Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758280
U. Lilhore, Sarita Simaiya, Jasminder Kaur Sandhu, N. K. Trivedi, A. Garg, Aditi Moudgil
In the perspective of the Industry 4.0 (IR 4.0) model, the Deep Learning (DL) domain now has a significant impact on the production industry. The IR 4.0 model promotes intelligent sensors, systems, and devices to build intelligent industries that gather information regularly. DL method enables the development of implementable intelligence by analyzing the gathered information to boost production efficiency without dramatically changing the necessary materials. Component defects and discrepancies that impact component reliability are particularly massive in industrial processes. This research introduces a novel framework based on the VGG-16 with CNN model that creates the Intelligent Production learning center into an I4.0 production system. We describe the issue of recognizing tiny defects in an industrial inspection. The primary objective is to classify the pixel value correlating to a defect with a minimal level of false-positive results. Destructive Vs. non-destructive testing and classification procedures are mainly utilized for product quality assurance after production. Convolutional neural networks (CNN) based on machine learning (ML) methods are frequently utilized for this activity. Complex transfer learning (TL) strategies are examined in this research, which allows for the automatic detection and categorization of product defects in the manufacturing process employing industrial product samples. All the known performance metrics have been evaluated to measure and compare the model performance. The proposed VGG16 with CNN model has better outcomes for precision, recall, and accuracy as compared to exisitng CNN, VGG-16, EfficientNetB0, and Inception V3 methods.
从工业4.0 (IR 4.0)模型的角度来看，深度学习(DL)领域现在对生产行业产生了重大影响。IR 4.0模型促进智能传感器、系统和设备，以建立定期收集信息的智能产业。DL方法通过分析收集到的信息来开发可实现的智能，从而提高生产效率，而无需大幅改变必要的材料。影响组件可靠性的组件缺陷和差异在工业过程中尤为严重。本研究提出了一种基于VGG-16和CNN模型的新框架，将智能生产学习中心创建为工业4.0生产系统。我们描述了在工业检查中识别微小缺陷的问题。主要目标是将与缺陷相关的像素值与最小的假阳性结果进行分类。破坏性与非破坏性检测和分类程序主要用于产品生产后的质量保证。基于机器学习(ML)方法的卷积神经网络(CNN)经常用于此活动。本文研究了复杂迁移学习(TL)策略，该策略允许使用工业产品样本对制造过程中的产品缺陷进行自动检测和分类。对所有已知的性能指标进行了评估，以度量和比较模型的性能。与现有的CNN、VGG-16、EfficientNetB0和Inception V3方法相比，本文提出的带有CNN模型的VGG16在精密度、查全率和准确率方面都有更好的结果。
{"title":"Deep Learning-Based Predictive Model for Defect Detection and Classification in Industry 4.0","authors":"U. Lilhore, Sarita Simaiya, Jasminder Kaur Sandhu, N. K. Trivedi, A. Garg, Aditi Moudgil","doi":"10.1109/ESCI53509.2022.9758280","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758280","url":null,"abstract":"In the perspective of the Industry 4.0 (IR 4.0) model, the Deep Learning (DL) domain now has a significant impact on the production industry. The IR 4.0 model promotes intelligent sensors, systems, and devices to build intelligent industries that gather information regularly. DL method enables the development of implementable intelligence by analyzing the gathered information to boost production efficiency without dramatically changing the necessary materials. Component defects and discrepancies that impact component reliability are particularly massive in industrial processes. This research introduces a novel framework based on the VGG-16 with CNN model that creates the Intelligent Production learning center into an I4.0 production system. We describe the issue of recognizing tiny defects in an industrial inspection. The primary objective is to classify the pixel value correlating to a defect with a minimal level of false-positive results. Destructive Vs. non-destructive testing and classification procedures are mainly utilized for product quality assurance after production. Convolutional neural networks (CNN) based on machine learning (ML) methods are frequently utilized for this activity. Complex transfer learning (TL) strategies are examined in this research, which allows for the automatic detection and categorization of product defects in the manufacturing process employing industrial product samples. All the known performance metrics have been evaluated to measure and compare the model performance. The proposed VGG16 with CNN model has better outcomes for precision, recall, and accuracy as compared to exisitng CNN, VGG-16, EfficientNetB0, and Inception V3 methods.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116480284","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 : 2022-03-09DOI: 10.1109/esci53509.2022.9758377
{"title":"ESCI 2022 Programme Schedule","authors":"","doi":"10.1109/esci53509.2022.9758377","DOIUrl":"https://doi.org/10.1109/esci53509.2022.9758377","url":null,"abstract":"","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133411744","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}
Particle swarm optimization (PSO) comes from a family of swarm optimization techniques that work iteratively to obtain an optimum solution for single or multi objective systems. For instance, teacher learner-based optimization (TLbO) when combined with PSO, fuses swarm intelligence behaviour with teacher-learner relationship for speeding up the learning process. However most of these algorithms do not modify the original PSO learning factors, due to which their performance is limited. In this work, a novel adaptive learning-based TLbO inspired PSO model is proposed. This model aims at improving the convergence speed and reduce solution error via adaptively learning from previous iteration error and modifying social and cognitive learning behaviour of the underlying PSO. The proposed model is 20% more efficient in terms of convergence delay, and 25% efficient in terms of final solution error when compared with existing highly efficient TLbO-PSO models.
{"title":"Enhancing the Convergence Speed and Accuracy of Particle Swarm Optimizers through Adaptive Learning","authors":"Santosh Lavate, Amol Avinash Joshi, Trupti Smit Shinde","doi":"10.1109/ESCI53509.2022.9758308","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758308","url":null,"abstract":"Particle swarm optimization (PSO) comes from a family of swarm optimization techniques that work iteratively to obtain an optimum solution for single or multi objective systems. For instance, teacher learner-based optimization (TLbO) when combined with PSO, fuses swarm intelligence behaviour with teacher-learner relationship for speeding up the learning process. However most of these algorithms do not modify the original PSO learning factors, due to which their performance is limited. In this work, a novel adaptive learning-based TLbO inspired PSO model is proposed. This model aims at improving the convergence speed and reduce solution error via adaptively learning from previous iteration error and modifying social and cognitive learning behaviour of the underlying PSO. The proposed model is 20% more efficient in terms of convergence delay, and 25% efficient in terms of final solution error when compared with existing highly efficient TLbO-PSO models.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123431538","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758293
Mehul Jain, Kajal Gupta, Rajni Jindal
The most substantial organ of the human body is the skin. It plays an essential role in the sustenance of life and health. It helps in providing an airtight, watertight and flexible barrier between the internal body organs and the adverse elements from outside environment. Skin conditions contribute 1.79% of the global burden of disease worldwide. Development in techniques to visually inspect a skin disease is essential to fasten diagnosis and minimise life-threatening situations. Automated classification of skin disorders via image processing and various machine learning algorithms have been proposed in the literature. Previous research has demonstrated that Convolutional Neural Networks (CNNs) have great ability to recognise specific regions in images without providing the annotated bounding boxes of those specific regions. Hence, we plan to compare a custom CNN model along with the Residual Attention Network model and a custom CNN model based on ResNet without any attention layers for skin classification problems. The attention layer would improve the localisation ability of a CNN model and consider only the relevant regions from the images. Moreover, the residual network works better for small sample learning problems. So, a combination of residual and attention units is suitable to tackle the concerned problems.
{"title":"Improving Skin Disease Classification using Residual Attention Network","authors":"Mehul Jain, Kajal Gupta, Rajni Jindal","doi":"10.1109/ESCI53509.2022.9758293","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758293","url":null,"abstract":"The most substantial organ of the human body is the skin. It plays an essential role in the sustenance of life and health. It helps in providing an airtight, watertight and flexible barrier between the internal body organs and the adverse elements from outside environment. Skin conditions contribute 1.79% of the global burden of disease worldwide. Development in techniques to visually inspect a skin disease is essential to fasten diagnosis and minimise life-threatening situations. Automated classification of skin disorders via image processing and various machine learning algorithms have been proposed in the literature. Previous research has demonstrated that Convolutional Neural Networks (CNNs) have great ability to recognise specific regions in images without providing the annotated bounding boxes of those specific regions. Hence, we plan to compare a custom CNN model along with the Residual Attention Network model and a custom CNN model based on ResNet without any attention layers for skin classification problems. The attention layer would improve the localisation ability of a CNN model and consider only the relevant regions from the images. Moreover, the residual network works better for small sample learning problems. So, a combination of residual and attention units is suitable to tackle the concerned problems.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129744688","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758379
D. Gupta, Aditi Moudgil, Shivani Wadhwa, Vikas Solanki
We live in a world where huge end devices execute computing on a daily basis. With the growing number of sophisticated apps (e.g., augmented reality and face recognition) that require considerably more computational capacity, they are shifting to mobile cloud computing (MCC), or offloading computation to the cloud. Unfortunately, because the cloud is typically located far away from end devices, latency and quality of experience (QoE) for delay-sensitive applications suffer. Mobile edge computing (MEC) is considered to be a viable solution for meeting the requirement for low latency. Prior works on edge computing mostly focused on computation offloading to support low latency. This paper Jointly considered data caching and computation offloading to support better QoE for end device users. With caching of completed tasks data and offloading of computations at edge cloud using an efficient approach termed as data caching and computation offloading at edge (DCCO-E), the simulation results proved outstanding performance of the DCCO-E against other schemes in terms of low energy consumption and reduced latency.
{"title":"Efficient Data Caching and Computation Offloading Strategy for Edge Network","authors":"D. Gupta, Aditi Moudgil, Shivani Wadhwa, Vikas Solanki","doi":"10.1109/ESCI53509.2022.9758379","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758379","url":null,"abstract":"We live in a world where huge end devices execute computing on a daily basis. With the growing number of sophisticated apps (e.g., augmented reality and face recognition) that require considerably more computational capacity, they are shifting to mobile cloud computing (MCC), or offloading computation to the cloud. Unfortunately, because the cloud is typically located far away from end devices, latency and quality of experience (QoE) for delay-sensitive applications suffer. Mobile edge computing (MEC) is considered to be a viable solution for meeting the requirement for low latency. Prior works on edge computing mostly focused on computation offloading to support low latency. This paper Jointly considered data caching and computation offloading to support better QoE for end device users. With caching of completed tasks data and offloading of computations at edge cloud using an efficient approach termed as data caching and computation offloading at edge (DCCO-E), the simulation results proved outstanding performance of the DCCO-E against other schemes in terms of low energy consumption and reduced latency.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117228564","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758266
Hariharan U, V. Dhanakoti, K. Rajkumar, J. Jeyavel
The decentralized nature of vulnerable health records can bring about cases where timely records are unavailable, worsening overall health results. Moreover, as patient participation in healthcare increases, there's a growing demand for individuals to control and access the data. Blockchain is a protected, decentralized online server which may be employed to handle electronic health records (EHRs) efficiently, therefore with the possibility to boost health outcomes by producing a stream for interoperability. Thus, it's of key-value to secure electronic overall health captures. Centralized storage space of comprehensive health information appeals to constant viewing, and cyber-attacks of affected person captures are complex. Thus, it's essential to develop a method while using the cloud, which enables you to guarantee authentication and offers the integrity of overall health captures. The keyless signature infrastructure utilized within the suggested method of ensuring electronic signatures' secrecy also guarantees elements of authentication. Besides, information integrity is handled through the proposed blockchain technologies. The functionality of the suggested framework is examined by evaluating the variables such as typical period, sizing, and then the price of information storage space and retrieval on the blockchain know-how with traditional details storage space methods. The result reveals that the resulting period of the suggested process, together with blockchain engineering, is practically 55% smaller than traditional strategies. Also, they voice the price of storage space is approximately 25% less when it comes to the method with Blockchain in deep comparability with all the pre-existing strategies.
{"title":"Safeguarding E-Healthcare Documents Utilizing an Infrastructure for Advanced Keyless Signature Infrastructure Blockchain Within the Cloud","authors":"Hariharan U, V. Dhanakoti, K. Rajkumar, J. Jeyavel","doi":"10.1109/ESCI53509.2022.9758266","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758266","url":null,"abstract":"The decentralized nature of vulnerable health records can bring about cases where timely records are unavailable, worsening overall health results. Moreover, as patient participation in healthcare increases, there's a growing demand for individuals to control and access the data. Blockchain is a protected, decentralized online server which may be employed to handle electronic health records (EHRs) efficiently, therefore with the possibility to boost health outcomes by producing a stream for interoperability. Thus, it's of key-value to secure electronic overall health captures. Centralized storage space of comprehensive health information appeals to constant viewing, and cyber-attacks of affected person captures are complex. Thus, it's essential to develop a method while using the cloud, which enables you to guarantee authentication and offers the integrity of overall health captures. The keyless signature infrastructure utilized within the suggested method of ensuring electronic signatures' secrecy also guarantees elements of authentication. Besides, information integrity is handled through the proposed blockchain technologies. The functionality of the suggested framework is examined by evaluating the variables such as typical period, sizing, and then the price of information storage space and retrieval on the blockchain know-how with traditional details storage space methods. The result reveals that the resulting period of the suggested process, together with blockchain engineering, is practically 55% smaller than traditional strategies. Also, they voice the price of storage space is approximately 25% less when it comes to the method with Blockchain in deep comparability with all the pre-existing strategies.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121589880","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758254
Ahmmad Musha, Rehnuma Hasnat, Abdullah Al Mamun, Tonmoy Ghosh
Polyps are one of the most common gastrointestinal diseases. It has the potential to cause fatal colon and rectal cancers. As a result, it must be removed during the primitive stage. In this paper, we developed an algorithm that uses endoscopy images to detect polyp removal status. We investigated convolutional neural networks such as DenseNet, ResNet, VGG, MobileNet, and others to extract features from images and then use those features to classify whether a polyp is completely removed or not. 1000 dyed resection margins and 1000 dyed and lifted polyps' images from a publicly available dataset were used to test and train the proposed models. On the testing dataset, we obtained 85% sensitivity, 88% precision, and 85% fl-scores by using MobileNet architecture. This computer-aided polyp removal method assists physicians in diagnosing polyp status in a reliable, quick, and cost-effective manner.
{"title":"Deep Learning-Based Comparative Study to Detect Polyp Removal in Endoscopic Images","authors":"Ahmmad Musha, Rehnuma Hasnat, Abdullah Al Mamun, Tonmoy Ghosh","doi":"10.1109/ESCI53509.2022.9758254","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758254","url":null,"abstract":"Polyps are one of the most common gastrointestinal diseases. It has the potential to cause fatal colon and rectal cancers. As a result, it must be removed during the primitive stage. In this paper, we developed an algorithm that uses endoscopy images to detect polyp removal status. We investigated convolutional neural networks such as DenseNet, ResNet, VGG, MobileNet, and others to extract features from images and then use those features to classify whether a polyp is completely removed or not. 1000 dyed resection margins and 1000 dyed and lifted polyps' images from a publicly available dataset were used to test and train the proposed models. On the testing dataset, we obtained 85% sensitivity, 88% precision, and 85% fl-scores by using MobileNet architecture. This computer-aided polyp removal method assists physicians in diagnosing polyp status in a reliable, quick, and cost-effective manner.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114541143","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758331
Nasmin Jiwani, Ketan Gupta, Neda Afreen
The EEG signal is made up of numerous frequency bands that characterize human behaviours like emotion, attentiveness, and sleep status, among others. In order to detect epileptical seizures, categorization based on discrete EEG segments is required. The performance of the theta band in an EEG signal is analyzed with the Short-Time Fourier Transform (STFT). It also analyses different categorization methodologies, demonstrating that some classification algorithms achieve extremely high accuracy. The analysis was done in stages, with STFT, theta frequency band extraction, statistical feature extraction, and then classification using LightGBM and Catboost classifier at the end. STFT is used in this study to extract statistical properties from 2-dimensional data and classify epilepsy in the low frequency range. The proposed LightGBM and CatBoost classifier got 98.33% accuracy.
{"title":"Automated Seizure Detection using Theta Band","authors":"Nasmin Jiwani, Ketan Gupta, Neda Afreen","doi":"10.1109/ESCI53509.2022.9758331","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758331","url":null,"abstract":"The EEG signal is made up of numerous frequency bands that characterize human behaviours like emotion, attentiveness, and sleep status, among others. In order to detect epileptical seizures, categorization based on discrete EEG segments is required. The performance of the theta band in an EEG signal is analyzed with the Short-Time Fourier Transform (STFT). It also analyses different categorization methodologies, demonstrating that some classification algorithms achieve extremely high accuracy. The analysis was done in stages, with STFT, theta frequency band extraction, statistical feature extraction, and then classification using LightGBM and Catboost classifier at the end. STFT is used in this study to extract statistical properties from 2-dimensional data and classify epilepsy in the low frequency range. The proposed LightGBM and CatBoost classifier got 98.33% accuracy.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123120365","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758284
Akshay Dhande, R. Malik
In crop-imagery different algorithms have been proposed over the years which determine crop-growth, crop-diseases, crop-yield etc., using a series of image processing steps. As large number of architectures are available in the area of crop imaging, selection of particular algorithm is a very much crucial task for getting optimum results from the set off application. A lot of research is required for this, which increases the delay in the system design, to reduce this delay this paper reviews the best algorithm set in terms of their statistical parameter. The error rate and accuracy of different algorithms is compared in order to understand performance of different algorithms. This will facilitate the investigator to search out the most effective practices in connection with crop disease detection and crop yield prediction.
{"title":"Empirical Study of Crop-disease Detection and Crop-yield Analysis Systems: A Statistical View","authors":"Akshay Dhande, R. Malik","doi":"10.1109/ESCI53509.2022.9758284","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758284","url":null,"abstract":"In crop-imagery different algorithms have been proposed over the years which determine crop-growth, crop-diseases, crop-yield etc., using a series of image processing steps. As large number of architectures are available in the area of crop imaging, selection of particular algorithm is a very much crucial task for getting optimum results from the set off application. A lot of research is required for this, which increases the delay in the system design, to reduce this delay this paper reviews the best algorithm set in terms of their statistical parameter. The error rate and accuracy of different algorithms is compared in order to understand performance of different algorithms. This will facilitate the investigator to search out the most effective practices in connection with crop disease detection and crop yield prediction.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124489066","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758333
Hamzah Shabbir, Ankita Chaturvedi
This paper aims to propose and analyze a method to combine the Machine learning model with the Time-series model for hybrid forecasting of Global Horizontal Irradiance (GHI). This hybrid model exploits the performance of the Time-series model and Machine learning model, which perform differently on a different set of weather conditions, to give a more accurate result. For this research, Random Forest has been used as a machine learning model, and for the Time-series model, Seasonal Autoregressive Integrated Moving Average with exogenous regressors (SARIMAX) model has been used. The machine learning model considers weather conditions such as humidity, cloud cover temp, etc., to predict GHI. The time series model only depends on past data values, which makes it independent of weather conditions. A hybrid forecast tends to exploit the advantages of both models and overcome limitations. The final estimates from the Hybrid model contain the weight of each model, which is calculated during the validation period using a regression algorithm.
本文旨在提出并分析一种将机器学习模型与时间序列模型相结合的全球水平辐照度(GHI)混合预测方法。这种混合模型利用了时间序列模型和机器学习模型的性能，它们在不同的天气条件下表现不同，从而给出更准确的结果。本研究采用随机森林作为机器学习模型，时间序列模型采用SARIMAX (Seasonal Autoregressive Integrated Moving Average with exogenous regressors)模型。机器学习模型考虑天气条件，如湿度、云层覆盖温度等，来预测GHI。时间序列模型只依赖于过去的数据值，这使得它独立于天气条件。混合预测倾向于利用两种模式的优点并克服其局限性。Hybrid模型的最终估计包含每个模型的权重，这是在验证期间使用回归算法计算的。
{"title":"Day Ahead Hybrid Forecasting of Global Horizontal Irradiance using Machine Learning (Random Forest Algorithm) and Time-Series Model (SARIMAX)","authors":"Hamzah Shabbir, Ankita Chaturvedi","doi":"10.1109/ESCI53509.2022.9758333","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758333","url":null,"abstract":"This paper aims to propose and analyze a method to combine the Machine learning model with the Time-series model for hybrid forecasting of Global Horizontal Irradiance (GHI). This hybrid model exploits the performance of the Time-series model and Machine learning model, which perform differently on a different set of weather conditions, to give a more accurate result. For this research, Random Forest has been used as a machine learning model, and for the Time-series model, Seasonal Autoregressive Integrated Moving Average with exogenous regressors (SARIMAX) model has been used. The machine learning model considers weather conditions such as humidity, cloud cover temp, etc., to predict GHI. The time series model only depends on past data values, which makes it independent of weather conditions. A hybrid forecast tends to exploit the advantages of both models and overcome limitations. The final estimates from the Hybrid model contain the weight of each model, which is calculated during the validation period using a regression algorithm.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121291887","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：