Pub Date : 2023-04-05DOI: 10.1109/ICNWC57852.2023.10127521
S. Malaiarasan, R. Ravi, D.R. Maheswari, C. Rubavathi, M. Ramnath, V. Hemamalini
Most people’s first experience with cancer will be with skin cancer, which is also the most prevalent and potentially fatal kind. Determining a skin cancer diagnosis also requires the use of information technologies. This highlights the need of developing and deploying highly effective deep-learning methods for the early and accurate diagnosis and detection of skin cancer. Deep Convolution Neural Network (DCNN) is proposed for automated skin cancer detection in this study. This study’s unique contribution is the use of a deep convolution neural network containing 12 nested processing layers to improve the accuracy of skin cancer diagnosis and detection. As a consequence of this study’s findings, researchers have determined that deep learning techniques are superior to machine learning for spotting skin cancer. As a consequence, pathologists’ precision and competence may be improved by using automated evidence-based detection of skin cancer. To accurately distinguish between benign and malignant skin lesions, we present a deep convolution neural network (DCNN) model in this research that uses a deep learning technique. First, we normalize the input photos and identify characteristics that aid in correct classification, then we apply a filter or Gaussian to eliminate noise and artifacts, and lastly, we supplement the data to increase the number of images, which enhances the accuracy of the classification rate.
{"title":"Towards Enhanced Deep CNN For Early And Precise Skin Cancer Diagnosis","authors":"S. Malaiarasan, R. Ravi, D.R. Maheswari, C. Rubavathi, M. Ramnath, V. Hemamalini","doi":"10.1109/ICNWC57852.2023.10127521","DOIUrl":"https://doi.org/10.1109/ICNWC57852.2023.10127521","url":null,"abstract":"Most people’s first experience with cancer will be with skin cancer, which is also the most prevalent and potentially fatal kind. Determining a skin cancer diagnosis also requires the use of information technologies. This highlights the need of developing and deploying highly effective deep-learning methods for the early and accurate diagnosis and detection of skin cancer. Deep Convolution Neural Network (DCNN) is proposed for automated skin cancer detection in this study. This study’s unique contribution is the use of a deep convolution neural network containing 12 nested processing layers to improve the accuracy of skin cancer diagnosis and detection. As a consequence of this study’s findings, researchers have determined that deep learning techniques are superior to machine learning for spotting skin cancer. As a consequence, pathologists’ precision and competence may be improved by using automated evidence-based detection of skin cancer. To accurately distinguish between benign and malignant skin lesions, we present a deep convolution neural network (DCNN) model in this research that uses a deep learning technique. First, we normalize the input photos and identify characteristics that aid in correct classification, then we apply a filter or Gaussian to eliminate noise and artifacts, and lastly, we supplement the data to increase the number of images, which enhances the accuracy of the classification rate.","PeriodicalId":197525,"journal":{"name":"2023 International Conference on Networking and Communications (ICNWC)","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":"114963747","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/ICNWC57852.2023.10127356
P. Kalaichelvi, T. Rani, S. Sakthy, G. Chidambara Raja, P. Charan Reddy
There are inefficient factors involved in agriculture like diseases in plants, plant nourishment product like inorganic fertilizers, insects and characteristics of soil in which the farmers cultivate crops. One of the profitable agriculture factors is to make proper treatment for plants by spraying organic fertilizers and planning to control the disease occurring in plants. The manual work of spraying fertilizers highly affects the farmers’ health and is time-consuming. Many farmers use drones to help them in their agricultural fields. Both fertilizers and pesticides can be sprinkled in the field with drone technology. Moreover, IoT sensors are being used to achieve a high performance of the technology. In our system, we have proposed the Even Height Maintaining (EHM) Algorithm to maintain the constant gap between plants and drones while spraying pesticides and fertilizing crops. This improves the speed of the fertilization process in agriculture and reduces the cost of drone agriculture technology. Moreover, the use of Artificial Intelligence and Deep learning in the disease detection of crops has been discussed.
{"title":"Improving Drone Technology Performance In Crop Fertilization","authors":"P. Kalaichelvi, T. Rani, S. Sakthy, G. Chidambara Raja, P. Charan Reddy","doi":"10.1109/ICNWC57852.2023.10127356","DOIUrl":"https://doi.org/10.1109/ICNWC57852.2023.10127356","url":null,"abstract":"There are inefficient factors involved in agriculture like diseases in plants, plant nourishment product like inorganic fertilizers, insects and characteristics of soil in which the farmers cultivate crops. One of the profitable agriculture factors is to make proper treatment for plants by spraying organic fertilizers and planning to control the disease occurring in plants. The manual work of spraying fertilizers highly affects the farmers’ health and is time-consuming. Many farmers use drones to help them in their agricultural fields. Both fertilizers and pesticides can be sprinkled in the field with drone technology. Moreover, IoT sensors are being used to achieve a high performance of the technology. In our system, we have proposed the Even Height Maintaining (EHM) Algorithm to maintain the constant gap between plants and drones while spraying pesticides and fertilizing crops. This improves the speed of the fertilization process in agriculture and reduces the cost of drone agriculture technology. Moreover, the use of Artificial Intelligence and Deep learning in the disease detection of crops has been discussed.","PeriodicalId":197525,"journal":{"name":"2023 International Conference on Networking and Communications (ICNWC)","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":"115201004","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/ICNWC57852.2023.10127446
M. Sabarimuthu, S. Gomathy, T. Prabhu, N. Senthilnathan, A. Harini, M. Kamaladevi
Farmers face numerous challenges on agricultural land such as planting, watering, fertilizing, etc. To overcome the issue of spraying the fertilizer, the proposed idea is to inject the fertilizer into the plants when the commands are given by the user.Here, the compost and water will be in the tank with the segment. The NPK fertilizer and water are in the tank with segment. It consists of three modes from which the user can select among these modes. In manual mode the ratio of fertilizers and water is given by the user. In Auto mode the ratio of fertilizer and water is selected automatically by knowing the name of plant. In Smart mode, the name of the plant, ratio of fertilizer and water is automatically taken by atmosphere temperature, moisture and crop data. If the user enters the type of plant using their mobile, the system will mix the needed compost with water and inject it into the plant through the drip irrigation.
{"title":"Integrated Compost Injector And Drip Irrigation For Agricultural Plant Using Iot System","authors":"M. Sabarimuthu, S. Gomathy, T. Prabhu, N. Senthilnathan, A. Harini, M. Kamaladevi","doi":"10.1109/ICNWC57852.2023.10127446","DOIUrl":"https://doi.org/10.1109/ICNWC57852.2023.10127446","url":null,"abstract":"Farmers face numerous challenges on agricultural land such as planting, watering, fertilizing, etc. To overcome the issue of spraying the fertilizer, the proposed idea is to inject the fertilizer into the plants when the commands are given by the user.Here, the compost and water will be in the tank with the segment. The NPK fertilizer and water are in the tank with segment. It consists of three modes from which the user can select among these modes. In manual mode the ratio of fertilizers and water is given by the user. In Auto mode the ratio of fertilizer and water is selected automatically by knowing the name of plant. In Smart mode, the name of the plant, ratio of fertilizer and water is automatically taken by atmosphere temperature, moisture and crop data. If the user enters the type of plant using their mobile, the system will mix the needed compost with water and inject it into the plant through the drip irrigation.","PeriodicalId":197525,"journal":{"name":"2023 International Conference on Networking and Communications (ICNWC)","volume":"9 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":"115397008","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/ICNWC57852.2023.10127360
V. Darthy Rabecka, J. Britto pari
A fundamental yet challenging problem in general image analysis is object detection. It has recently generated a lot of interest and is essential for many applications. Despite the fact that there are numerous methods, a comprehensive study of the current identity research is still necessary. This study offers a thorough overview of recent advancements in discernible image retrieval that are based on deep learning.1) Techniques for identifying objects in a region, including Fast R-CNN (fast region-based convolutional neural network), R-CNN (region-based convolutional neural network), and Mask R-CNN (mask region-based convolutional neural network) are investigated.2) Classifications in addition to regression-based object identification methods like YOLO (you only look once), SSD (single-shot detector), and Retina Net. Several benchmark sets of data from free sources that include their usual evaluation metrics. We primarily focus on deep learning algorithms in core applications, such as object identification in monitoring, combat, transport, healthcare and quotidian life.In the scrutiny, we look closely at a number of challenges, such as limited storage space and computational power, an extensive range of denominations and based on discrepancy. The survey’s conclusion is reached by demonstrating how object detection can be applied to autonomous vehicles and by enhancing the current findings in the ensuing years.
{"title":"Assessing The Performance Of Advanced Object Detection Techniques For Autonomous Cars","authors":"V. Darthy Rabecka, J. Britto pari","doi":"10.1109/ICNWC57852.2023.10127360","DOIUrl":"https://doi.org/10.1109/ICNWC57852.2023.10127360","url":null,"abstract":"A fundamental yet challenging problem in general image analysis is object detection. It has recently generated a lot of interest and is essential for many applications. Despite the fact that there are numerous methods, a comprehensive study of the current identity research is still necessary. This study offers a thorough overview of recent advancements in discernible image retrieval that are based on deep learning.1) Techniques for identifying objects in a region, including Fast R-CNN (fast region-based convolutional neural network), R-CNN (region-based convolutional neural network), and Mask R-CNN (mask region-based convolutional neural network) are investigated.2) Classifications in addition to regression-based object identification methods like YOLO (you only look once), SSD (single-shot detector), and Retina Net. Several benchmark sets of data from free sources that include their usual evaluation metrics. We primarily focus on deep learning algorithms in core applications, such as object identification in monitoring, combat, transport, healthcare and quotidian life.In the scrutiny, we look closely at a number of challenges, such as limited storage space and computational power, an extensive range of denominations and based on discrepancy. The survey’s conclusion is reached by demonstrating how object detection can be applied to autonomous vehicles and by enhancing the current findings in the ensuing years.","PeriodicalId":197525,"journal":{"name":"2023 International Conference on Networking and Communications (ICNWC)","volume":"46 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":"123067722","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}
Augmented Reality (AR) technology has the potential to revolutionize education by providing a new way for students to visualize and interact with complex concepts. In this project, a system is proposed to develop an AR smartphone application that allows students to visualize objects and scenarios that the teacher is teaching in real-time. The application will employ the smartphone’s camera and sensors to materialize a user-friendly and easy-to-use dynamic AR experience, with the teacher allowing the students to simply access their smartphone to project 3D models of objects or scenarios onto a flat surface. Students will be able to view these models from any angle and interact with them in a variety of ways, such as by rotating them or zooming in on specific details. In addition to enhancing student’s understanding of the material being taught, the AR application will also provide an engaging and immersive learning experience. The distinguishing factor is the storage of the 3D assets on the cloud that will equip the educator with the option of pre-planning and customizing their entire lesson as well as storing any number of models. This can help to increase student engagement and motivation, leading to better retention of the material being taught. Overall, the proposed AR smartphone application has the potential to significantly improve the way students learn and understand complex concepts, making education more effective and enjoyable for all.
{"title":"Augmented Reality For Education Based On Markerless Dynamic Rendering","authors":"Soumik Rakshit, Aarthi Iyer, Sunil Retmin Raj.C, Shiloah Elizabeth.D, Aditya Vaidyanathan","doi":"10.1109/ICNWC57852.2023.10127337","DOIUrl":"https://doi.org/10.1109/ICNWC57852.2023.10127337","url":null,"abstract":"Augmented Reality (AR) technology has the potential to revolutionize education by providing a new way for students to visualize and interact with complex concepts. In this project, a system is proposed to develop an AR smartphone application that allows students to visualize objects and scenarios that the teacher is teaching in real-time. The application will employ the smartphone’s camera and sensors to materialize a user-friendly and easy-to-use dynamic AR experience, with the teacher allowing the students to simply access their smartphone to project 3D models of objects or scenarios onto a flat surface. Students will be able to view these models from any angle and interact with them in a variety of ways, such as by rotating them or zooming in on specific details. In addition to enhancing student’s understanding of the material being taught, the AR application will also provide an engaging and immersive learning experience. The distinguishing factor is the storage of the 3D assets on the cloud that will equip the educator with the option of pre-planning and customizing their entire lesson as well as storing any number of models. This can help to increase student engagement and motivation, leading to better retention of the material being taught. Overall, the proposed AR smartphone application has the potential to significantly improve the way students learn and understand complex concepts, making education more effective and enjoyable for all.","PeriodicalId":197525,"journal":{"name":"2023 International Conference on Networking and Communications (ICNWC)","volume":"61 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":"125841240","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/ICNWC57852.2023.10127403
Jeelakarra Teja, K. Thilak, K. P. Reddy
The introduction of invasive species, often referred to as foreign species, to the native species occurs frequently through a variety of channels, including the air, birds, and insects. This might harm the environment in the area. Invasive plants can have a negative impact on natural ecosystems by reducing native biodiversity, altering species composition, removing habitat from native and dependent species, changing biogeochemical cycling, and changing disturbance regimes. There are a few ideas that have been made in earlier studies to prevent this, but in this study, we approach to solving this issue by combining artificial intelligence with an anomaly detection technique and image processing. We compile sample photos of each species of flower in the ecosystem and create a dataset of all local flower species. In order to create a dataset of all native flower species, we first collect sample pictures of each flower species in the environment. Analyse the image dataset quantitatively and programme a machine learning model to identify the species. In order for a qualified botanist to examine the plant and decide whether it is hazardous to the park’s ecology, it is important to identify any outlier or anomalous flower species that are found. Finding flowers in pictures is one of CNNs’ most well-known applications. For instance, a producer of sunglasses employed CNNs to recognise floral images in advertising photos. The training set in this instance included thousands of photographs of actual flowers. The photos were then appropriately recognised as flowers by the network. This is a great example of how effective CNNs can be when used properly. The user so they can look into the image’s origin.
{"title":"Detection And Alert System Of Invasive Flower Species Using Cnn","authors":"Jeelakarra Teja, K. Thilak, K. P. Reddy","doi":"10.1109/ICNWC57852.2023.10127403","DOIUrl":"https://doi.org/10.1109/ICNWC57852.2023.10127403","url":null,"abstract":"The introduction of invasive species, often referred to as foreign species, to the native species occurs frequently through a variety of channels, including the air, birds, and insects. This might harm the environment in the area. Invasive plants can have a negative impact on natural ecosystems by reducing native biodiversity, altering species composition, removing habitat from native and dependent species, changing biogeochemical cycling, and changing disturbance regimes. There are a few ideas that have been made in earlier studies to prevent this, but in this study, we approach to solving this issue by combining artificial intelligence with an anomaly detection technique and image processing. We compile sample photos of each species of flower in the ecosystem and create a dataset of all local flower species. In order to create a dataset of all native flower species, we first collect sample pictures of each flower species in the environment. Analyse the image dataset quantitatively and programme a machine learning model to identify the species. In order for a qualified botanist to examine the plant and decide whether it is hazardous to the park’s ecology, it is important to identify any outlier or anomalous flower species that are found. Finding flowers in pictures is one of CNNs’ most well-known applications. For instance, a producer of sunglasses employed CNNs to recognise floral images in advertising photos. The training set in this instance included thousands of photographs of actual flowers. The photos were then appropriately recognised as flowers by the network. This is a great example of how effective CNNs can be when used properly. The user so they can look into the image’s origin.","PeriodicalId":197525,"journal":{"name":"2023 International Conference on Networking and Communications (ICNWC)","volume":"73 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":"126108940","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/ICNWC57852.2023.10127269
G. Pradeep, T. D. V. Rayen, A. Pushpalatha, P. K. Rani
Crop production forecasting is a huge challenge nowadays, resulting in inaccurate results such as food shortages, economic instability, inefficient resource allocation, environmental impact, and lower farmer profitability. Our proposed machine-learning algorithm forecasting yield can help address these difficulties and enhance agricultural outcomes. Crop yield prediction is used to estimate the potential harvest of crops, providing valuable information to farmers, policymakers, and agribusinesses for planning, resource management, and making informed crop production decisions. It helps to improve food security, reduce food waste, and increase the efficiency of food production. Gradient Boosting Agricultural Yield Prediction is a machine learning approach that employs decision trees and gradient descent optimization to create accurate crop yield predictions. This approach and strategy are useful in predicting crop yields. They can assist farmers and agricultural organizations in making better-educated planting, harvesting, and resource allocation decisions. The results of crop yield prediction based on gradient boosting with an accuracy rate of 87.2%, precision of0.84, recall ofO.90, and F1-Score of0.87 indicate that the model is making accurate predictions about crop yields with a good balance of precision and recall. Our work suggests that the model performs efficiently and makes accurate predictions for crop yields. It increases crop production prediction, which improves decision-making, increases efficiency, effectively allocates resources, supports planning, and reduces agriculture’s environmental impact. It has a tremendous impact on the agriculture sector because it promotes sustainability, reduces waste, and improves overall performance.
{"title":"Effective Crop Yield Prediction Using Gradient Boosting To Improve Agricultural Outcomes","authors":"G. Pradeep, T. D. V. Rayen, A. Pushpalatha, P. K. Rani","doi":"10.1109/ICNWC57852.2023.10127269","DOIUrl":"https://doi.org/10.1109/ICNWC57852.2023.10127269","url":null,"abstract":"Crop production forecasting is a huge challenge nowadays, resulting in inaccurate results such as food shortages, economic instability, inefficient resource allocation, environmental impact, and lower farmer profitability. Our proposed machine-learning algorithm forecasting yield can help address these difficulties and enhance agricultural outcomes. Crop yield prediction is used to estimate the potential harvest of crops, providing valuable information to farmers, policymakers, and agribusinesses for planning, resource management, and making informed crop production decisions. It helps to improve food security, reduce food waste, and increase the efficiency of food production. Gradient Boosting Agricultural Yield Prediction is a machine learning approach that employs decision trees and gradient descent optimization to create accurate crop yield predictions. This approach and strategy are useful in predicting crop yields. They can assist farmers and agricultural organizations in making better-educated planting, harvesting, and resource allocation decisions. The results of crop yield prediction based on gradient boosting with an accuracy rate of 87.2%, precision of0.84, recall ofO.90, and F1-Score of0.87 indicate that the model is making accurate predictions about crop yields with a good balance of precision and recall. Our work suggests that the model performs efficiently and makes accurate predictions for crop yields. It increases crop production prediction, which improves decision-making, increases efficiency, effectively allocates resources, supports planning, and reduces agriculture’s environmental impact. It has a tremendous impact on the agriculture sector because it promotes sustainability, reduces waste, and improves overall performance.","PeriodicalId":197525,"journal":{"name":"2023 International Conference on Networking and Communications (ICNWC)","volume":"19 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":"124839526","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/ICNWC57852.2023.10127264
Venkata Sai P Bhamidipati, Ishi Saxena, D. Saisanthiya, Mervin Retnadhas
Robust Intelligent Posture Estimation is an important aspect of an AI Gym Trainer that can help fitness enthusiasts improve their workout technique and prevent injuries. This research presents an approach to achieve accurate posture estimation using Mediapipe and OpenCV. Mediapipe is a machine learning framework that provides pre-trained models for human posture estimation, while OpenCV is a popular computer vision library that offers a range of functions for image and video processing. The proposed solution integrates the strengths of both tools to develop a robust posture estimation system. The system first captures the user’s video feed and passes it through MediaPipe to detect the human body landmarks, then, OpenCV is used to calculate the angles between the detected landmarks in order to analyze the posture. The system provides real-time feedback to the user on their posture and suggests reparative measures. The use case that has been used for this research was repetitions for bicep curls. The proposed system can be tested on various exercises, such as squats, push-ups, and lunges. It can accurately estimate the posture of the user in different lighting conditions and is robust to occlusions and background clutter. The system can be deployed as an AI Gym Trainer and can help fitness enthusiasts improve their form and technique while reducing the risk of injury.
{"title":"Robust Intelligent Posture Estimation for an AI Gym Trainer using Mediapipe and OpenCV","authors":"Venkata Sai P Bhamidipati, Ishi Saxena, D. Saisanthiya, Mervin Retnadhas","doi":"10.1109/ICNWC57852.2023.10127264","DOIUrl":"https://doi.org/10.1109/ICNWC57852.2023.10127264","url":null,"abstract":"Robust Intelligent Posture Estimation is an important aspect of an AI Gym Trainer that can help fitness enthusiasts improve their workout technique and prevent injuries. This research presents an approach to achieve accurate posture estimation using Mediapipe and OpenCV. Mediapipe is a machine learning framework that provides pre-trained models for human posture estimation, while OpenCV is a popular computer vision library that offers a range of functions for image and video processing. The proposed solution integrates the strengths of both tools to develop a robust posture estimation system. The system first captures the user’s video feed and passes it through MediaPipe to detect the human body landmarks, then, OpenCV is used to calculate the angles between the detected landmarks in order to analyze the posture. The system provides real-time feedback to the user on their posture and suggests reparative measures. The use case that has been used for this research was repetitions for bicep curls. The proposed system can be tested on various exercises, such as squats, push-ups, and lunges. It can accurately estimate the posture of the user in different lighting conditions and is robust to occlusions and background clutter. The system can be deployed as an AI Gym Trainer and can help fitness enthusiasts improve their form and technique while reducing the risk of injury.","PeriodicalId":197525,"journal":{"name":"2023 International Conference on Networking and Communications (ICNWC)","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":"125369279","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/ICNWC57852.2023.10127447
Rishit Nagar, Nitish Chaturvedi, J. Prabakaran
Every well-known business has scammers who sell counterfeit goods at reduced prices. Due to a lack of transparency, supply chain management has frequently encountered issues such as service redundancy, poor departmental collaboration, and a compromise of standards. Counterfeiters in the market generate major challenges for legitimate businesses. Still, a significant number of individuals are unaware of the greatest extent of the harm that these products have on brands. As a result, it is essential to have a system that allows the end user to verify all details about the products purchased for the customer to determine the product’s authenticity. Combining these features with blockchain-based technology can create a coherent, effective counterfeit-reduction strategy.
{"title":"Product Authentication System using Blockchain*","authors":"Rishit Nagar, Nitish Chaturvedi, J. Prabakaran","doi":"10.1109/ICNWC57852.2023.10127447","DOIUrl":"https://doi.org/10.1109/ICNWC57852.2023.10127447","url":null,"abstract":"Every well-known business has scammers who sell counterfeit goods at reduced prices. Due to a lack of transparency, supply chain management has frequently encountered issues such as service redundancy, poor departmental collaboration, and a compromise of standards. Counterfeiters in the market generate major challenges for legitimate businesses. Still, a significant number of individuals are unaware of the greatest extent of the harm that these products have on brands. As a result, it is essential to have a system that allows the end user to verify all details about the products purchased for the customer to determine the product’s authenticity. Combining these features with blockchain-based technology can create a coherent, effective counterfeit-reduction strategy.","PeriodicalId":197525,"journal":{"name":"2023 International Conference on Networking and Communications (ICNWC)","volume":"27 10 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":"125682045","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/ICNWC57852.2023.10127467
Ranjith, V. M, Berin Shalu S
In India, the sewage system is the most serious issue. Since the drainage system isn•t properly maintained, drainage water periodically mixes with drinking water, putting people•s health in peril. We suggest the use of a smart drainage monitoring system to solve this issue. The proposed device would keep an eye on water levels in the sewage system as well as the movement of water and potentially harmful gasses. The value set will be stored in the cloud and later reviewed. The Blynk server will send an SMS with the drainage status to a point close to the corporate office. The officials of the company will then take the necessary steps.
{"title":"System for Monitoring and Controlling Drainage using Internet of Things","authors":"Ranjith, V. M, Berin Shalu S","doi":"10.1109/ICNWC57852.2023.10127467","DOIUrl":"https://doi.org/10.1109/ICNWC57852.2023.10127467","url":null,"abstract":"In India, the sewage system is the most serious issue. Since the drainage system isn•t properly maintained, drainage water periodically mixes with drinking water, putting people•s health in peril. We suggest the use of a smart drainage monitoring system to solve this issue. The proposed device would keep an eye on water levels in the sewage system as well as the movement of water and potentially harmful gasses. The value set will be stored in the cloud and later reviewed. The Blynk server will send an SMS with the drainage status to a point close to the corporate office. The officials of the company will then take the necessary steps.","PeriodicalId":197525,"journal":{"name":"2023 International Conference on Networking and Communications (ICNWC)","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":"128692235","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：