Pub Date : 2023-07-19DOI: 10.1109/ICECAA58104.2023.10212315
W. Auccahuasi, Oscar Linares, Luis Vivanco-Aldon, Martin Campos-Martinez, Humberto Quispe-Peña, Julia Sobrino-Mesias
Epilepsy is one of the characteristic diseases of brain activity. People with this disease live with the symptoms and try to control the events to mitigate the effects that they may cause, such as a fall, blow, or any other consequence. In these times of pandemic caused by COVID-19, many of the people who have this disease present many events in a row, which can be caused by many factors, such as being at home most of the day, without being able to go out to get distracted. In this research work, a practical method is presented to monitor and predict an epilepsy event, based on the measurement of stress and meditation levels by using low-cost devices. For the evaluation of the method, measurements were made on a patient who constantly presents epilepsy events. The evaluation was carried out when she took her classes online, where the students present greater pressure. The method uses smartwatches to evaluate the stress level and BCI devices to measure the level of meditation. The results found in the data analysis present grouped values for the positive and negative cases of happening of epilepsy events. In the evaluation, possible threshold values that can be used to classify epilepsy events were determined. The determined threshold value can be used independently if only one device can be counted on. The reference threshold value is determined between values of 41 and 79, on a scale of 1 to 100. It is concluded that the device that can be counted on in the market at a low cost is the smartwatch that measures the stress level, compared to the best-known brain signal analysis device. As for the BCI device, the presented method is easy to implement; it can be easily used by the patients themselves or their relatives.
{"title":"Method for Monitoring Patterns in the Behavior of Brain Activity Prior to an Episode of Epilepsy, Applied to Young People in Times of the Covid-19 Pandemic, using Low-Cost BCI Devices","authors":"W. Auccahuasi, Oscar Linares, Luis Vivanco-Aldon, Martin Campos-Martinez, Humberto Quispe-Peña, Julia Sobrino-Mesias","doi":"10.1109/ICECAA58104.2023.10212315","DOIUrl":"https://doi.org/10.1109/ICECAA58104.2023.10212315","url":null,"abstract":"Epilepsy is one of the characteristic diseases of brain activity. People with this disease live with the symptoms and try to control the events to mitigate the effects that they may cause, such as a fall, blow, or any other consequence. In these times of pandemic caused by COVID-19, many of the people who have this disease present many events in a row, which can be caused by many factors, such as being at home most of the day, without being able to go out to get distracted. In this research work, a practical method is presented to monitor and predict an epilepsy event, based on the measurement of stress and meditation levels by using low-cost devices. For the evaluation of the method, measurements were made on a patient who constantly presents epilepsy events. The evaluation was carried out when she took her classes online, where the students present greater pressure. The method uses smartwatches to evaluate the stress level and BCI devices to measure the level of meditation. The results found in the data analysis present grouped values for the positive and negative cases of happening of epilepsy events. In the evaluation, possible threshold values that can be used to classify epilepsy events were determined. The determined threshold value can be used independently if only one device can be counted on. The reference threshold value is determined between values of 41 and 79, on a scale of 1 to 100. It is concluded that the device that can be counted on in the market at a low cost is the smartwatch that measures the stress level, compared to the best-known brain signal analysis device. As for the BCI device, the presented method is easy to implement; it can be easily used by the patients themselves or their relatives.","PeriodicalId":114624,"journal":{"name":"2023 2nd International Conference on Edge Computing and Applications (ICECAA)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115748485","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 : 2023-07-19DOI: 10.1109/ICECAA58104.2023.10212192
S. Kavitha, Safreen, Padmasree, Niskriyaa, Keerthana
People require transportation to move from one place to another place nowadays. As the population expands, so does transportation across the country. Despite the high demand for vehicles, the number of accidents are increasing due to the lack of safety precautions. Moreover, major accidents occur for two-wheeler drivers because two-wheelers lack safety precautions that four-wheelers have. Many research studies that are conducted recently to determine the cause of accidents concluded that driving recklessly in crowded areas causes road accidents. The major goal of this research work is to minimize the road accidents and prevent road accidents while ensuring driver safety. The proposed model enhances human safety while driving is by employing Arduino UNO and IoT. The vehicle will only start if the driver wears helmet and not intoxicated. To track the vehicle's current location, GPS module is used. To avoid accidents due to vehicle overspeed, the speed of vehicle is continuously monitored. In a particular zone, the driver will be alerted to reduce the speed.
{"title":"Human Safety and Vehicle Speed Controlling using Internet of Things (IoT)","authors":"S. Kavitha, Safreen, Padmasree, Niskriyaa, Keerthana","doi":"10.1109/ICECAA58104.2023.10212192","DOIUrl":"https://doi.org/10.1109/ICECAA58104.2023.10212192","url":null,"abstract":"People require transportation to move from one place to another place nowadays. As the population expands, so does transportation across the country. Despite the high demand for vehicles, the number of accidents are increasing due to the lack of safety precautions. Moreover, major accidents occur for two-wheeler drivers because two-wheelers lack safety precautions that four-wheelers have. Many research studies that are conducted recently to determine the cause of accidents concluded that driving recklessly in crowded areas causes road accidents. The major goal of this research work is to minimize the road accidents and prevent road accidents while ensuring driver safety. The proposed model enhances human safety while driving is by employing Arduino UNO and IoT. The vehicle will only start if the driver wears helmet and not intoxicated. To track the vehicle's current location, GPS module is used. To avoid accidents due to vehicle overspeed, the speed of vehicle is continuously monitored. In a particular zone, the driver will be alerted to reduce the speed.","PeriodicalId":114624,"journal":{"name":"2023 2nd International Conference on Edge Computing and Applications (ICECAA)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124929086","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 : 2023-07-19DOI: 10.1109/ICECAA58104.2023.10212215
Sohail Hussain, R. Karthikeyan, S. Ramamoorthi, I. Arafat, S. M. Gani
Internet helps the users to develop their skills as well as for doing their day to day activities. Regular work of the users is generally done via online. If people want to do money based transaction or sending important message to their neighbor, they use online network. This can be used and interfere by the attackers. Many attackers utilize the tools to disturb the authorized user job. Major attack utilized by the attacker to disturb the service get by the authorized users is denial of service. It becomes major kind of procedure done by the intruder to take control over system or to down the server. In this paper we analyze the DoS attack including its distributed version. Impacts of DoS are also discussed. Based on the survey and analysis we suggest the model for handling the DoS attack dynamically is also mentioned.
{"title":"Denial of Service Attack Analysis Using Machine Learning Techniques","authors":"Sohail Hussain, R. Karthikeyan, S. Ramamoorthi, I. Arafat, S. M. Gani","doi":"10.1109/ICECAA58104.2023.10212215","DOIUrl":"https://doi.org/10.1109/ICECAA58104.2023.10212215","url":null,"abstract":"Internet helps the users to develop their skills as well as for doing their day to day activities. Regular work of the users is generally done via online. If people want to do money based transaction or sending important message to their neighbor, they use online network. This can be used and interfere by the attackers. Many attackers utilize the tools to disturb the authorized user job. Major attack utilized by the attacker to disturb the service get by the authorized users is denial of service. It becomes major kind of procedure done by the intruder to take control over system or to down the server. In this paper we analyze the DoS attack including its distributed version. Impacts of DoS are also discussed. Based on the survey and analysis we suggest the model for handling the DoS attack dynamically is also mentioned.","PeriodicalId":114624,"journal":{"name":"2023 2nd International Conference on Edge Computing and Applications (ICECAA)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125881837","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 : 2023-07-19DOI: 10.1109/ICECAA58104.2023.10212348
C. A. Kumar, Dr. Y. Chalapathi Rao, Vondala Naveen, Kumar, R. K. Senapati
Text-to-speech (TTS) synthesis involves creating machines that can talk. It is the process of automatically converting written text into spoken speech. TTS systems are important in human-computer interaction, and they have become more prominent with the advancement of computer technology. There has been a growing demand for speech applications in various areas such as telemedicine, robotics, Google Assistant, and other fields, especially during the Covid-19 era. It is clear that the existing speech technology is not good enough for users. Many speech applications still use pre-recorded prompts instead of text-to-speech because the quality of text-to-speech isn't good, and it can't pronounce non-standard words. In this study, a dependable and effective approach is proposed for creating speech synthesis voices that are specialized for specific domains and can handle non-standard words. This involves developing lexicon rules for non-standard words. By constructing databases that are closely related to the specific domain of the speech application, concatenative unit selection synthesis techniques, implemented using the festival/festvox tool, can be utilized to achieve high-quality, intelligible, and natural speech synthesis within the targeted domain. And the quality of proposed voice model is measured using Melcepstral distortion as an objective metric and Mean Opinion score as a subjective metric.
{"title":"Building a Light Weight Intelligible Text-to-Speech Voice Model for Indian Accent Telugu","authors":"C. A. Kumar, Dr. Y. Chalapathi Rao, Vondala Naveen, Kumar, R. K. Senapati","doi":"10.1109/ICECAA58104.2023.10212348","DOIUrl":"https://doi.org/10.1109/ICECAA58104.2023.10212348","url":null,"abstract":"Text-to-speech (TTS) synthesis involves creating machines that can talk. It is the process of automatically converting written text into spoken speech. TTS systems are important in human-computer interaction, and they have become more prominent with the advancement of computer technology. There has been a growing demand for speech applications in various areas such as telemedicine, robotics, Google Assistant, and other fields, especially during the Covid-19 era. It is clear that the existing speech technology is not good enough for users. Many speech applications still use pre-recorded prompts instead of text-to-speech because the quality of text-to-speech isn't good, and it can't pronounce non-standard words. In this study, a dependable and effective approach is proposed for creating speech synthesis voices that are specialized for specific domains and can handle non-standard words. This involves developing lexicon rules for non-standard words. By constructing databases that are closely related to the specific domain of the speech application, concatenative unit selection synthesis techniques, implemented using the festival/festvox tool, can be utilized to achieve high-quality, intelligible, and natural speech synthesis within the targeted domain. And the quality of proposed voice model is measured using Melcepstral distortion as an objective metric and Mean Opinion score as a subjective metric.","PeriodicalId":114624,"journal":{"name":"2023 2nd International Conference on Edge Computing and Applications (ICECAA)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125435021","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 : 2023-07-19DOI: 10.1109/ICECAA58104.2023.10212269
Xinli Lyu, Fangli Ying, Pintusorn Onpium
In the recent years, image diversity generation algorithms based on deep learning have gradually mined feature information that can describe the essential content of images through hierarchical learning, which has become a current research hotspot. In particular, the emergence of GANs has made the qualitative leap in the speed and quality of image data generation due to its good autonomous generation ability and ingenious weakly supervised learning mode. In this study, the focus is deep learning based scene style conversion algorithm of the AI digital host. The proposed model has 2 aspects: (1) Efficient face modelling. In this step, the face features are extracted and combined with deep neural networks to obtain the efficient representations, then, the rough stitching of point cloud data is applied to construct the different perspectives the faces. (2) Efficient style conversion algorithm. This study designs an improved GAN algorithm to create new image conversion scheme to complete the design of the AI host. In the experiment section, the mainstream evaluation methods are adopted to test the performance.
{"title":"Scene Style Conversion Algorithm of AI Digital Host: A Deep Learning Approach","authors":"Xinli Lyu, Fangli Ying, Pintusorn Onpium","doi":"10.1109/ICECAA58104.2023.10212269","DOIUrl":"https://doi.org/10.1109/ICECAA58104.2023.10212269","url":null,"abstract":"In the recent years, image diversity generation algorithms based on deep learning have gradually mined feature information that can describe the essential content of images through hierarchical learning, which has become a current research hotspot. In particular, the emergence of GANs has made the qualitative leap in the speed and quality of image data generation due to its good autonomous generation ability and ingenious weakly supervised learning mode. In this study, the focus is deep learning based scene style conversion algorithm of the AI digital host. The proposed model has 2 aspects: (1) Efficient face modelling. In this step, the face features are extracted and combined with deep neural networks to obtain the efficient representations, then, the rough stitching of point cloud data is applied to construct the different perspectives the faces. (2) Efficient style conversion algorithm. This study designs an improved GAN algorithm to create new image conversion scheme to complete the design of the AI host. In the experiment section, the mainstream evaluation methods are adopted to test the performance.","PeriodicalId":114624,"journal":{"name":"2023 2nd International Conference on Edge Computing and Applications (ICECAA)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125573006","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 : 2023-07-19DOI: 10.1109/ICECAA58104.2023.10212121
S. S, Shakunthala M, C. V. Josphine, M. T. Kingslin, S. Sivarajan, Chairma Lakshmi K R
Energy wastage, traffic jams, and traffic monitoring are the three main problems the world's population is confronting. The traditional roadway management method is ineffective at controlling the flow of traffic when one lane has significantly more traffic than the others. Additionally, the present street light controller uses lights that are not turned off in the morning even though it is sufficiently light outside and that are turned on in the late afternoon before even the sun sets, which causes some power loss. This proposed system will offer an advanced smart traffic management system and intelligent street lighting system to bring a standardized approach to solving this issue. The enhanced smart traffic control system is accomplished by modifying the signal timings based on the comparison of the densities on all roads, i.e., when a specific lane has much more density compared to the other lanes, that specific lane is modified to a green signal for a specified length of time. By computerizing solar-powered street lights with IoT, energy wastage in the traditional street lighting system is prevented. The street lights are activated whenever an item moves, or they remain off otherwise. This proposed method reduces traffic issues, and energy wastage and improves the traffic monitoring process also.
{"title":"IoT based on Smart Traffic Lights and Streetlight System","authors":"S. S, Shakunthala M, C. V. Josphine, M. T. Kingslin, S. Sivarajan, Chairma Lakshmi K R","doi":"10.1109/ICECAA58104.2023.10212121","DOIUrl":"https://doi.org/10.1109/ICECAA58104.2023.10212121","url":null,"abstract":"Energy wastage, traffic jams, and traffic monitoring are the three main problems the world's population is confronting. The traditional roadway management method is ineffective at controlling the flow of traffic when one lane has significantly more traffic than the others. Additionally, the present street light controller uses lights that are not turned off in the morning even though it is sufficiently light outside and that are turned on in the late afternoon before even the sun sets, which causes some power loss. This proposed system will offer an advanced smart traffic management system and intelligent street lighting system to bring a standardized approach to solving this issue. The enhanced smart traffic control system is accomplished by modifying the signal timings based on the comparison of the densities on all roads, i.e., when a specific lane has much more density compared to the other lanes, that specific lane is modified to a green signal for a specified length of time. By computerizing solar-powered street lights with IoT, energy wastage in the traditional street lighting system is prevented. The street lights are activated whenever an item moves, or they remain off otherwise. This proposed method reduces traffic issues, and energy wastage and improves the traffic monitoring process also.","PeriodicalId":114624,"journal":{"name":"2023 2nd International Conference on Edge Computing and Applications (ICECAA)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126602250","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 : 2023-07-19DOI: 10.1109/ICECAA58104.2023.10212415
K. Meenakshi, S. Sivasubramanian, B. Bharathi, S. John, Justin Thangaraj
Cloud computing is now widely employed across a variety of IT-related fields. It has become an essential technology to address infrastructure and data service requirements with minimal effort, low-cost, and with a high degree of scalability. Although the use of cloud computing has increased significantly, information security concerns have not yet been fully resolved. There are still certain information security concerns that must be resolved in order for cloud computing to advance. In terms of integrity, authenticity, and secrecy, in particular, Blockchain has developed into an essential security instrument. The application of Blockchain in cloud computing security is further examined in this article after a review of the different security concerns in blockchain and cloud computing. The main challenge in the cloud computing is cloud computing security. Because data in the cloud must be transported via the Internet, data security becomes a top priority. It is necessary to maintain the primary data protection methods, such as integrity, accountability, privacies, access controls, authentications, and authorizations. Blockchain is a technological advancement that enhances cloud computing. Blockchain resolves cloud computing security problems. This study seeks to analyze and contrast various cloud environment security challenges with blockchain security difficulties.
{"title":"Cloud Security Analysis using Blockchain Technology","authors":"K. Meenakshi, S. Sivasubramanian, B. Bharathi, S. John, Justin Thangaraj","doi":"10.1109/ICECAA58104.2023.10212415","DOIUrl":"https://doi.org/10.1109/ICECAA58104.2023.10212415","url":null,"abstract":"Cloud computing is now widely employed across a variety of IT-related fields. It has become an essential technology to address infrastructure and data service requirements with minimal effort, low-cost, and with a high degree of scalability. Although the use of cloud computing has increased significantly, information security concerns have not yet been fully resolved. There are still certain information security concerns that must be resolved in order for cloud computing to advance. In terms of integrity, authenticity, and secrecy, in particular, Blockchain has developed into an essential security instrument. The application of Blockchain in cloud computing security is further examined in this article after a review of the different security concerns in blockchain and cloud computing. The main challenge in the cloud computing is cloud computing security. Because data in the cloud must be transported via the Internet, data security becomes a top priority. It is necessary to maintain the primary data protection methods, such as integrity, accountability, privacies, access controls, authentications, and authorizations. Blockchain is a technological advancement that enhances cloud computing. Blockchain resolves cloud computing security problems. This study seeks to analyze and contrast various cloud environment security challenges with blockchain security difficulties.","PeriodicalId":114624,"journal":{"name":"2023 2nd International Conference on Edge Computing and Applications (ICECAA)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115032642","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 : 2023-07-19DOI: 10.1109/ICECAA58104.2023.10212386
K. M, N. M, Naresh Kumar A, Nitheesh Gokul M
The sewage system is the mechanism for efficiently dumping sewage from buildings and other locations. Sewage is collected using various components such as drains, manholes, pumping stations, and sanitary sewers. In cities, sewage is transported separately in sanitary sewers, while street runoff is managed by storm drains. In cities, the sewage system is accessible by a manhole. Due to high precipitation seasons, sewage is discharged straight into receiving waterways, posing a concern to public health. To decrease human labour and fatalities, a robot is deployed to remove blocked garbage from the canal and warns crew members if the container used for waste collection becomes totally full. To maintain the environment clean, the sewage canals should be monitored by performing live data comparison.
{"title":"Raspberry Pi Controller based Sewage Treating Mobile Robot using Machine Learning","authors":"K. M, N. M, Naresh Kumar A, Nitheesh Gokul M","doi":"10.1109/ICECAA58104.2023.10212386","DOIUrl":"https://doi.org/10.1109/ICECAA58104.2023.10212386","url":null,"abstract":"The sewage system is the mechanism for efficiently dumping sewage from buildings and other locations. Sewage is collected using various components such as drains, manholes, pumping stations, and sanitary sewers. In cities, sewage is transported separately in sanitary sewers, while street runoff is managed by storm drains. In cities, the sewage system is accessible by a manhole. Due to high precipitation seasons, sewage is discharged straight into receiving waterways, posing a concern to public health. To decrease human labour and fatalities, a robot is deployed to remove blocked garbage from the canal and warns crew members if the container used for waste collection becomes totally full. To maintain the environment clean, the sewage canals should be monitored by performing live data comparison.","PeriodicalId":114624,"journal":{"name":"2023 2nd International Conference on Edge Computing and Applications (ICECAA)","volume":"184 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122835118","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 : 2023-07-19DOI: 10.1109/ICECAA58104.2023.10212227
P. N. Bhushanam, S. S
Unconstraint video analytics are important in visual learning. Unconstrained videos contain complex content with various artifacts, variable lengths, and different operating environments. Human activity plays an important role in the video, abundant in archives and becoming more prevalent on the Internet. Various methods are employed for action recognition under constrained conditions, but huge attention to complex actions and realistic applications is substantially needed. Complex movements consist of sequences of simple movements that have long temporal structures. Complex actions in the same class exhibit large variations in class-interior appearance and behavior due to complex temporal structures, confusing backgrounds, changes in viewpoint, and changes in movement speed. Thus, feature representation and classification of complex motions are determined to be challenging in unconstrained video analysisdue to complex temporal structures. This article presents a comprehensive analysis to solve this problem using deep learningapproaches.
{"title":"A Comprehensive Analysis on Unconstraint Video Analysis Using Deep Learning Approaches","authors":"P. N. Bhushanam, S. S","doi":"10.1109/ICECAA58104.2023.10212227","DOIUrl":"https://doi.org/10.1109/ICECAA58104.2023.10212227","url":null,"abstract":"Unconstraint video analytics are important in visual learning. Unconstrained videos contain complex content with various artifacts, variable lengths, and different operating environments. Human activity plays an important role in the video, abundant in archives and becoming more prevalent on the Internet. Various methods are employed for action recognition under constrained conditions, but huge attention to complex actions and realistic applications is substantially needed. Complex movements consist of sequences of simple movements that have long temporal structures. Complex actions in the same class exhibit large variations in class-interior appearance and behavior due to complex temporal structures, confusing backgrounds, changes in viewpoint, and changes in movement speed. Thus, feature representation and classification of complex motions are determined to be challenging in unconstrained video analysisdue to complex temporal structures. This article presents a comprehensive analysis to solve this problem using deep learningapproaches.","PeriodicalId":114624,"journal":{"name":"2023 2nd International Conference on Edge Computing and Applications (ICECAA)","volume":"105 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114032554","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 : 2023-07-19DOI: 10.1109/ICECAA58104.2023.10212410
A. M., S. Suvarna, Rajgopal K T
The use of a signature as a biometric authentication method is crucial in banking and financial systems. The two forms of signatures are offline and online, with offline signatures being the more popular choice because of how straightforward and distinctive they are. The payer and payee must both sign digital cheques, just like paper ones. In this proposed approach we build a security system, verifying entrance applications and evaluating password substitutes. The proposed approach will create a system that can authenticate digital signatures online while clearing checks for banks and detecting any flaws, guaranteeing confidentiality and preventing fraud. The goal is to successfully validate the validity of online digital checks that contain signatures.
{"title":"Online Digital Cheque Signature Verification using Deep Learning Approach","authors":"A. M., S. Suvarna, Rajgopal K T","doi":"10.1109/ICECAA58104.2023.10212410","DOIUrl":"https://doi.org/10.1109/ICECAA58104.2023.10212410","url":null,"abstract":"The use of a signature as a biometric authentication method is crucial in banking and financial systems. The two forms of signatures are offline and online, with offline signatures being the more popular choice because of how straightforward and distinctive they are. The payer and payee must both sign digital cheques, just like paper ones. In this proposed approach we build a security system, verifying entrance applications and evaluating password substitutes. The proposed approach will create a system that can authenticate digital signatures online while clearing checks for banks and detecting any flaws, guaranteeing confidentiality and preventing fraud. The goal is to successfully validate the validity of online digital checks that contain signatures.","PeriodicalId":114624,"journal":{"name":"2023 2nd International Conference on Edge Computing and Applications (ICECAA)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122031626","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：