Pub Date : 2023-04-05DOI: 10.1109/PCEMS58491.2023.10136094
Ankush D. Sawarkar, D. Shrimankar, S. Sahu, Lal Singh, N. Bokde, Manish Kumar
Bamboo is a grass that grows quickly (lm/days) and is very important to India’s social, economic, and environmental growth. India has a big market for bamboo because of the gap between its imports (3,306 thousand ${$}$) and exports (607 thousand ${$}$) of its products. Clustering bamboo for business will be important in the near future since it can be used in over 1,000 ways and is worth USD 2.969 billion in trade. Machine learning (ML) clustering models play a vital role in achieving this task for the commercial clustering of bamboo. In this research, we have presented details of twenty commercial species of bamboo in India has been identified, and data on the 2000 bamboo species have been collected. Although many algorithms have been introduced for clustering in recent years, not on bamboo, especially on morphological data. The target of this paper is to cluster these different bamboo species based on its commercial value using various ML algorithms such as K-means, Gaussian Mixture Models (GMM), and Balance Iterative Reducing and Clustering using Hierarchies (BIRCH). Clustering comes under unsupervised learning; there is no direct accuracy count. To evaluate the performance of clustering and determine which algorithm is best for clustering, we have used other indirect performance measures such as Silhouette Score, Calinski-Harabasz Index (CHI), and the Davies-Bouldin Index (DBI). K-mean shows the highest performance measures among all the other clustering ML models, with achieve a silhouette score of 0.5126, CHI of 17315 and DBI of 0.6633.
{"title":"Commercial Clustering of Indian Bamboo Species Using Machine Learning Techniques","authors":"Ankush D. Sawarkar, D. Shrimankar, S. Sahu, Lal Singh, N. Bokde, Manish Kumar","doi":"10.1109/PCEMS58491.2023.10136094","DOIUrl":"https://doi.org/10.1109/PCEMS58491.2023.10136094","url":null,"abstract":"Bamboo is a grass that grows quickly (lm/days) and is very important to India’s social, economic, and environmental growth. India has a big market for bamboo because of the gap between its imports (3,306 thousand ${$}$) and exports (607 thousand ${$}$) of its products. Clustering bamboo for business will be important in the near future since it can be used in over 1,000 ways and is worth USD 2.969 billion in trade. Machine learning (ML) clustering models play a vital role in achieving this task for the commercial clustering of bamboo. In this research, we have presented details of twenty commercial species of bamboo in India has been identified, and data on the 2000 bamboo species have been collected. Although many algorithms have been introduced for clustering in recent years, not on bamboo, especially on morphological data. The target of this paper is to cluster these different bamboo species based on its commercial value using various ML algorithms such as K-means, Gaussian Mixture Models (GMM), and Balance Iterative Reducing and Clustering using Hierarchies (BIRCH). Clustering comes under unsupervised learning; there is no direct accuracy count. To evaluate the performance of clustering and determine which algorithm is best for clustering, we have used other indirect performance measures such as Silhouette Score, Calinski-Harabasz Index (CHI), and the Davies-Bouldin Index (DBI). K-mean shows the highest performance measures among all the other clustering ML models, with achieve a silhouette score of 0.5126, CHI of 17315 and DBI of 0.6633.","PeriodicalId":330870,"journal":{"name":"2023 2nd International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125746588","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 : 2023-04-05DOI: 10.1109/PCEMS58491.2023.10136037
P Pedda Sadhu Naik, M. Padukudru, Jeny Rajan
Asthma is a chronic respiratory disorder characterised by airway inflammation and constriction, leading to difficulty in breathing and recurrent attacks of wheezing, coughing, and shortness of breath. In asthma, various cytokines, including interleukins (IL-4, IL-5, and IL-13) and tumor necrosis factoralpha (TNF-alpha), have been found to be increased in the airways of individuals. These cytokines are involved in the recruitment and activation of immune cells, such as eosinophils and T-lymphocytes, which contribute to the inflammation and airway hyperresponsiveness. Dysregulation of cytokine production and signaling has been implicated in the pathogenesis of asthma and may be targeted by therapies to alleviate symptoms and improve outcomes in individuals with this disease. We propose a predictive binary and multi-class machine learning model analysis that efficiently classify the asthma and healthy control patients by detecting cytokines in bronchoalveolar lavage (BAL) fluid which achieved better F1-score than existing approaches.
{"title":"Assessment of Asthma BAL Cytokines using Machine Learning Techniques","authors":"P Pedda Sadhu Naik, M. Padukudru, Jeny Rajan","doi":"10.1109/PCEMS58491.2023.10136037","DOIUrl":"https://doi.org/10.1109/PCEMS58491.2023.10136037","url":null,"abstract":"Asthma is a chronic respiratory disorder characterised by airway inflammation and constriction, leading to difficulty in breathing and recurrent attacks of wheezing, coughing, and shortness of breath. In asthma, various cytokines, including interleukins (IL-4, IL-5, and IL-13) and tumor necrosis factoralpha (TNF-alpha), have been found to be increased in the airways of individuals. These cytokines are involved in the recruitment and activation of immune cells, such as eosinophils and T-lymphocytes, which contribute to the inflammation and airway hyperresponsiveness. Dysregulation of cytokine production and signaling has been implicated in the pathogenesis of asthma and may be targeted by therapies to alleviate symptoms and improve outcomes in individuals with this disease. We propose a predictive binary and multi-class machine learning model analysis that efficiently classify the asthma and healthy control patients by detecting cytokines in bronchoalveolar lavage (BAL) fluid which achieved better F1-score than existing approaches.","PeriodicalId":330870,"journal":{"name":"2023 2nd International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129256903","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 : 2023-04-05DOI: 10.1109/PCEMS58491.2023.10136059
Mani Deepika Kalava, Likhita Kadiyala, S. Kommineni, Ramana Reddy Atla, V. S. G. Thadikemalla
Humans need to communicate in order to emphasize their thoughts and feelings, collaborate with others, and elevate society as a whole. A hearing-impaired person uses sign language to communicate and this language develops naturally within them. However, the non-signer community doesn’t somehow acknowledge it and hence this remains as a significant barrier that negatively impacts living quality. To bridge the gap, effective sign-language recognition (SLR) system is required and is still an unsolved research issue. New technologies have been developed for the past few years to counter the problem of recognition and were mainly developed using sensors and hardware equipment based on gloves. As contrary to earlier technology, this review presents that, there is no need for expensive and complex hardware in order to recognize sign language, only a modern device with a camera is sufficient. This is accomplished by using Google’s MediaPipe framework and machine learning techniques. In this paper, we had presented various techniques developed for Indian sign language and our future goal is to deliver a reliable SLR system with computer vision and AI due to its self-learning capabilities and increased accuracy.
{"title":"Indian Sign Language Recognition Using Classical And Machine Learning Techniques – A Review","authors":"Mani Deepika Kalava, Likhita Kadiyala, S. Kommineni, Ramana Reddy Atla, V. S. G. Thadikemalla","doi":"10.1109/PCEMS58491.2023.10136059","DOIUrl":"https://doi.org/10.1109/PCEMS58491.2023.10136059","url":null,"abstract":"Humans need to communicate in order to emphasize their thoughts and feelings, collaborate with others, and elevate society as a whole. A hearing-impaired person uses sign language to communicate and this language develops naturally within them. However, the non-signer community doesn’t somehow acknowledge it and hence this remains as a significant barrier that negatively impacts living quality. To bridge the gap, effective sign-language recognition (SLR) system is required and is still an unsolved research issue. New technologies have been developed for the past few years to counter the problem of recognition and were mainly developed using sensors and hardware equipment based on gloves. As contrary to earlier technology, this review presents that, there is no need for expensive and complex hardware in order to recognize sign language, only a modern device with a camera is sufficient. This is accomplished by using Google’s MediaPipe framework and machine learning techniques. In this paper, we had presented various techniques developed for Indian sign language and our future goal is to deliver a reliable SLR system with computer vision and AI due to its self-learning capabilities and increased accuracy.","PeriodicalId":330870,"journal":{"name":"2023 2nd International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130061225","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 : 2023-04-05DOI: 10.1109/PCEMS58491.2023.10136111
Manasa Korumilli, Koppula Sai Lasya, Naveen Cheggoju, V. Kamble, V. Satpute
Unintentional falls of people, when left without serve may cause severe injuries and in extreme cases, they may even lead to loss of lives. In order to provide timely medication, detection of fall events when occurred is necessary.Any action can be considered as the specific motion of various bone key points. So, in our work, we considered bone key points as feature extractors. The MediaPipe framework developed by Google is used to detect the bone key features and its coordinates on the human skeleton. The data obtained is then normalized with respect to the boundary box drawn around humans. Machine learning classifiers, RF, SVM and Deep Learning model, DNN are then used individually to recognise and classify the action into fall or non-fall events. NTU-RGB+D dataset is used in our work. Real time detection using a webcam is also tested. The accuracy achieved by DNN model is 97.63% and that of SVM and RF classifiers is 83.3% and 99.34% respectively. Thus, the highest accuracy is achieved by RF classifier which is 99.34%.
{"title":"Human Fall Detection using Skeleton Features","authors":"Manasa Korumilli, Koppula Sai Lasya, Naveen Cheggoju, V. Kamble, V. Satpute","doi":"10.1109/PCEMS58491.2023.10136111","DOIUrl":"https://doi.org/10.1109/PCEMS58491.2023.10136111","url":null,"abstract":"Unintentional falls of people, when left without serve may cause severe injuries and in extreme cases, they may even lead to loss of lives. In order to provide timely medication, detection of fall events when occurred is necessary.Any action can be considered as the specific motion of various bone key points. So, in our work, we considered bone key points as feature extractors. The MediaPipe framework developed by Google is used to detect the bone key features and its coordinates on the human skeleton. The data obtained is then normalized with respect to the boundary box drawn around humans. Machine learning classifiers, RF, SVM and Deep Learning model, DNN are then used individually to recognise and classify the action into fall or non-fall events. NTU-RGB+D dataset is used in our work. Real time detection using a webcam is also tested. The accuracy achieved by DNN model is 97.63% and that of SVM and RF classifiers is 83.3% and 99.34% respectively. Thus, the highest accuracy is achieved by RF classifier which is 99.34%.","PeriodicalId":330870,"journal":{"name":"2023 2nd International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131308016","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 : 2023-04-05DOI: 10.1109/PCEMS58491.2023.10136065
N. Sakthipriya, V. Govindasamy, V. Akila
Internet of Things plays a vital role in our everyday lives in terms of economic, social, and commercial aspects. The widespread use of IoT devices has made them a prime target for cyber-attacks. IoT botnet attacks usually have a greater sensitivity to the consequences that might result from launching other attacks such as DDoS attacks and dissemination of sensitive information. For botnet detection in the IoT environment, deep learning mechanisms have emerged. But processing high-dimensional data is difficult, and it adversely affects DL-based botnet detection systems. Various dimensionality reduction methods have been proposed by researchers to address this concern. The purpose of this study is to examine and compare current mainstream dimensionality reduction methods. This will enable us to understand how reducing the number of features may lead to higher classification accuracy. Extensive tests are conducted on the NBaIoT dataset to verify the viability of PCA and auto encoder dimensionality reduction strategies. Results show that Auto encoder algorithm outperform PCA dimensionality reduction methods by the accuracy of 95.02%.
{"title":"A Comparative Analysis of various Dimensionality Reduction Techniques on N-BaIoT Dataset for IoT Botnet Detection","authors":"N. Sakthipriya, V. Govindasamy, V. Akila","doi":"10.1109/PCEMS58491.2023.10136065","DOIUrl":"https://doi.org/10.1109/PCEMS58491.2023.10136065","url":null,"abstract":"Internet of Things plays a vital role in our everyday lives in terms of economic, social, and commercial aspects. The widespread use of IoT devices has made them a prime target for cyber-attacks. IoT botnet attacks usually have a greater sensitivity to the consequences that might result from launching other attacks such as DDoS attacks and dissemination of sensitive information. For botnet detection in the IoT environment, deep learning mechanisms have emerged. But processing high-dimensional data is difficult, and it adversely affects DL-based botnet detection systems. Various dimensionality reduction methods have been proposed by researchers to address this concern. The purpose of this study is to examine and compare current mainstream dimensionality reduction methods. This will enable us to understand how reducing the number of features may lead to higher classification accuracy. Extensive tests are conducted on the NBaIoT dataset to verify the viability of PCA and auto encoder dimensionality reduction strategies. Results show that Auto encoder algorithm outperform PCA dimensionality reduction methods by the accuracy of 95.02%.","PeriodicalId":330870,"journal":{"name":"2023 2nd International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132519857","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 : 2023-04-05DOI: 10.1109/PCEMS58491.2023.10136071
Rishi Rakesh Shrivastava, G. Deepak
There is a need for Ontology modelling and automatic generation of Ontologies in order to assimilate knowledge World Wide Web knowledge as a strategic model. Ontologies are the best knowledge descriptor model as they have some amount of human cognition associated with them because either humans are major contributors when they are generated manually or are indirect contributors when they are semi automatically generated. Internet of Things is a domain which has strategically evolved in the last few years, and there is a need for integrating several facets of Internet of Things Ontology. In this paper a strategic scheme for Internet of Things Ontology integration for Internet of Things domain with different perspective are proposed wherein the dataset are subjected to tag generation which is further classified using the AdaBoost classifier which are aligned with the random core classes of the existing variational Ontologies in the Internet of Things domain using Shannon’s entropy and the pointwise mutual information measure with differential step deviation measure. Which yields average precision and recall of 96.83 and 97.95 respectively.
{"title":"AIOIML: Automatic Integration of Ontologies for IoT Domain Using Hybridized Machine Learning Techniques","authors":"Rishi Rakesh Shrivastava, G. Deepak","doi":"10.1109/PCEMS58491.2023.10136071","DOIUrl":"https://doi.org/10.1109/PCEMS58491.2023.10136071","url":null,"abstract":"There is a need for Ontology modelling and automatic generation of Ontologies in order to assimilate knowledge World Wide Web knowledge as a strategic model. Ontologies are the best knowledge descriptor model as they have some amount of human cognition associated with them because either humans are major contributors when they are generated manually or are indirect contributors when they are semi automatically generated. Internet of Things is a domain which has strategically evolved in the last few years, and there is a need for integrating several facets of Internet of Things Ontology. In this paper a strategic scheme for Internet of Things Ontology integration for Internet of Things domain with different perspective are proposed wherein the dataset are subjected to tag generation which is further classified using the AdaBoost classifier which are aligned with the random core classes of the existing variational Ontologies in the Internet of Things domain using Shannon’s entropy and the pointwise mutual information measure with differential step deviation measure. Which yields average precision and recall of 96.83 and 97.95 respectively.","PeriodicalId":330870,"journal":{"name":"2023 2nd International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124569097","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 : 2023-04-05DOI: 10.1109/PCEMS58491.2023.10136101
Poongkuyil Muse, M. S., Hamil Stanly
Efficient log analysis involves collecting, evaluating, and managing raw data from computer-generated records. As security vulnerabilities increase, the analysis of logs has become vital and crucial in multidisciplinary domains. Maintaining and analyzing the log is a pivotal part of every organization as tons of logs are generated every millisecond. However, anomaly detection and log parsing addressed so far, rely on a time-consuming training algorithm based on a Machine Learning framework. The proposed method detects anomalies from real-time data generated from the data centre without the need for a training algorithm. Detection and visualization of malicious activities are done by Elasticsearch, Logstash, and Kibana (ELK) framework. The process of shipping, parsing, indexing, and anomaly detection is carried out using an unsupervised machine learning algorithm which gives a clear inference to detect bots and perform unique log session classification. A real-time Apache HTTP Server log is accessed and anomalous behavior is identified based on the incoming requests. Experiments on real-time data show that 13.76% of anomalies are detected on per weekly basis.
{"title":"Online Log Analysis(OLA) for Malicious User Activities","authors":"Poongkuyil Muse, M. S., Hamil Stanly","doi":"10.1109/PCEMS58491.2023.10136101","DOIUrl":"https://doi.org/10.1109/PCEMS58491.2023.10136101","url":null,"abstract":"Efficient log analysis involves collecting, evaluating, and managing raw data from computer-generated records. As security vulnerabilities increase, the analysis of logs has become vital and crucial in multidisciplinary domains. Maintaining and analyzing the log is a pivotal part of every organization as tons of logs are generated every millisecond. However, anomaly detection and log parsing addressed so far, rely on a time-consuming training algorithm based on a Machine Learning framework. The proposed method detects anomalies from real-time data generated from the data centre without the need for a training algorithm. Detection and visualization of malicious activities are done by Elasticsearch, Logstash, and Kibana (ELK) framework. The process of shipping, parsing, indexing, and anomaly detection is carried out using an unsupervised machine learning algorithm which gives a clear inference to detect bots and perform unique log session classification. A real-time Apache HTTP Server log is accessed and anomalous behavior is identified based on the incoming requests. Experiments on real-time data show that 13.76% of anomalies are detected on per weekly basis.","PeriodicalId":330870,"journal":{"name":"2023 2nd International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128439337","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 : 2023-04-05DOI: 10.1109/PCEMS58491.2023.10136095
Biby Joseph, Gopireddy Chaithanyakumar Reddy, R. Kavitha
Address decoders play a vital role in Static Random-Access Memory (SRAM) memory array architecture to fetch the data in less span of time. As compared to other memory devices, SRAM based Artificial Intelligence (AI) accelerator possess high speed in which memory array address decoder plays a vital role. As a result, the address decoder is the key element for SRAM performance. In this paper, an energy efficient address decoder with low power dissipation is proposed, which can be used for SRAM based AI accelerator. Major part of power consumption of memory chip depends on address decoders. As we go down from higher technology nodes to lower technology nodes, leakage power increases which results in total power consumption. Source biasing technique is used to reduce static power consumption. This paper compares the proposed Address decoder in UMC 65nm technology with existing architectures in terms of power, delay and energy. This decoder design has an improvement of 85.8% average power and 87.46% energy as compared with existing conventional 6-64 decoder circuit using pre-decoding methodology.
{"title":"Energy Efficient Memory Decoder for SRAM Based AI Accelerator","authors":"Biby Joseph, Gopireddy Chaithanyakumar Reddy, R. Kavitha","doi":"10.1109/PCEMS58491.2023.10136095","DOIUrl":"https://doi.org/10.1109/PCEMS58491.2023.10136095","url":null,"abstract":"Address decoders play a vital role in Static Random-Access Memory (SRAM) memory array architecture to fetch the data in less span of time. As compared to other memory devices, SRAM based Artificial Intelligence (AI) accelerator possess high speed in which memory array address decoder plays a vital role. As a result, the address decoder is the key element for SRAM performance. In this paper, an energy efficient address decoder with low power dissipation is proposed, which can be used for SRAM based AI accelerator. Major part of power consumption of memory chip depends on address decoders. As we go down from higher technology nodes to lower technology nodes, leakage power increases which results in total power consumption. Source biasing technique is used to reduce static power consumption. This paper compares the proposed Address decoder in UMC 65nm technology with existing architectures in terms of power, delay and energy. This decoder design has an improvement of 85.8% average power and 87.46% energy as compared with existing conventional 6-64 decoder circuit using pre-decoding methodology.","PeriodicalId":330870,"journal":{"name":"2023 2nd International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS)","volume":"449 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115931360","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 : 2023-04-05DOI: 10.1109/PCEMS58491.2023.10136117
Omkar Gupta, Irika Nikhita Kalapala, K. Bhurchandi
Images are an integral and indispensable aspect of various disciplines, such as medicine, surveillance, and the entertainment industry. However, the quality of images can be severely compromised by the presence of sensor noise, quantization errors, or transmission errors. This research proposes a novel approach that combines wavelet thresholding and BM3D (Block-Matching and 3D Filtering) techniques for effective image denoising.The efficacy of the methodologies is evaluated and compared to cutting-edge denoising techniques, demonstrating superior performance in both quantitative metrics and visual quality. Furthermore, the study delves into the intricate mechanisms underlying the denoising process and the impact of various parameters on the denoising performance, contributing significantly to the field of image denoising.
{"title":"Enhancing Image Denoising Performance through a Family of Algorithms","authors":"Omkar Gupta, Irika Nikhita Kalapala, K. Bhurchandi","doi":"10.1109/PCEMS58491.2023.10136117","DOIUrl":"https://doi.org/10.1109/PCEMS58491.2023.10136117","url":null,"abstract":"Images are an integral and indispensable aspect of various disciplines, such as medicine, surveillance, and the entertainment industry. However, the quality of images can be severely compromised by the presence of sensor noise, quantization errors, or transmission errors. This research proposes a novel approach that combines wavelet thresholding and BM3D (Block-Matching and 3D Filtering) techniques for effective image denoising.The efficacy of the methodologies is evaluated and compared to cutting-edge denoising techniques, demonstrating superior performance in both quantitative metrics and visual quality. Furthermore, the study delves into the intricate mechanisms underlying the denoising process and the impact of various parameters on the denoising performance, contributing significantly to the field of image denoising.","PeriodicalId":330870,"journal":{"name":"2023 2nd International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116415609","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 : 2023-04-05DOI: 10.1109/PCEMS58491.2023.10136033
Bhavesh Dhake, C. Shetye, Pratik Borhade, Devish Gawas, Amit K. Nerurkar
Cyber hazards have emerged as a significant cause of worry for society. Firms are beginning to invest heavily in developing Cyber Threat Intelligence in recent years in order to combat the rising threat of cyber-attacks (CTI). Predominantly, many businesses gathered and analyzed data from internal log files, leading in reactive CTI, which is essentially a data-driven process. The internet hacker community may provide significant proactive CTI value by alerting enterprises about risks that they were previously unaware of. Forums, more than any other platform, give the most metadata, data persistence, and tens of thousands of publicly available Tools, Techniques, and Procedures. Anticrawling techniques, including as authentication, throttling, and obfuscation, are commonly employed in forums. This study intends to create a unique web crawler, as well as use machine learning and deep learning approaches with neural networks to automatically categorize hacker forum data into predetermined categories and anticipate probable future cyber risks for proactive and timely CTI.
{"title":"Stratification of Hacker Forums and Predicting Cyber Assaults for Proactive Cyber Threat Intelligence","authors":"Bhavesh Dhake, C. Shetye, Pratik Borhade, Devish Gawas, Amit K. Nerurkar","doi":"10.1109/PCEMS58491.2023.10136033","DOIUrl":"https://doi.org/10.1109/PCEMS58491.2023.10136033","url":null,"abstract":"Cyber hazards have emerged as a significant cause of worry for society. Firms are beginning to invest heavily in developing Cyber Threat Intelligence in recent years in order to combat the rising threat of cyber-attacks (CTI). Predominantly, many businesses gathered and analyzed data from internal log files, leading in reactive CTI, which is essentially a data-driven process. The internet hacker community may provide significant proactive CTI value by alerting enterprises about risks that they were previously unaware of. Forums, more than any other platform, give the most metadata, data persistence, and tens of thousands of publicly available Tools, Techniques, and Procedures. Anticrawling techniques, including as authentication, throttling, and obfuscation, are commonly employed in forums. This study intends to create a unique web crawler, as well as use machine learning and deep learning approaches with neural networks to automatically categorize hacker forum data into predetermined categories and anticipate probable future cyber risks for proactive and timely CTI.","PeriodicalId":330870,"journal":{"name":"2023 2nd International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115589556","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：