Pub Date : 2023-03-01DOI: 10.1109/ESCI56872.2023.10099903
Gargee Athalye, Atharva Sarde, Mayur Badgujar, Vijay Gaikwad, S. Sondkar
Heart diseases are prevalent in today's world due to many factors like lipid disorder (hypercholesterolemia), corpulence (obesity), increase in triglycerides levels (lipids obtained from esterification fatty acids to glycerol), hypertension, etc. It is estimated that nearly 18 million lives are affected yearly due to various heart diseases. Early detection of such diseases could help save several lives. In the proposed system, heart failure prediction is estimated using the combination of Gradient boost detection and decision trees. The parallel handling approach is used for feature processing to speed up the results and for optimal performance. The generation and discrimination approach are used to verify the outcomes concerning other algorithms and pseudo-codes. This paper uses the data file from the University of California, Irvine Intelligent Systems Repository to test the results. It is observed from several experiments that it provides optimal performance compared to the remaining predictors in the context of f1 score, recall, and accuracy. The ROC curve of Gradient Boost provides a higher deviation for low false positives. The Gradient Boost shows a 0.919 ROC value and 92 % of accuracy with an F1 score of 0.928 and a recall of 0.934.
{"title":"Hybrid Gradient Boost based Heart Failure Prediction System","authors":"Gargee Athalye, Atharva Sarde, Mayur Badgujar, Vijay Gaikwad, S. Sondkar","doi":"10.1109/ESCI56872.2023.10099903","DOIUrl":"https://doi.org/10.1109/ESCI56872.2023.10099903","url":null,"abstract":"Heart diseases are prevalent in today's world due to many factors like lipid disorder (hypercholesterolemia), corpulence (obesity), increase in triglycerides levels (lipids obtained from esterification fatty acids to glycerol), hypertension, etc. It is estimated that nearly 18 million lives are affected yearly due to various heart diseases. Early detection of such diseases could help save several lives. In the proposed system, heart failure prediction is estimated using the combination of Gradient boost detection and decision trees. The parallel handling approach is used for feature processing to speed up the results and for optimal performance. The generation and discrimination approach are used to verify the outcomes concerning other algorithms and pseudo-codes. This paper uses the data file from the University of California, Irvine Intelligent Systems Repository to test the results. It is observed from several experiments that it provides optimal performance compared to the remaining predictors in the context of f1 score, recall, and accuracy. The ROC curve of Gradient Boost provides a higher deviation for low false positives. The Gradient Boost shows a 0.919 ROC value and 92 % of accuracy with an F1 score of 0.928 and a recall of 0.934.","PeriodicalId":441215,"journal":{"name":"2023 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126896066","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 : 2023-03-01DOI: 10.1109/ESCI56872.2023.10099759
Smita S. Kulkarni, Sangeeta Jadhav
This work proposes a video understanding technique that primarily focuses on the individual action recognition appearing in the video. The state-of-the-art showed promising work in video understanding. Though, it's essential to require inclusive information on human action in real-time CCTV video surveillance, sports video analysis, health care, etc. This paper proposed a transfer learning deep neural network model designed for recognizing individual actions accomplished by multiple people in a video sequence. This research established a deep model which uses Region-Of-Interest (RoI) pooling layer to capture automated features from a specified video frame to recognize individual actions. The MobileNet model accomplishes this as the backbone to recognize individual actions from each video frame. The accuracy score of the model was compared with the CNN models VGG-19,InceptionV3, and MobileNet. The MobileNet is computationally low-cost and enhances the performance of individual action recognition performed by multiple humans in a video frame. The investigational results were evaluated by varying learning parameters, and optimizer of deep neural network. The experimental results of the proposed model for individual action recognition demonstrate the improved efficiency of the standard benchmark collective activity dataset. This research illustrates the progress of action recognition by employing the transfer learning CNN model along with RoI pooling layer.
{"title":"Insight on Human Activity Recognition Using the Deep Learning Approach","authors":"Smita S. Kulkarni, Sangeeta Jadhav","doi":"10.1109/ESCI56872.2023.10099759","DOIUrl":"https://doi.org/10.1109/ESCI56872.2023.10099759","url":null,"abstract":"This work proposes a video understanding technique that primarily focuses on the individual action recognition appearing in the video. The state-of-the-art showed promising work in video understanding. Though, it's essential to require inclusive information on human action in real-time CCTV video surveillance, sports video analysis, health care, etc. This paper proposed a transfer learning deep neural network model designed for recognizing individual actions accomplished by multiple people in a video sequence. This research established a deep model which uses Region-Of-Interest (RoI) pooling layer to capture automated features from a specified video frame to recognize individual actions. The MobileNet model accomplishes this as the backbone to recognize individual actions from each video frame. The accuracy score of the model was compared with the CNN models VGG-19,InceptionV3, and MobileNet. The MobileNet is computationally low-cost and enhances the performance of individual action recognition performed by multiple humans in a video frame. The investigational results were evaluated by varying learning parameters, and optimizer of deep neural network. The experimental results of the proposed model for individual action recognition demonstrate the improved efficiency of the standard benchmark collective activity dataset. This research illustrates the progress of action recognition by employing the transfer learning CNN model along with RoI pooling layer.","PeriodicalId":441215,"journal":{"name":"2023 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114174375","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 : 2023-03-01DOI: 10.1109/ESCI56872.2023.10100064
Vipan Kusla, Gurbinder Singh Brar, Vikas K. Garg, Ankit Bansal, R. Kaushal
A wireless sensor network (WSN) improves wireless communication by using hundreds or thousands of nodes to gather data. The lifespan of the nodes and balanced energy consumption are the major issues in the WSN. Long-term WSN efficiency requires optimising node energy. Selecting the optimal node as the cluster head improves energy usage in wireless sensor networks. The Artificial Hummingbird algorithm is used in this paper to identify the best cluster head selection in homogenous wireless sensor networks. The proposed algorithm's innovation lies in the fact that it takes into account a number of parameters like residual energy, intra-cluster distance, and balanced cluster formation while choosing a CH from a homogeneous sensor network. The performance analysis of the proposed algorithm considers four parameters: average energy consumption, total energy consumption, first node death, and residual energy. When compared to other algorithms, MATLAB-based simulation analyses show that the proposed algorithm MAHA-EECHS outperforms them.
{"title":"Meta-heuristic Artificial Humming Bird Algorithm Based Energy Efficient Cluster Head Selection (MAHA-EECHS) in Wireless Sensor Networks","authors":"Vipan Kusla, Gurbinder Singh Brar, Vikas K. Garg, Ankit Bansal, R. Kaushal","doi":"10.1109/ESCI56872.2023.10100064","DOIUrl":"https://doi.org/10.1109/ESCI56872.2023.10100064","url":null,"abstract":"A wireless sensor network (WSN) improves wireless communication by using hundreds or thousands of nodes to gather data. The lifespan of the nodes and balanced energy consumption are the major issues in the WSN. Long-term WSN efficiency requires optimising node energy. Selecting the optimal node as the cluster head improves energy usage in wireless sensor networks. The Artificial Hummingbird algorithm is used in this paper to identify the best cluster head selection in homogenous wireless sensor networks. The proposed algorithm's innovation lies in the fact that it takes into account a number of parameters like residual energy, intra-cluster distance, and balanced cluster formation while choosing a CH from a homogeneous sensor network. The performance analysis of the proposed algorithm considers four parameters: average energy consumption, total energy consumption, first node death, and residual energy. When compared to other algorithms, MATLAB-based simulation analyses show that the proposed algorithm MAHA-EECHS outperforms them.","PeriodicalId":441215,"journal":{"name":"2023 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115214896","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 : 2023-03-01DOI: 10.1109/ESCI56872.2023.10100307
Vijaykumar R. Bhanuse, S. Patankar, J. Kulkarni
Metallic plates are used in structural work and therefore it is necessary to analyze plate vibrations. Measurement of vibration analysis is important in preventive maintenance in many process to avoid failure of structural elements due to vibrations. Fundamental frequency is one of the important characteristic of material. These frequencies are used to determine material properties. This paper presents harmonic analysis of metallic by ball falling with different height. Vibration caused by impact is detected by a piezoelectric acceleration sensor. A spectral analysis of the detected vibration signal is performed using the Matlab platform and the fundamental frequency plate is estimated. It is observed that fundamental frequency remains nearly constant. Compare fundamental frequency estimated by impact testing with ANSYS software. It is observed average estimated fundamental frequency error is 4.152%.
{"title":"Harmonic Analysis of Mild steel plate","authors":"Vijaykumar R. Bhanuse, S. Patankar, J. Kulkarni","doi":"10.1109/ESCI56872.2023.10100307","DOIUrl":"https://doi.org/10.1109/ESCI56872.2023.10100307","url":null,"abstract":"Metallic plates are used in structural work and therefore it is necessary to analyze plate vibrations. Measurement of vibration analysis is important in preventive maintenance in many process to avoid failure of structural elements due to vibrations. Fundamental frequency is one of the important characteristic of material. These frequencies are used to determine material properties. This paper presents harmonic analysis of metallic by ball falling with different height. Vibration caused by impact is detected by a piezoelectric acceleration sensor. A spectral analysis of the detected vibration signal is performed using the Matlab platform and the fundamental frequency plate is estimated. It is observed that fundamental frequency remains nearly constant. Compare fundamental frequency estimated by impact testing with ANSYS software. It is observed average estimated fundamental frequency error is 4.152%.","PeriodicalId":441215,"journal":{"name":"2023 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"43 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116560469","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 : 2023-03-01DOI: 10.1109/ESCI56872.2023.10099923
Neelima K, Kumar Raja Meruva, C. Subhas
This paper aims at development of Fusion algorithms for Multimodal Images using Xilinx System Generator for ease of using available built-in Field Programmable Gate Array based hardware algorithms for image fusion. This facilitates the development of new enhanced algorithms easily by using built-in blocks. Further this reduces the design effort, enhances hardware utilization. Further the comparison parameters like mean, standard deviation, peak signal to noise ratio, variance, root mean square error, kurtosis etc are compared with the existing fusion algorithms like DWT, SIDWT, PCA, DCT, etc. Xilinx ISE 14.5 is used as synthesis tool for Zynq 7000 Series 28nm FPGA board with part number XC7Z100-1FFG1156. The existing and modified architectures are implemented using Xilinx System Generator (XSG) as a cosimulation with MATLAB 2015a Simulink workspace. The proposed XSG based FPGA Image Fusion Algorithm proves to be a better choice of implementation with a scope of reconfigurability.
本文旨在使用Xilinx System Generator开发多模态图像的融合算法，以便于使用可用的内置基于现场可编程门阵列的硬件算法进行图像融合。这有助于通过使用内置块轻松开发新的增强算法。进一步减少了设计工作量，提高了硬件利用率。并将均值、标准差、峰值信噪比、方差、均方根误差、峰度等比较参数与现有的DWT、SIDWT、PCA、DCT等融合算法进行比较。Xilinx ISE 14.5用作Zynq 7000系列28nm FPGA板的合成工具，零件号为XC7Z100-1FFG1156。使用Xilinx System Generator (XSG)作为MATLAB 2015a Simulink工作空间的协同仿真来实现现有和修改的架构。实验证明，基于XSG的FPGA图像融合算法具有较好的可重构性，是一种较好的实现选择。
{"title":"Image Fusion using Xilinx System Generator for MRI and CT Medical Image Modalities","authors":"Neelima K, Kumar Raja Meruva, C. Subhas","doi":"10.1109/ESCI56872.2023.10099923","DOIUrl":"https://doi.org/10.1109/ESCI56872.2023.10099923","url":null,"abstract":"This paper aims at development of Fusion algorithms for Multimodal Images using Xilinx System Generator for ease of using available built-in Field Programmable Gate Array based hardware algorithms for image fusion. This facilitates the development of new enhanced algorithms easily by using built-in blocks. Further this reduces the design effort, enhances hardware utilization. Further the comparison parameters like mean, standard deviation, peak signal to noise ratio, variance, root mean square error, kurtosis etc are compared with the existing fusion algorithms like DWT, SIDWT, PCA, DCT, etc. Xilinx ISE 14.5 is used as synthesis tool for Zynq 7000 Series 28nm FPGA board with part number XC7Z100-1FFG1156. The existing and modified architectures are implemented using Xilinx System Generator (XSG) as a cosimulation with MATLAB 2015a Simulink workspace. The proposed XSG based FPGA Image Fusion Algorithm proves to be a better choice of implementation with a scope of reconfigurability.","PeriodicalId":441215,"journal":{"name":"2023 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125250384","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 : 2023-03-01DOI: 10.1109/ESCI56872.2023.10099666
Masaru Sasaki, T. Shibanoki, H. Tonooka
In this study, a tactile feedback method is proposed for virtual repulsive force based on non-contact impedance and applied to collision avoidance for a biological signal-controlled mobile robot. Virtual walls based on mechanical impedance are placed around the robot, it can avoid obstacles using virtual repulsive forces when obstacles meet the virtual walls. The proposed method provides tactile feedback about the force to the operator, which enables the operator to recognize the environment around the robot. In the experiments, a blindfolded participant controlled a mobile robot using myoelectric signals. The results demonstrated that the robot could be operated stably.
{"title":"Mobile Robot Control Based on Virtual Impedance Force Feedback","authors":"Masaru Sasaki, T. Shibanoki, H. Tonooka","doi":"10.1109/ESCI56872.2023.10099666","DOIUrl":"https://doi.org/10.1109/ESCI56872.2023.10099666","url":null,"abstract":"In this study, a tactile feedback method is proposed for virtual repulsive force based on non-contact impedance and applied to collision avoidance for a biological signal-controlled mobile robot. Virtual walls based on mechanical impedance are placed around the robot, it can avoid obstacles using virtual repulsive forces when obstacles meet the virtual walls. The proposed method provides tactile feedback about the force to the operator, which enables the operator to recognize the environment around the robot. In the experiments, a blindfolded participant controlled a mobile robot using myoelectric signals. The results demonstrated that the robot could be operated stably.","PeriodicalId":441215,"journal":{"name":"2023 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122862380","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 : 2023-03-01DOI: 10.1109/ESCI56872.2023.10100323
Praveen Sankarasubramanian
Extreme precautions must be observed to handle toxic wastes, radioactive substances, chemical raw materials, chemical wastes, and bio-products in different industries. Any malfunction in a dangerous traffic network can lead to serious accidents, deaths and / or serious damage. Direct monitoring and analysis, and preventive measures to prevent the spread of failures, can significantly reduce the recurrence of adverse effects. Current research suggests that detailed publicity and information on the latest developments in pipeline monitoring and research may help modernize the oil industry in the future. We also propose a framework to detect timely leakage in pipelines, especially in oil and gas sector.
{"title":"Protection of Hazardous Places in Industries using Machine Learning","authors":"Praveen Sankarasubramanian","doi":"10.1109/ESCI56872.2023.10100323","DOIUrl":"https://doi.org/10.1109/ESCI56872.2023.10100323","url":null,"abstract":"Extreme precautions must be observed to handle toxic wastes, radioactive substances, chemical raw materials, chemical wastes, and bio-products in different industries. Any malfunction in a dangerous traffic network can lead to serious accidents, deaths and / or serious damage. Direct monitoring and analysis, and preventive measures to prevent the spread of failures, can significantly reduce the recurrence of adverse effects. Current research suggests that detailed publicity and information on the latest developments in pipeline monitoring and research may help modernize the oil industry in the future. We also propose a framework to detect timely leakage in pipelines, especially in oil and gas sector.","PeriodicalId":441215,"journal":{"name":"2023 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122881838","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 : 2023-03-01DOI: 10.1109/ESCI56872.2023.10099753
Bhuvana Kumbhare, K. Akant, M. Khanapurkar, P. Chandankhede
In level 2 automated cars, functioning breaks are currently present. New systems must be evaluated in a broad range of difficult scenarios in order to boost automation while assuring all-around safety. There are several disadvantages to validating these systems on real cars, including the time required to drive millions of kilometers, the danger involved in particular circumstances, and the high expense. Platforms for simulation show up as a suitable solution. In order to evaluate autonomous driving maneuvers and control methods, strong and trustworthy virtual environments are required. To that end, this study offers strategies which are created, adjusted, and verified using a custom simulation framework before being implemented in a real vehicle. A multibody vehicle model is used to calculate the simulation's dynamics. The usefulness of the suggested approach for creating and verifying longitudinal controllers for actual automated vehicles is demonstrated by a comparison of outcomes.
{"title":"Longitudinal Control for closed loop simulation of Autonomous driving Vehicle","authors":"Bhuvana Kumbhare, K. Akant, M. Khanapurkar, P. Chandankhede","doi":"10.1109/ESCI56872.2023.10099753","DOIUrl":"https://doi.org/10.1109/ESCI56872.2023.10099753","url":null,"abstract":"In level 2 automated cars, functioning breaks are currently present. New systems must be evaluated in a broad range of difficult scenarios in order to boost automation while assuring all-around safety. There are several disadvantages to validating these systems on real cars, including the time required to drive millions of kilometers, the danger involved in particular circumstances, and the high expense. Platforms for simulation show up as a suitable solution. In order to evaluate autonomous driving maneuvers and control methods, strong and trustworthy virtual environments are required. To that end, this study offers strategies which are created, adjusted, and verified using a custom simulation framework before being implemented in a real vehicle. A multibody vehicle model is used to calculate the simulation's dynamics. The usefulness of the suggested approach for creating and verifying longitudinal controllers for actual automated vehicles is demonstrated by a comparison of outcomes.","PeriodicalId":441215,"journal":{"name":"2023 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131407831","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}
In the contemporary world, numerous nations are developing various technologies for national security. Therefore, it is important to think about the safety of military personnel who defend the national security of their individual nations. Numerous soldiers perish during a battle in desolate places or near borders. Landmines are explosive weapons that can be hidden in the ground and are set off when someone steps on them with just 9 kg of pressure. Additionally, it harms the structure of the soil, lowers soil productivity, and makes the soil more susceptible to wind and water erosion. After a battle is over, mines continue to harm innocent civilians and soldiers. This study suggests a robotic vehicle with a metal detector that can identify mines in front of it in order to save soldiers' lives and defuse minefields. Additionally, it will provide latitude and longitude information after locating the location of a landmine using a GPS and GSM module.
{"title":"Mine Detecting Military Bot Using IoT","authors":"M. Rane, Manas Jain, Aryan Kashyap, Adhip Jajoo, Harshvardhan Kadam, Devika Kadam","doi":"10.1109/ESCI56872.2023.10100211","DOIUrl":"https://doi.org/10.1109/ESCI56872.2023.10100211","url":null,"abstract":"In the contemporary world, numerous nations are developing various technologies for national security. Therefore, it is important to think about the safety of military personnel who defend the national security of their individual nations. Numerous soldiers perish during a battle in desolate places or near borders. Landmines are explosive weapons that can be hidden in the ground and are set off when someone steps on them with just 9 kg of pressure. Additionally, it harms the structure of the soil, lowers soil productivity, and makes the soil more susceptible to wind and water erosion. After a battle is over, mines continue to harm innocent civilians and soldiers. This study suggests a robotic vehicle with a metal detector that can identify mines in front of it in order to save soldiers' lives and defuse minefields. Additionally, it will provide latitude and longitude information after locating the location of a landmine using a GPS and GSM module.","PeriodicalId":441215,"journal":{"name":"2023 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132287746","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 : 2023-03-01DOI: 10.1109/ESCI56872.2023.10099639
B. Solunke, S. Gengaje
Fast and accurate object detection systems are in high demand due to the advent of autonomous vehicles, smart video surveillance, facial detection, and numerous people counting applications. These systems not only detect and classify every object in an image or video, but also locate each one by creating a bounding box around it. This paper analyses the traditional and recent deep learning-based object detection methods from different perspectives, incorporating features recognition on many scales, data expansion, training approach, and perspective detection, in order to make it easier to deeply understand object detection. Some commonly used standard datasets for object detection are discussed. It also addressed the challenges and possible research scope in the future from the perspective of evolving object detection datasets and the framework for object detection tasks. From the analysis, it is observed that the performance of the methods in use for object detection is moderate and requires improvement, especially in difficult environments such as large object scale variance, obstructed object view, and horrific mild prerequisites. Therefore, the possible research scope for inventions and implementation of more novel deep learning methods to enhance object detection and classification accuracy is discussed.
{"title":"A Review on Traditional and Deep Learning based Object Detection Methods","authors":"B. Solunke, S. Gengaje","doi":"10.1109/ESCI56872.2023.10099639","DOIUrl":"https://doi.org/10.1109/ESCI56872.2023.10099639","url":null,"abstract":"Fast and accurate object detection systems are in high demand due to the advent of autonomous vehicles, smart video surveillance, facial detection, and numerous people counting applications. These systems not only detect and classify every object in an image or video, but also locate each one by creating a bounding box around it. This paper analyses the traditional and recent deep learning-based object detection methods from different perspectives, incorporating features recognition on many scales, data expansion, training approach, and perspective detection, in order to make it easier to deeply understand object detection. Some commonly used standard datasets for object detection are discussed. It also addressed the challenges and possible research scope in the future from the perspective of evolving object detection datasets and the framework for object detection tasks. From the analysis, it is observed that the performance of the methods in use for object detection is moderate and requires improvement, especially in difficult environments such as large object scale variance, obstructed object view, and horrific mild prerequisites. Therefore, the possible research scope for inventions and implementation of more novel deep learning methods to enhance object detection and classification accuracy is discussed.","PeriodicalId":441215,"journal":{"name":"2023 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"43 13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114299177","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：