICN50933.2020.9296192","DOIUrl":"https://doi.org/10.1109/ICRCICN50933.2020.9296192","url":null,"abstract":"We propose a method to reconstruct natural grayscale images and handwritten characters from Functional Magnetic Resonance Imaging (fMRI) data and achieve a high degree of similarity to the original stimuli images. The approach utilizes a pre-trained Deep Convolutional Generative Adversarial Network (DCGAN) to reconstruct images and provide visual confirmations regarding the resemblance between the reconstructed and original images. A linear regressor is used to elicit information from the fMRI data and estimate a latent space representation for the formerly trained generative model. A composite loss function combining the Perceptual and Multi-Scale Structural Similarity Index (MS-SSIM) losses is used to train the regressor. The advantages of both functions are evident with the Perceptual loss capturing semantic information and the MS-SSIM loss carrying information about objects in a scene. With this loss function, we were able to reconstruct human objects in the stimuli to a degree of accuracy. The reconstructions obtained were then validated using the Scale Invariant Feature Transform (SIFT) method to elucidate the number of features matched between the original and recreated images. The SSIM scores for the reconstructed images are observed to be higher than state-of-the-art methods. Parallels are drawn between the distortions produced in images submerged underwater and those in the reconstructed images using the Contrast Limited Adaptive Histogram Equalization (CLAHE), an image enhancement technique. A sharp increase in the number of SIFT features matched, is observed with the application of CLAHE on the reconstructed images.","PeriodicalId":138966,"journal":{"name":"2020 Fifth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126433318","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 : 2020-11-26DOI: 10.1109/ICRCICN50933.2020.9296161
S. Manavi, Vinuthna Nekkanti, Ram Shankar Choudhary, N. Jayapandian
The Internet of Things (IoT) has been gaining attention in various disciplines ranging from agriculture, health, industries and home automation. When a pandemic first breaks out early detection, isolating the infected, and tracing the contacts are the most important challenges. IoT protocols like Radio-frequency identification (RFID), Wireless Fidelity (WiFi), Global Positioning System (GPS) are gaining popularity for providing solutions to these challenges. IoT based applications in the health sector are benefitting COVID-19 (coronavirus disease of 2019) patients during this pandemic situation. This article explores and reviews the various Internet of Things enabled technologies and applications used in screening, contact tracing, and surveillance. IoT based telemedicine processes are very useful during the pandemic COVID-19. The purpose of this paper is to deliver an overall understanding of the existing and proposed technologies of IoT based solutions to make the situations better during COVID-19.
{"title":"Review on Emerging Internet of Things Technologies to Fight the COVID-19","authors":"S. Manavi, Vinuthna Nekkanti, Ram Shankar Choudhary, N. Jayapandian","doi":"10.1109/ICRCICN50933.2020.9296161","DOIUrl":"https://doi.org/10.1109/ICRCICN50933.2020.9296161","url":null,"abstract":"The Internet of Things (IoT) has been gaining attention in various disciplines ranging from agriculture, health, industries and home automation. When a pandemic first breaks out early detection, isolating the infected, and tracing the contacts are the most important challenges. IoT protocols like Radio-frequency identification (RFID), Wireless Fidelity (WiFi), Global Positioning System (GPS) are gaining popularity for providing solutions to these challenges. IoT based applications in the health sector are benefitting COVID-19 (coronavirus disease of 2019) patients during this pandemic situation. This article explores and reviews the various Internet of Things enabled technologies and applications used in screening, contact tracing, and surveillance. IoT based telemedicine processes are very useful during the pandemic COVID-19. The purpose of this paper is to deliver an overall understanding of the existing and proposed technologies of IoT based solutions to make the situations better during COVID-19.","PeriodicalId":138966,"journal":{"name":"2020 Fifth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128228030","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 : 2020-11-26DOI: 10.1109/ICRCICN50933.2020.9296180
Dweepa Honnavalli, Kavya Varma, G. Srinivasa
In recent times, the need for virtual screening of chemical compounds has grown with the advent of computational synthesis and de-novo generation of drugs. The state-of-the-art benchmarks of virtual screening today, incorporate chemical and physiological properties, binding affinities, along with the targets of known chemical compounds. However, benchmarks for the classification of drugs into overarching functional groups based purely on the structure of the compound are yet to be explored. In this paper, we introduce VIRTECS: a tool that leverages the simplified molecular-input line-entry system (SMILES) – a structural representation of a drug – to enable virtual screening of large scale chemical databases, based on the therapeutic classes of drugs. The only input required by the system is the SMILES representation, one that is readily available with most computational generation approaches. The experimental results on multiple datasets demonstrate the potency of structural information in determining the functional groups of chemical compounds. VIRTECS holds enormous potential in yielding insights into various properties of novel molecules when an embedding of the SMILES input is used and paired with an apposite graph algorithm, and tested with known molecules. We present a framework that allows for multiple combinations of the input (SMILES with or without the embedding) and a choice of models and databases that can be tested based on the desired output: insight to the function or potential therapeutic value of a chemical compound.
{"title":"VIRTECS: Virtual Screening Of Therapeutic Classes Using Encodings Of Chemical Structures","authors":"Dweepa Honnavalli, Kavya Varma, G. Srinivasa","doi":"10.1109/ICRCICN50933.2020.9296180","DOIUrl":"https://doi.org/10.1109/ICRCICN50933.2020.9296180","url":null,"abstract":"In recent times, the need for virtual screening of chemical compounds has grown with the advent of computational synthesis and de-novo generation of drugs. The state-of-the-art benchmarks of virtual screening today, incorporate chemical and physiological properties, binding affinities, along with the targets of known chemical compounds. However, benchmarks for the classification of drugs into overarching functional groups based purely on the structure of the compound are yet to be explored. In this paper, we introduce VIRTECS: a tool that leverages the simplified molecular-input line-entry system (SMILES) – a structural representation of a drug – to enable virtual screening of large scale chemical databases, based on the therapeutic classes of drugs. The only input required by the system is the SMILES representation, one that is readily available with most computational generation approaches. The experimental results on multiple datasets demonstrate the potency of structural information in determining the functional groups of chemical compounds. VIRTECS holds enormous potential in yielding insights into various properties of novel molecules when an embedding of the SMILES input is used and paired with an apposite graph algorithm, and tested with known molecules. We present a framework that allows for multiple combinations of the input (SMILES with or without the embedding) and a choice of models and databases that can be tested based on the desired output: insight to the function or potential therapeutic value of a chemical compound.","PeriodicalId":138966,"journal":{"name":"2020 Fifth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","volume":"217 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116173074","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}
Mental workload contributes considerably to the outcome or the performance of any task. The concern of human workload increases during a human-machine collaboration task or in a multitasking environment. This paper presents a comparative study of machine learning algorithms used to estimate workload using Electroencephalography (EEG) data. An open-access EEG dataset acquired during a “simultaneous capacity (SIMKAP) experiment” and “no task” is used to create and validate models for binary classification of workload as present and absent respectively. The paper presents an implementation of various classification models that use EEG data to predict the workload. In this paper, implementation for KNN classifier (57.3%), Random Forest classifier (57.19%), MLP network classifier (58.2%), CNN+ LSTM network classifier (58.68%), and LSTM network classifier (61.08%) has been reported. The paper can be further extended to study operator workload in real-time using a brain-computer interface paradigm for any kind of task in a real-world application. The workload classification can be further used in human-machine tasks to decide task allocation between the system to achieve optimal performance in a complex critical system.
{"title":"Mental Workload Estimation Using EEG","authors":"Vishal Pandey, Dhirendra Kumar Choudhary, Vinita Verma, Greeshma Sharma, Ram Singh, Sushil Chandra","doi":"10.1109/ICRCICN50933.2020.9296150","DOIUrl":"https://doi.org/10.1109/ICRCICN50933.2020.9296150","url":null,"abstract":"Mental workload contributes considerably to the outcome or the performance of any task. The concern of human workload increases during a human-machine collaboration task or in a multitasking environment. This paper presents a comparative study of machine learning algorithms used to estimate workload using Electroencephalography (EEG) data. An open-access EEG dataset acquired during a “simultaneous capacity (SIMKAP) experiment” and “no task” is used to create and validate models for binary classification of workload as present and absent respectively. The paper presents an implementation of various classification models that use EEG data to predict the workload. In this paper, implementation for KNN classifier (57.3%), Random Forest classifier (57.19%), MLP network classifier (58.2%), CNN+ LSTM network classifier (58.68%), and LSTM network classifier (61.08%) has been reported. The paper can be further extended to study operator workload in real-time using a brain-computer interface paradigm for any kind of task in a real-world application. The workload classification can be further used in human-machine tasks to decide task allocation between the system to achieve optimal performance in a complex critical system.","PeriodicalId":138966,"journal":{"name":"2020 Fifth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","volume":"173 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116003764","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 : 2020-11-26DOI: 10.1109/ICRCICN50933.2020.9296198
M. Jambhale, S. Joshi, Sakshi Amrutkar, Neha Baisane
Taking public transport into consideration the seething crowd of India follows no or negligible discipline resulting in fatal accidents and mishaps. If people are provided with live information of the crowd, they can plan their journey accordingly with a prior knowledge of the vacancies at that moment. This paper intends to solve the problem of overcrowding by providing a real time system that will efficiently monitor the people count in every local compartment using ToF sensors. It uses Arduino and Ethernet to retrieve and transmit data respectively from sensors at the client side to the railway server system of the platform.
{"title":"Efficient Solution to Avoid Overcrowding in Local Train Compartments","authors":"M. Jambhale, S. Joshi, Sakshi Amrutkar, Neha Baisane","doi":"10.1109/ICRCICN50933.2020.9296198","DOIUrl":"https://doi.org/10.1109/ICRCICN50933.2020.9296198","url":null,"abstract":"Taking public transport into consideration the seething crowd of India follows no or negligible discipline resulting in fatal accidents and mishaps. If people are provided with live information of the crowd, they can plan their journey accordingly with a prior knowledge of the vacancies at that moment. This paper intends to solve the problem of overcrowding by providing a real time system that will efficiently monitor the people count in every local compartment using ToF sensors. It uses Arduino and Ethernet to retrieve and transmit data respectively from sensors at the client side to the railway server system of the platform.","PeriodicalId":138966,"journal":{"name":"2020 Fifth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127279724","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：