M. Agrawal, Rutuja Ainapure, Shrushti Agrawal, Simran Bhosale, S. Desai
{"title":"Models for Hand Gesture Recognition using Deep Learning","authors":"M. Agrawal, Rutuja Ainapure, Shrushti Agrawal, Simran Bhosale, S. Desai","doi":"10.1109/ICCCA49541.2020.9250846","DOIUrl":"https://doi.org/10.1109/ICCCA49541.2020.9250846","url":null,"abstract":"According to the World Health organization, 5% of the world population, approximately 466 million people, is deaf and/or mute or has disabling hearing loss. There is often a wall of distinction between handicapped people and normal people. We communicate to share our thoughts but for a disabled person (mainly deaf and dumb), it becomes difficult to communicate. Inability to speak is considered to be a true form of disability. For such people, sign language or Braille is the only means of communication. Sign Language is a way of communication using hand gestures. However, it becomes difficult for them to communicate with others as most don’t understand sign language. Hence, we aim at bridging this communication gap, between a deaf/mute person and others, by developing a system which acts as a mediator between both. We propose a hand gesture recognition system which works in 4 steps: 1. Generate a live stream of hand gestures using web-cam. 2. Form images from the video using video frames. 3. Preprocess these images. 4. Recognize sign language hand-gestures and convert into text/audio output. The system is implemented using the concepts of image processing and neural networks. We have tested the proposed models using Kaggle dataset, our dataset and a dataset formed after combining both. We propose to eliminate the ambiguity introduced in the results by inculcating variation in the background. Most of the models give similar accuracy of the test results for both plain and cluttered background.","PeriodicalId":260353,"journal":{"name":"2020 IEEE 5th International Conference on Computing Communication and Automation (ICCCA)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124089081","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}
{"title":"DriveCare: A Real-Time Vision Based Driver Drowsiness Detection Using Multiple Convolutional Neural Networks With Kernelized Correlation Filters (MCNN-KCF)","authors":"U. Gopikrishnan, R. Jose","doi":"10.1109/ICCCA49541.2020.9250868","DOIUrl":"https://doi.org/10.1109/ICCCA49541.2020.9250868","url":null,"abstract":"Driver Drowsiness is one of the most significant safety issues facing the road transport industry today, and it reduces vigilance, alertness, and concentration to perform the attention-based activities such as driving. When the driver is in a fatigue state, the facial expressions are a bit different from those in the normal state. Here a new system called DriveCare, which determines the drivers’ fatigue state, like duration of eyes closure, blinking, and yawning, using video images, without equipping their bodies with any devices, is introduced. A novel face-tracking algorithm called Multiple Convolutional Neural Networks with Kernelized Correlation Filters (MCNN-KCF) is used to improve the tracking accuracy. Further, a detection method is used for extracting the facial features of the driver. Then with the help of these facial features, we evaluate the Driver’s state. By combining the features extracted by the eyes and mouth, DriveCare can alert the driver using a drowsiness warning tone. Different experimental results showed that the DriveCare achieved around a 95% accuracy.","PeriodicalId":260353,"journal":{"name":"2020 IEEE 5th International Conference on Computing Communication and Automation (ICCCA)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121238203","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}
Paras Narendranath, Harika Jayanthi, N. Shreyas, B. Shambhavi, P. Jayarekha, Javaid Nabi
{"title":"Recognition of Named Entities and Categories in Text using Stacked Embeddings","authors":"Paras Narendranath, Harika Jayanthi, N. Shreyas, B. Shambhavi, P. Jayarekha, Javaid Nabi","doi":"10.1109/ICCCA49541.2020.9250886","DOIUrl":"https://doi.org/10.1109/ICCCA49541.2020.9250886","url":null,"abstract":"Named entities enable the identification of key elements in text while sentence classification provides for a summary of the same. Sequential labeling and sentence classification tasks together enable deeper extraction of information from text. Embeddings trained over a corpus pertaining to a specific domain, tend to generate strong vector representations thereby providing for the creation of better classification models. We propose custom fastText embeddings trained on a large Indian English news corpus. These embeddings are stacked with state-of-the-art Pooled Flair embeddings to generate an f1-score of 79 on a custom FIRE English NER dataset and 93.05 f1-score on a subset of the OntoNotes 5.0 dataset. The embeddings were also used for sentence classification on 20 news categories, to generate the best multi-class accuracy of 88.1%. We also propose two Indian news datasets, one based on the FIRE NER dataset and a custom multi-class sentence classification dataset.","PeriodicalId":260353,"journal":{"name":"2020 IEEE 5th International Conference on Computing Communication and Automation (ICCCA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129165456","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}
{"title":"Impact of Network Density on AODV protocol in VANET","authors":"Santosh Kumar, Achintya Kumar, Vishu Tyagi, Ashwini Kumar","doi":"10.1109/ICCCA49541.2020.9250898","DOIUrl":"https://doi.org/10.1109/ICCCA49541.2020.9250898","url":null,"abstract":"Vehicular Ad-hoc Network (VANET) is attracting and significant consideration in the automotive technology to advance the solutions of the Intelligent Transportation System (ITS). Since routing plays a major role in the highly mobile node in VANET. Therefore, the Study of an Adhoc routing protocol in various network density is essential because it contribute significantly in various real applications such as safeguarding drivers and passengers mutually. There are various routing protocols with the aim of proposing and assessing to enhance the effectiveness of VANET. In this work, AODV protocol has taken as a routing services to create a scenario for VANET simulation task using various tools for their specific purpose viz SUMO, MOVE and NS 2.35. The effective implementation of the AODV has studied and analyzed with various QoS parameters against different network density.","PeriodicalId":260353,"journal":{"name":"2020 IEEE 5th International Conference on Computing Communication and Automation (ICCCA)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129283570","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}
{"title":"Unsupervised video summarization framework using keyframe extraction and video skimming","authors":"Shruti Jadon, Mahmood Jasim","doi":"10.1109/ICCCA49541.2020.9250764","DOIUrl":"https://doi.org/10.1109/ICCCA49541.2020.9250764","url":null,"abstract":"Video is a robust sources of information and the consumption of online and offline videos has reached an unprecedented level in the last few years. However, the extraction of information from a video presents more challenges than the extraction of information from a picture. To extract the context of the video, a viewer has to go through the whole video. Apart from context understanding, it almost impossible to create a universal summarized video for everyone as everyone has their own bias of keyframe. Example, in a soccer game, a coach might prefer frames which consist of information on player placement and techniques. However, a person with less knowledge about soccer will focus more on frames which consist of goals and score-board. Therefore, if we were to tackle problem video summarization through a supervised learning path, it will require extensive personalized data labeling. In this paper, we attempt to solve video summarization through unsupervised learning by employing traditional vision-based algorithmic methodologies for accurate feature extraction from video frames. We have also proposed a deep learning based feature extraction followed by multiple clustering methods to find an effective way of summarizing a video by interesting keyframe extraction. We have compared the performance of these approaches on the SumMe dataset and showcased that using deep learning-based feature extraction has been proven to perform better in dynamic viewpoint videos.","PeriodicalId":260353,"journal":{"name":"2020 IEEE 5th International Conference on Computing Communication and Automation (ICCCA)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114322976","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}
Srimanta Kundu, Utsav Bandyopadhyay Maulik, Arnabi Bej, U. Maulik
{"title":"Deep Learning based Pollution Detection in Intelligent Transportation System","authors":"Srimanta Kundu, Utsav Bandyopadhyay Maulik, Arnabi Bej, U. Maulik","doi":"10.1109/ICCCA49541.2020.9250883","DOIUrl":"https://doi.org/10.1109/ICCCA49541.2020.9250883","url":null,"abstract":"Vehicle pollution is a major concern in the current world. The technologies are improving day by day to reduce the pollution caused by the car. However, we are still lagging from addressing this critical issue completely. Therefore, road surveillance should be made stringent to capture those vehicles causing this serious problem. In this article, we have proposed a deep learning based framework which will identify the vehicle pollution from the images captured by the on-road surveillance camera. We have prepared an enriched large data set with significant variations which has been used in the training phase while deploying the deep model. The experiment has dealt with three base line Deep Learning CNN models, i.e. Inception-V3, MobileNet-V2 and InceptionResNet-V2. Transfer learning concept has been exploited to identify the on-road polluting vehicle. The experimental outcome has demonstrated the supremacy of the proposed approach in the traffic surveillance domain.","PeriodicalId":260353,"journal":{"name":"2020 IEEE 5th International Conference on Computing Communication and Automation (ICCCA)","volume":"83 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126072261","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}
Zameer A. Khan, D. Mishra, Vardaan Sharma, Akshay Sharma
{"title":"Empirical Study of Various Classification Techniques for Heart Disease Prediction","authors":"Zameer A. Khan, D. Mishra, Vardaan Sharma, Akshay Sharma","doi":"10.1109/ICCCA49541.2020.9250852","DOIUrl":"https://doi.org/10.1109/ICCCA49541.2020.9250852","url":null,"abstract":"It is a known fact that heart acts as one of the most vital as well as complex organ of the body. Without it, majority of living beings will cease to exist. The main function that is performed by the heart is pumping the blood and supplying it to all organs of the whole body, which makes it extremely important. And this is the reason the heart related diseases are extremely sensitive and need to be handled with utmost care. Often it is said that prevention is better than cure. For majority of heart related problems, we often find it after the problem has occurred. But research has shown that around 90% of the cardiovascular diseases can be cured if predicted in advance. From the findings of this study that we have conducted we can find the relation between various factors like age, blood pressure, gender etc. The analysis of these things will help us in developing a deeper understanding of these factors on heart disease. Furthermore, we are modelling the various classification models, KNN, Decision Trees, Logistic Regression, Gaussian Naïve Bias, SVM and Random Forests. It is extremely important to predict early on if there is an ailment of heart or not. Heart is the most crucial and sensitive organ of our body and this is a field of critical and crucial work. It provides a medical foresight and helps avid many life-threatening situations. his paper provides an insight of the existing algorithm and it gives an overall summary of the existing work","PeriodicalId":260353,"journal":{"name":"2020 IEEE 5th International Conference on Computing Communication and Automation (ICCCA)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126111599","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}
Chinthaka Amarasinghe, A. Ratnaweera, Sanjeeva Maitripala
{"title":"Hybrid Feature Enhance Filter (HFEF) for Underwater Vision Navigation","authors":"Chinthaka Amarasinghe, A. Ratnaweera, Sanjeeva Maitripala","doi":"10.1109/ICCCA49541.2020.9250713","DOIUrl":"https://doi.org/10.1109/ICCCA49541.2020.9250713","url":null,"abstract":"This research presents a real-time visibility enhancement algorithm for effective underwater visual navigation. Unlike an aerial environment, an underwater environment is a poor visibility as light travels in the water and the resulting scenes are poorly contrasted and hazy. At present, several vision-based navigation algorithms were introduced by the ground robotic communities. However, most of them fail in underwater due to image degradation. But, there is a possibility to use the same vision navigation methods with image preprocessing which addresses image degradation. Presented paper address the turbidity of the water, the most common problem in underwater image preprocessing, and improved feature point detection and matching which can be used in a vision-based navigation algorithm effectively. The proposed enhancement method combining with, noise removing, artificial light estimation, de-hazing, adaptive depth aware contrast stretching, and image sharpening. The proposed method is validated using two underwater turbidity datasets, one for different turbidity levels and one for visual SLAM with OpenVSLAM. The result showed the proposed visibility enhancement algorithm improved the underwater visual SLAM navigation significantly.","PeriodicalId":260353,"journal":{"name":"2020 IEEE 5th International Conference on Computing Communication and Automation (ICCCA)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125713459","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}
{"title":"Signature-Based Video Watermarking using SVD DTCWT","authors":"Rammohannath Mishra, Shelly Sachdeva","doi":"10.1109/ICCCA49541.2020.9250854","DOIUrl":"https://doi.org/10.1109/ICCCA49541.2020.9250854","url":null,"abstract":"With the development in the electronics and communication industry, i.e., super high-speed internet connectivity and super-fast hardware and software, it is not so hard to get access to most of the content on the internet. Most of the multimedia content on the internet, either it is copyrighted or not, is on the fingertips; because of some websites and torrent sites. Cinema industry is on its knees because of this much illegal distribution of millions of copies of unreleased movies, pirated copies of movies. Therefore, to limit the unlawful distribution of multimedia content, watermarking techniques have been used for years. In this paper, a blind video watermarking technique based on Dual-Tree Complex Wavelet Transform (DTCWT), Singular Value Decomposition (SVD), and chrominance channel, is presented, which uses a signature-based approach for authentication.","PeriodicalId":260353,"journal":{"name":"2020 IEEE 5th International Conference on Computing Communication and Automation (ICCCA)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121960674","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}
Sampurna Mandal, Swagatam Biswas, V. Balas, R. Shaw, Ankush Ghosh
{"title":"Motion Prediction for Autonomous Vehicles from Lyft Dataset using Deep Learning","authors":"Sampurna Mandal, Swagatam Biswas, V. Balas, R. Shaw, Ankush Ghosh","doi":"10.1109/ICCCA49541.2020.9250790","DOIUrl":"https://doi.org/10.1109/ICCCA49541.2020.9250790","url":null,"abstract":"Autonomous Vehicles are expected to change the future of worldwide transportation system. As self-driving cars are facing a lot of engineering challenges, it is one of the hottest topics in recent research. One such challenge is to build models to predict the movements of traffic agents such as cars, cyclists, pedestrians etc around the self-driving cars. The objective of this paper is to analyse the prediction efficiency of various deep learning models by calculating root mean square error score. This deep learning models takes a current state of the surrounding and depending on that predict the motion for the agents.","PeriodicalId":260353,"journal":{"name":"2020 IEEE 5th International Conference on Computing Communication and Automation (ICCCA)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123825883","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}