Pub Date : 2021-12-01DOI: 10.1109/ComPE53109.2021.9752237
Ambika Lakhera, Priyansh Jain, Ruchi Gajjar, Manish I. Patel
The coronavirus pandemic (COVID-19) has unfolded hastily throughout the entire world. This pandemic disease can spread through droplets and can be airborne. Hence, the use of face masks in public places is crucial to stop its spread. The present study aims to develop a system that can identify masked or non-masked faces; whether it is a normal mask, transparent mask, or a face alike mask. The face mask detection system is developed with the help of Convolutional Neural Networks (CNN). The model compression technique of Knowledge Distillation has been used to make the machine lesser computation and memory intensive so that it is simple to install the model on a few embedded gadgets and cell computing platforms. Using the model compression technique and GPU systems will help boom the calculation velocity of the model and drop the storage space required for calculations. The experimental outcomes show that the developed detector is capable to classify diverse types of masks. Also, it can classify video images in real-time. Using the Knowledge Distillation on the baseline model can improve the testing accuracy from 88.79% to 90.13%. The proposed unique system can be implemented to assist in the prevention of COVID-19 spread and detect various mask types.
{"title":"Face Mask Detection for Preventing the Spread of Covid-19 using Knowledge Distillation","authors":"Ambika Lakhera, Priyansh Jain, Ruchi Gajjar, Manish I. Patel","doi":"10.1109/ComPE53109.2021.9752237","DOIUrl":"https://doi.org/10.1109/ComPE53109.2021.9752237","url":null,"abstract":"The coronavirus pandemic (COVID-19) has unfolded hastily throughout the entire world. This pandemic disease can spread through droplets and can be airborne. Hence, the use of face masks in public places is crucial to stop its spread. The present study aims to develop a system that can identify masked or non-masked faces; whether it is a normal mask, transparent mask, or a face alike mask. The face mask detection system is developed with the help of Convolutional Neural Networks (CNN). The model compression technique of Knowledge Distillation has been used to make the machine lesser computation and memory intensive so that it is simple to install the model on a few embedded gadgets and cell computing platforms. Using the model compression technique and GPU systems will help boom the calculation velocity of the model and drop the storage space required for calculations. The experimental outcomes show that the developed detector is capable to classify diverse types of masks. Also, it can classify video images in real-time. Using the Knowledge Distillation on the baseline model can improve the testing accuracy from 88.79% to 90.13%. The proposed unique system can be implemented to assist in the prevention of COVID-19 spread and detect various mask types.","PeriodicalId":211704,"journal":{"name":"2021 International Conference on Computational Performance Evaluation (ComPE)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114868604","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 : 2021-12-01DOI: 10.1109/ComPE53109.2021.9752410
T. R. Ganesh Babu, K. Shenbagadevi, V. S. Shoba, S. Shrinidhi, J. Sabitha, U. Saravanakumar
The face is one of the simplest ways to distinguish one another's personal image. Face recognition is a personal identification system which uses a person's personal features to recognize the identity of the individual. Human facial identification is basically a two-phase procedure, including face detection, where the process is carried out very rapidly in people, whereas the second is the implementation of environments that classify the face as persons, when the eye is positioned within a short distance. Stage is then repeated and established to be one of the most researched biometric strategies and established by experts for facial expression recognition. In this study, we implemented the area of face detection and face recognition image processing MTCNN techniques while utilizing the VGG face model dataset. In this initiative, python framework is the program necessity.
{"title":"Image Processing Methods for Face Recognition using Machine Learning Techniques","authors":"T. R. Ganesh Babu, K. Shenbagadevi, V. S. Shoba, S. Shrinidhi, J. Sabitha, U. Saravanakumar","doi":"10.1109/ComPE53109.2021.9752410","DOIUrl":"https://doi.org/10.1109/ComPE53109.2021.9752410","url":null,"abstract":"The face is one of the simplest ways to distinguish one another's personal image. Face recognition is a personal identification system which uses a person's personal features to recognize the identity of the individual. Human facial identification is basically a two-phase procedure, including face detection, where the process is carried out very rapidly in people, whereas the second is the implementation of environments that classify the face as persons, when the eye is positioned within a short distance. Stage is then repeated and established to be one of the most researched biometric strategies and established by experts for facial expression recognition. In this study, we implemented the area of face detection and face recognition image processing MTCNN techniques while utilizing the VGG face model dataset. In this initiative, python framework is the program necessity.","PeriodicalId":211704,"journal":{"name":"2021 International Conference on Computational Performance Evaluation (ComPE)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115004237","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 : 2021-12-01DOI: 10.1109/ComPE53109.2021.9752053
J. Patni, Saurabh Agarwal, M. Kumar, Priyal Agarwal
Picture improvement has been discovered to be perhaps the main vision applications since it can upgrade the computerized pictures with the goal that the outcomes are more appropriate for show or further picture examination. To work on the nature of computerized pictures there are remarkable strategies that have been proposed. The goal is to manage picture handling and its major strides after that we had zeroed in on the diverse picture upgrade procedures. Since picture clearness is effectively influenced by lighting, climate, or gear that has been utilized to catch the picture. These conditions lead to loss of data. The principle motivation behind picture improvement is to bring out detail that is covered up in a picture or to expand contrast during a low difference picture. It gives countless decisions for working on the visual nature of pictures. Its article is to dissect the specific picture attributes for examination, end and further use.
{"title":"Image Quality Enhancement : A Linear Programming Approach","authors":"J. Patni, Saurabh Agarwal, M. Kumar, Priyal Agarwal","doi":"10.1109/ComPE53109.2021.9752053","DOIUrl":"https://doi.org/10.1109/ComPE53109.2021.9752053","url":null,"abstract":"Picture improvement has been discovered to be perhaps the main vision applications since it can upgrade the computerized pictures with the goal that the outcomes are more appropriate for show or further picture examination. To work on the nature of computerized pictures there are remarkable strategies that have been proposed. The goal is to manage picture handling and its major strides after that we had zeroed in on the diverse picture upgrade procedures. Since picture clearness is effectively influenced by lighting, climate, or gear that has been utilized to catch the picture. These conditions lead to loss of data. The principle motivation behind picture improvement is to bring out detail that is covered up in a picture or to expand contrast during a low difference picture. It gives countless decisions for working on the visual nature of pictures. Its article is to dissect the specific picture attributes for examination, end and further use.","PeriodicalId":211704,"journal":{"name":"2021 International Conference on Computational Performance Evaluation (ComPE)","volume":"139 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114656425","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}
A Wireless Sensor Network (WSN) consist of huge number of Sensor Nodes (SN’s) which are powered by irreplaceable battery source. So as to get best outcome of WSN, better network lifetime, equal load balancing among all SN’s and scalability of the network among the network are to be taken care off. For getting optimal lifetime from sensor network, clustering is one of the best techniques. Basically it is a process of placing similar kind of nodes together which works parallel with each other. One elected node from a cluster is responsible for transmitting data cluster’s aggregated data to the resource opulence node known as sink or Base Station (BS). An efficient scheme is required for formation of clusters & electing CH for network optimization. This work focuses on analysing network performance by varying number of clusters during the network operation. In the end it checks the lifetime (stable & Overall) of the network by varying probability factor for number of CH’s (Cluster Head) in network. Result analysis shows that lifetime of the network decreases gradually while increasing the probability of CH (Cluster head) however there is not as much difference observed in Stable Lifetime.
无线传感器网络(WSN)由大量的传感器节点(SN)组成，这些节点由不可替代的电池供电。为了获得最佳的WSN效果，需要注意更好的网络生存时间、各SN之间均衡的负载均衡以及网络之间的可扩展性。为了从传感器网络中获得最优的生存期，聚类是最好的技术之一。基本上，它是一个将相似类型的节点放置在一起的过程，这些节点彼此并行工作。从集群中选出的一个节点负责将数据集群的聚合数据传输到称为sink或Base Station (BS)的资源丰富节点。网络优化需要一个有效的簇的形成和CH的选择方案。这项工作的重点是在网络运行过程中通过不同数量的集群来分析网络性能。最后，它通过改变网络中CH(簇头)数量的概率因子来检查网络的生命周期(稳定和总体)。结果分析表明，随着簇头概率的增加，网络的生存期逐渐减小，而稳定生存期差异不大。
{"title":"Performance Analysis of WSN by varying number of clusters","authors":"Gaurav Bathla, Lokesh Pawar, Rohit Bajaj, Harjeet Kaur, Navjot Singh","doi":"10.1109/ComPE53109.2021.9752419","DOIUrl":"https://doi.org/10.1109/ComPE53109.2021.9752419","url":null,"abstract":"A Wireless Sensor Network (WSN) consist of huge number of Sensor Nodes (SN’s) which are powered by irreplaceable battery source. So as to get best outcome of WSN, better network lifetime, equal load balancing among all SN’s and scalability of the network among the network are to be taken care off. For getting optimal lifetime from sensor network, clustering is one of the best techniques. Basically it is a process of placing similar kind of nodes together which works parallel with each other. One elected node from a cluster is responsible for transmitting data cluster’s aggregated data to the resource opulence node known as sink or Base Station (BS). An efficient scheme is required for formation of clusters & electing CH for network optimization. This work focuses on analysing network performance by varying number of clusters during the network operation. In the end it checks the lifetime (stable & Overall) of the network by varying probability factor for number of CH’s (Cluster Head) in network. Result analysis shows that lifetime of the network decreases gradually while increasing the probability of CH (Cluster head) however there is not as much difference observed in Stable Lifetime.","PeriodicalId":211704,"journal":{"name":"2021 International Conference on Computational Performance Evaluation (ComPE)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134129034","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 : 2021-12-01DOI: 10.1109/ComPE53109.2021.9752185
Sarapu Likith, B. R. Reddy, K. Sripal Reddy
This paper's central theme is the use of YOLO and CNN to detect and classify pests. The quick extension of the human population opens on to a growth in food requirements. We lose a lot of crops owing to weather conditions and pests because of our country's illiteracy and hardship. Pests wreak havoc on a huge number of crops each and every year. As a result, in order to ensure excellent production in agricultural fields, the pest must be recognized and categorized. Early detection of pests in images is critical for pest reduction and elimination in the agricultural fields. As a result, classification of the Bug present in photographs has been difficult. The major goal of the proposed work is the classification of pests and implement pest- control strategies to safeguard crops from pests. We employ the YOLO (You Only Look Once) algorithm for pest detection and CNN for pest classification (Convolution Neural Network).
本文的中心主题是利用YOLO和CNN对害虫进行检测和分类。人口的迅速增长导致了粮食需求的增长。由于我们国家的文盲和艰苦，天气条件和害虫使我们损失了很多庄稼。害虫每年都对大量农作物造成严重破坏。因此，为了保证农业领域的优质生产，必须对害虫进行识别和分类。图像中害虫的早期发现对于减少和消除农田害虫至关重要。因此，对照片中的虫子进行分类是很困难的。提出的工作的主要目标是害虫的分类和实施害虫防治策略，以保护作物免受害虫的侵害。我们使用YOLO (You Only Look Once)算法进行害虫检测，使用CNN(卷积神经网络)进行害虫分类。
{"title":"A Smart System for Detection and Classification of Pests Using YOLO AND CNN Techniques","authors":"Sarapu Likith, B. R. Reddy, K. Sripal Reddy","doi":"10.1109/ComPE53109.2021.9752185","DOIUrl":"https://doi.org/10.1109/ComPE53109.2021.9752185","url":null,"abstract":"This paper's central theme is the use of YOLO and CNN to detect and classify pests. The quick extension of the human population opens on to a growth in food requirements. We lose a lot of crops owing to weather conditions and pests because of our country's illiteracy and hardship. Pests wreak havoc on a huge number of crops each and every year. As a result, in order to ensure excellent production in agricultural fields, the pest must be recognized and categorized. Early detection of pests in images is critical for pest reduction and elimination in the agricultural fields. As a result, classification of the Bug present in photographs has been difficult. The major goal of the proposed work is the classification of pests and implement pest- control strategies to safeguard crops from pests. We employ the YOLO (You Only Look Once) algorithm for pest detection and CNN for pest classification (Convolution Neural Network).","PeriodicalId":211704,"journal":{"name":"2021 International Conference on Computational Performance Evaluation (ComPE)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133576002","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 : 2021-12-01DOI: 10.1109/ComPE53109.2021.9752260
S. Sharma, D. Sharma, J. Verma
Prior and well-grounded produces evaluation is vital in quantifying a well and financial assessment at the field level for discovering agricultural commodity strategic action plans for import-export policies and increasing farmer incomes. Crop production projections are performed utilizing machine learning algorithms to estimate a higher crop yield, which is one of the most difficult challenges in the agriculture business. Because of the growing importance of agricultural yield prediction, this article takes an in-depth look at how Machine Learning (ML) approaches may be utilized to forecast crop production. The present state of agricultural yield worldwide is discussed first, followed by a brief introduction of extensively utilized features and forecasting procedures. Forecasting crop yields is a serious issue in agriculture, plus there is a large dataset that makes it arduous for farmers to select seeds and forecast yields. In today’s circumstances, since the extension in population, agricultural production must be raised simultaneously to fulfill people’s wants. This paper is a detailed study of various aspects of crop yielding in India using machine learning techniques and artificial intelligence.
{"title":"Study on Machine-Learning Algorithms in Crop Yield Predictions specific to Indian Agricultural Contexts","authors":"S. Sharma, D. Sharma, J. Verma","doi":"10.1109/ComPE53109.2021.9752260","DOIUrl":"https://doi.org/10.1109/ComPE53109.2021.9752260","url":null,"abstract":"Prior and well-grounded produces evaluation is vital in quantifying a well and financial assessment at the field level for discovering agricultural commodity strategic action plans for import-export policies and increasing farmer incomes. Crop production projections are performed utilizing machine learning algorithms to estimate a higher crop yield, which is one of the most difficult challenges in the agriculture business. Because of the growing importance of agricultural yield prediction, this article takes an in-depth look at how Machine Learning (ML) approaches may be utilized to forecast crop production. The present state of agricultural yield worldwide is discussed first, followed by a brief introduction of extensively utilized features and forecasting procedures. Forecasting crop yields is a serious issue in agriculture, plus there is a large dataset that makes it arduous for farmers to select seeds and forecast yields. In today’s circumstances, since the extension in population, agricultural production must be raised simultaneously to fulfill people’s wants. This paper is a detailed study of various aspects of crop yielding in India using machine learning techniques and artificial intelligence.","PeriodicalId":211704,"journal":{"name":"2021 International Conference on Computational Performance Evaluation (ComPE)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130557425","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 : 2021-12-01DOI: 10.1109/ComPE53109.2021.9752238
S. R., Sharath Cherian Thomas, Achint Mathews, Nishtha, Swathi C. Prabhu
Climate factors like temperature and humidity has a major role to play in the productivity in an office workspace. Similarly, products like medicine and food have to be manufactured and stored in optimum environmental conditions. Thus, monitoring the factors like temperature and humidity of different part of an office space or manufacturing unit is very important. Studying these factors over time will help in finding the optimum temperature for different parts of the office. Periodic storage of the data on secure remote location such as a SQL database on the cloud opens possibilities for future analysis. Presence of an app-based and web-based portal for monitoring the temperature and humidity data makes remote monitoring easy. Such a system can ensure the conditions are monitored at all times and alerts can be given in case of a sudden peak or drop. Security and confidentiality of the data is very important and thus limiting the access to the database and even the web and app-based portals can ensure the data is visible only to authorized personal.
{"title":"Secure Cloud-based Remote Monitoring of Environmental Factors using Mobile and Web Apps for Industry Automation","authors":"S. R., Sharath Cherian Thomas, Achint Mathews, Nishtha, Swathi C. Prabhu","doi":"10.1109/ComPE53109.2021.9752238","DOIUrl":"https://doi.org/10.1109/ComPE53109.2021.9752238","url":null,"abstract":"Climate factors like temperature and humidity has a major role to play in the productivity in an office workspace. Similarly, products like medicine and food have to be manufactured and stored in optimum environmental conditions. Thus, monitoring the factors like temperature and humidity of different part of an office space or manufacturing unit is very important. Studying these factors over time will help in finding the optimum temperature for different parts of the office. Periodic storage of the data on secure remote location such as a SQL database on the cloud opens possibilities for future analysis. Presence of an app-based and web-based portal for monitoring the temperature and humidity data makes remote monitoring easy. Such a system can ensure the conditions are monitored at all times and alerts can be given in case of a sudden peak or drop. Security and confidentiality of the data is very important and thus limiting the access to the database and even the web and app-based portals can ensure the data is visible only to authorized personal.","PeriodicalId":211704,"journal":{"name":"2021 International Conference on Computational Performance Evaluation (ComPE)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125849886","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 : 2021-12-01DOI: 10.1109/ComPE53109.2021.9751909
Y. Kumar, Lokesh Chouhan, Basant Subba
Information security has become one of the significant concerns with the advancement of technology and digital assistance. An Intrusion Detection System(IDS) plays a substantial role in guarding the systems from security threats. However, existing IDS frameworks have faced challenges such as high false alarm rate, low detection rate, raw and huge dataset handling, etc. The Deep Learning techniques has grown as a reliable methodology to address such issues. This paper presents a taxonomy of anomaly based IDS frameworks. It also includes a detailed analysis of Deep Learning algorithms used in IDS frameworks and their comparison based on different characteristics. In addition, this study indicates critical challenges of the anomaly based IDS frameworks followed by possible future directions to improve their performances.
{"title":"Deep Learning Techniques for Anomaly based Intrusion Detection System: A Survey","authors":"Y. Kumar, Lokesh Chouhan, Basant Subba","doi":"10.1109/ComPE53109.2021.9751909","DOIUrl":"https://doi.org/10.1109/ComPE53109.2021.9751909","url":null,"abstract":"Information security has become one of the significant concerns with the advancement of technology and digital assistance. An Intrusion Detection System(IDS) plays a substantial role in guarding the systems from security threats. However, existing IDS frameworks have faced challenges such as high false alarm rate, low detection rate, raw and huge dataset handling, etc. The Deep Learning techniques has grown as a reliable methodology to address such issues. This paper presents a taxonomy of anomaly based IDS frameworks. It also includes a detailed analysis of Deep Learning algorithms used in IDS frameworks and their comparison based on different characteristics. In addition, this study indicates critical challenges of the anomaly based IDS frameworks followed by possible future directions to improve their performances.","PeriodicalId":211704,"journal":{"name":"2021 International Conference on Computational Performance Evaluation (ComPE)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122302004","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 : 2021-12-01DOI: 10.1109/ComPE53109.2021.9752254
Chaitanya Kaul, Neeraj Sharma
This research article is based on the ensemble approach of different supervised machine learning algorithms to identify the early stages of breast cancer problems. The World Health Organization (WHO) approved that existence of the breast tumor is high for the women in developing countries and it is one of the significant research issues in current scenario in the real world. In this research article researcher used the 30 features to extract and predict accurate prediction on breast cancer using ensemble approach of supervised machine learning algorithms. It is a great challenge in designing a machine learning model to evaluate the performance of the classification of breast tumor. Implementing an efficient classification methodology will support in resolving the complications in analyzing breast cancer. This proposed model employs four machine learning (ML) algorithms Decision tree classifiers, Random Forest KNN, and support vector machine (SVM) and found support vector machine (SVM) which given the high accuracy of 0.976688 among them for the categorization of breast tumor in women. This classification includes the two levels of disease as benign or malignant. The researcher also used the other parameters and evaluated this predictive model using Precision, Recall and F1-Score. The data analysis report is proved that this predictive model is having 98% accuracy level to predict the cancer at early stages in women.
{"title":"High Accuracy Predictive Model on Breast Cancer Using Ensemble Approach of Supervised Machine Learning Algorithms","authors":"Chaitanya Kaul, Neeraj Sharma","doi":"10.1109/ComPE53109.2021.9752254","DOIUrl":"https://doi.org/10.1109/ComPE53109.2021.9752254","url":null,"abstract":"This research article is based on the ensemble approach of different supervised machine learning algorithms to identify the early stages of breast cancer problems. The World Health Organization (WHO) approved that existence of the breast tumor is high for the women in developing countries and it is one of the significant research issues in current scenario in the real world. In this research article researcher used the 30 features to extract and predict accurate prediction on breast cancer using ensemble approach of supervised machine learning algorithms. It is a great challenge in designing a machine learning model to evaluate the performance of the classification of breast tumor. Implementing an efficient classification methodology will support in resolving the complications in analyzing breast cancer. This proposed model employs four machine learning (ML) algorithms Decision tree classifiers, Random Forest KNN, and support vector machine (SVM) and found support vector machine (SVM) which given the high accuracy of 0.976688 among them for the categorization of breast tumor in women. This classification includes the two levels of disease as benign or malignant. The researcher also used the other parameters and evaluated this predictive model using Precision, Recall and F1-Score. The data analysis report is proved that this predictive model is having 98% accuracy level to predict the cancer at early stages in women.","PeriodicalId":211704,"journal":{"name":"2021 International Conference on Computational Performance Evaluation (ComPE)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122347182","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 : 2021-12-01DOI: 10.1109/ComPE53109.2021.9752105
Sneh Lata Aswal, A. Negi, A. Saxena
Unless a powerful algorithm protects data transmitted over a wireless network, it is public. It is still vulnerable to attacks despite several solutions. The paper provides a robust M-encrypt technique for securing a wireless network. M-encrypt is the best option photos sending images over a wireless network. The image is partitioned into four sections in the first step of the method. Using L-Fractal, the proposed approach encrypts each of the four sections separately. The segments are then encrypted and transformed into linear arrays before being sent across the network. To test SITF, an IEEE 802.11g and 802.16 simulation was done. Kali Linux and hacking software running on the Redmi Note 7 were used to test the susceptibility to attacks. There was no evidence of vulnerability to the attacks in the packets. Four parameters were used to assess the quality of the encrypted images. The results show that the image quality transmitted is excellent. Finally, we calculated the encryption, decryption, and overall packet transmission times on Wi-Fi and Wi-Max networks.
{"title":"SITF: an algorithm for secured image transmission using Fractals","authors":"Sneh Lata Aswal, A. Negi, A. Saxena","doi":"10.1109/ComPE53109.2021.9752105","DOIUrl":"https://doi.org/10.1109/ComPE53109.2021.9752105","url":null,"abstract":"Unless a powerful algorithm protects data transmitted over a wireless network, it is public. It is still vulnerable to attacks despite several solutions. The paper provides a robust M-encrypt technique for securing a wireless network. M-encrypt is the best option photos sending images over a wireless network. The image is partitioned into four sections in the first step of the method. Using L-Fractal, the proposed approach encrypts each of the four sections separately. The segments are then encrypted and transformed into linear arrays before being sent across the network. To test SITF, an IEEE 802.11g and 802.16 simulation was done. Kali Linux and hacking software running on the Redmi Note 7 were used to test the susceptibility to attacks. There was no evidence of vulnerability to the attacks in the packets. Four parameters were used to assess the quality of the encrypted images. The results show that the image quality transmitted is excellent. Finally, we calculated the encryption, decryption, and overall packet transmission times on Wi-Fi and Wi-Max networks.","PeriodicalId":211704,"journal":{"name":"2021 International Conference on Computational Performance Evaluation (ComPE)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125563841","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：