Pub Date : 2021-05-19DOI: 10.1109/ETI4.051663.2021.9619354
Rameez R. Chowdhary, M. K. Chattopadhyay
The paper presents a model for Automated Guided Vehicle (AGV) like mobile Robots (RBs). The model is based on our Orchestrated approach. RB uses this model to perform a transportation task in environments such as a warehouse or a factory. The RB utilises D* Lite algorithm for path trajectory generation and implements our proposed modified extended navigation (ENG) algorithm to follow the path trajectory. Additionally, ENG algorithm helps RBs to avoid collisions during transportation between start and end point. We have improved the efficiency, consistency and capability of ENG algorithm by adding new method. The RB employs sensor data-fusion technique. The technique helps in reducing the position error during transportation. Our algorithm also helps the RBs to avoid the deadlock situation and make the model fault-tolerant. The performance of model has been validated with the help of new experiments. The Orchestration of Robotic Platform (ORP) with four robots is used to perform the experiments.
{"title":"Orchestration of Automated Guided Mobile Robots for Transportation Task in a Warehouse like Environment","authors":"Rameez R. Chowdhary, M. K. Chattopadhyay","doi":"10.1109/ETI4.051663.2021.9619354","DOIUrl":"https://doi.org/10.1109/ETI4.051663.2021.9619354","url":null,"abstract":"The paper presents a model for Automated Guided Vehicle (AGV) like mobile Robots (RBs). The model is based on our Orchestrated approach. RB uses this model to perform a transportation task in environments such as a warehouse or a factory. The RB utilises D* Lite algorithm for path trajectory generation and implements our proposed modified extended navigation (ENG) algorithm to follow the path trajectory. Additionally, ENG algorithm helps RBs to avoid collisions during transportation between start and end point. We have improved the efficiency, consistency and capability of ENG algorithm by adding new method. The RB employs sensor data-fusion technique. The technique helps in reducing the position error during transportation. Our algorithm also helps the RBs to avoid the deadlock situation and make the model fault-tolerant. The performance of model has been validated with the help of new experiments. The Orchestration of Robotic Platform (ORP) with four robots is used to perform the experiments.","PeriodicalId":129682,"journal":{"name":"2021 Emerging Trends in Industry 4.0 (ETI 4.0)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133544129","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 : 2021-05-19DOI: 10.1109/ETI4.051663.2021.9619298
Devipriya G P, G. V., S. K. Nair
An unequal arm interferometer beyond the coherence length of laser diode led to the development of laser instrumentation which is studied theoretically and experimentally in this paper. By using a low-cost diode laser having 650 nm wavelength and short coherence length, a modified Michelson’s interferometer is constructed for the measurements of various physical parameters for long distance measurements. The coherence length is not an abrupt distance. Beyond the coherence length the fringe contrast reduces gradually and becomes indistinguishable. Using better detection techniques, the fringe visibility can be improved to a greater extend. This paper further investigates the other applications of lasers beyond the coherence length.
{"title":"Study on the Development of Laser Instrumentation Beyond the Coherence Length of Laser Diode","authors":"Devipriya G P, G. V., S. K. Nair","doi":"10.1109/ETI4.051663.2021.9619298","DOIUrl":"https://doi.org/10.1109/ETI4.051663.2021.9619298","url":null,"abstract":"An unequal arm interferometer beyond the coherence length of laser diode led to the development of laser instrumentation which is studied theoretically and experimentally in this paper. By using a low-cost diode laser having 650 nm wavelength and short coherence length, a modified Michelson’s interferometer is constructed for the measurements of various physical parameters for long distance measurements. The coherence length is not an abrupt distance. Beyond the coherence length the fringe contrast reduces gradually and becomes indistinguishable. Using better detection techniques, the fringe visibility can be improved to a greater extend. This paper further investigates the other applications of lasers beyond the coherence length.","PeriodicalId":129682,"journal":{"name":"2021 Emerging Trends in Industry 4.0 (ETI 4.0)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133902070","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 : 2021-05-19DOI: 10.1109/ETI4.051663.2021.9619334
Jagadishwari V
The year 2020 began with the outbreak of covid-19 Pandemic, it originated in China and very quickly spread to all the other parts of the world. The deadly virus badly affected the health and economy of Mankind. This work aims to build Machine learning models to predict the spread of Covid -19. The up to date Time series data set of Covid 19 is used in the analytics. Three prediction models namely Regression, SVM and FBProphet are implemented. The results obtained from these models are investigated. FBProphet gives promising results as compared to the other models. The trend and seasonality components of FBProphet are shown to be very useful in the analysis of Time Series Data.
{"title":"Time series Covid 19 Predictions with Machine Learning Models","authors":"Jagadishwari V","doi":"10.1109/ETI4.051663.2021.9619334","DOIUrl":"https://doi.org/10.1109/ETI4.051663.2021.9619334","url":null,"abstract":"The year 2020 began with the outbreak of covid-19 Pandemic, it originated in China and very quickly spread to all the other parts of the world. The deadly virus badly affected the health and economy of Mankind. This work aims to build Machine learning models to predict the spread of Covid -19. The up to date Time series data set of Covid 19 is used in the analytics. Three prediction models namely Regression, SVM and FBProphet are implemented. The results obtained from these models are investigated. FBProphet gives promising results as compared to the other models. The trend and seasonality components of FBProphet are shown to be very useful in the analysis of Time Series Data.","PeriodicalId":129682,"journal":{"name":"2021 Emerging Trends in Industry 4.0 (ETI 4.0)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133968000","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 : 2021-05-19DOI: 10.1109/ETI4.051663.2021.9619295
Yahye Omar, S. B. Goyal, Vijayakumar Varadarajan
The world is only beginning to see the value and potential impact of the internet of things (IoT). Until recently, access to the internet was bounded via desktop, tablet, or smartphone. With the (IoT), practically all devices and objects can be connected to the internet and monitored remotely. IoT devices simplify our lives and make organizations more efficient; however, there are still challenges to address, particularly in the security context. As we continue to embed these connected objects and a wider variety of wireless devices, it is mandatory to provide confidence in this vast incoming information source. Blockchain has emerged as a disruptive technology that will transform the way we store, share information, and impose restrictions to know the authentications. The data distribution and robust level of encryption will remove the need for trust among the involved parties and add another security layer for IoT data. IoT devices generate too much data using sensors and stored, processed, accessed the same using cloud computing and achieve security some extend using big-data. Big-data security mechanism is not sufficient to meet the security requirements of IoT devices. We have proposed the Blockchain encryption mechanism using different layers architecture for the IoT devices to achieve the desired security level. In this paper, we have focused on how Blockchain could possibly improve IoT security. We also survey the most relevant work to investigate challenges associate with IoT Blockchain convergence. This proposed mechanism will achieve the security mechanism in IoT devices some extend.
{"title":"Apply Blockchain Technology for Security of IoT Devices","authors":"Yahye Omar, S. B. Goyal, Vijayakumar Varadarajan","doi":"10.1109/ETI4.051663.2021.9619295","DOIUrl":"https://doi.org/10.1109/ETI4.051663.2021.9619295","url":null,"abstract":"The world is only beginning to see the value and potential impact of the internet of things (IoT). Until recently, access to the internet was bounded via desktop, tablet, or smartphone. With the (IoT), practically all devices and objects can be connected to the internet and monitored remotely. IoT devices simplify our lives and make organizations more efficient; however, there are still challenges to address, particularly in the security context. As we continue to embed these connected objects and a wider variety of wireless devices, it is mandatory to provide confidence in this vast incoming information source. Blockchain has emerged as a disruptive technology that will transform the way we store, share information, and impose restrictions to know the authentications. The data distribution and robust level of encryption will remove the need for trust among the involved parties and add another security layer for IoT data. IoT devices generate too much data using sensors and stored, processed, accessed the same using cloud computing and achieve security some extend using big-data. Big-data security mechanism is not sufficient to meet the security requirements of IoT devices. We have proposed the Blockchain encryption mechanism using different layers architecture for the IoT devices to achieve the desired security level. In this paper, we have focused on how Blockchain could possibly improve IoT security. We also survey the most relevant work to investigate challenges associate with IoT Blockchain convergence. This proposed mechanism will achieve the security mechanism in IoT devices some extend.","PeriodicalId":129682,"journal":{"name":"2021 Emerging Trends in Industry 4.0 (ETI 4.0)","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114404282","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 : 2021-05-19DOI: 10.1109/ETI4.051663.2021.9619357
Z. Rustam, S. Hartini
Since the first case of COVID-19 appeared in Wuhan city, China, in December 2019, the disease has affected more than millions of people worldwide. Therefore, early detection of COVID-19 is important to prevent transmission to more people. One method widely used to detect COVID-19 through X-ray images is Convolutional Neural Networks (CNN). However, CNN needs large amounts of image data to build models with high accuracy, while the medical image has limited amounts of data. To overcome this problem, transfer learning technique where CNN is used as a feature extraction method is usually be chosen as an alternative. However, most studies use the extraction results of the final layers such as fully connected layer or the last convolutional layer. In this study, all layers will be used by turns to analyze how the extraction results affect the performance of classification method. The CNN models used are pre-trained models VGG16 and VGG19, while the classification method used is Support Vector Machines (SVM). Based on the results of the study, the extraction results by the initial layer gave a better performance on SVM compared to the layers that are deeper in the selected CNN architecture. Several layers in CNN model did not analyze due to limited source capability in doing computation. Therefore, as the future work, the rest layers of CNN in this study can be analyzed as well as the other CNN models and the classification method.
{"title":"Performance Analysis of Deep Convolutional Features using Support Vector Machines for COVID-19 Diagnosis on X-ray Images","authors":"Z. Rustam, S. Hartini","doi":"10.1109/ETI4.051663.2021.9619357","DOIUrl":"https://doi.org/10.1109/ETI4.051663.2021.9619357","url":null,"abstract":"Since the first case of COVID-19 appeared in Wuhan city, China, in December 2019, the disease has affected more than millions of people worldwide. Therefore, early detection of COVID-19 is important to prevent transmission to more people. One method widely used to detect COVID-19 through X-ray images is Convolutional Neural Networks (CNN). However, CNN needs large amounts of image data to build models with high accuracy, while the medical image has limited amounts of data. To overcome this problem, transfer learning technique where CNN is used as a feature extraction method is usually be chosen as an alternative. However, most studies use the extraction results of the final layers such as fully connected layer or the last convolutional layer. In this study, all layers will be used by turns to analyze how the extraction results affect the performance of classification method. The CNN models used are pre-trained models VGG16 and VGG19, while the classification method used is Support Vector Machines (SVM). Based on the results of the study, the extraction results by the initial layer gave a better performance on SVM compared to the layers that are deeper in the selected CNN architecture. Several layers in CNN model did not analyze due to limited source capability in doing computation. Therefore, as the future work, the rest layers of CNN in this study can be analyzed as well as the other CNN models and the classification method.","PeriodicalId":129682,"journal":{"name":"2021 Emerging Trends in Industry 4.0 (ETI 4.0)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134300314","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 : 2021-05-19DOI: 10.1109/ETI4.051663.2021.9619353
Lipismita Panigrahi, K. Verma
Reliability and accuracy is the key concern of an automated image classification process. However, the impact of background or surrounding area is very less in compared to object features, which create ambiguity while assigning the appropriate class label and reduce the classification accuracy. This paper presents a new model to address this issue which select the relevant features from the segmented images based on the inner and outer regions. The key idea of this model is that the texture features inside the objects are more relevant than the surrounding or outside region of the objects. The proposed model applying a segmentation method for automated segment the image. These segmented images are further partition into two parts (i.e. inner and outer). The 463 shape and texture features are extracted from the inner, outer parts of the segmented images and also from the whole image. Next, these extracted features are used to train the classifier using support vector machine (SVM). A database of 644 images that consists of 8 classes is used to verify the efficacy of the proposed model. The result proves the efficacy of the proposed model which achieves classification accuracy up to 97.79 % from the inner part of the image. The classification accuracy of inner features is increased by 9.58% from surroundings features.
{"title":"Segmented Region based Feature Extraction for Image Classification","authors":"Lipismita Panigrahi, K. Verma","doi":"10.1109/ETI4.051663.2021.9619353","DOIUrl":"https://doi.org/10.1109/ETI4.051663.2021.9619353","url":null,"abstract":"Reliability and accuracy is the key concern of an automated image classification process. However, the impact of background or surrounding area is very less in compared to object features, which create ambiguity while assigning the appropriate class label and reduce the classification accuracy. This paper presents a new model to address this issue which select the relevant features from the segmented images based on the inner and outer regions. The key idea of this model is that the texture features inside the objects are more relevant than the surrounding or outside region of the objects. The proposed model applying a segmentation method for automated segment the image. These segmented images are further partition into two parts (i.e. inner and outer). The 463 shape and texture features are extracted from the inner, outer parts of the segmented images and also from the whole image. Next, these extracted features are used to train the classifier using support vector machine (SVM). A database of 644 images that consists of 8 classes is used to verify the efficacy of the proposed model. The result proves the efficacy of the proposed model which achieves classification accuracy up to 97.79 % from the inner part of the image. The classification accuracy of inner features is increased by 9.58% from surroundings features.","PeriodicalId":129682,"journal":{"name":"2021 Emerging Trends in Industry 4.0 (ETI 4.0)","volume":"62 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134543629","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 : 2021-05-19DOI: 10.1109/ETI4.051663.2021.9619350
Asha Sara Thomas, E. Sasikala
In the last ten years, Lung Cancer and Chronic Obstructive Pulmonary Disease (COPD) have become two major diseases in the category of Respiratory Diseases which have lead to a large number of death rates in India and also in other countries. The main reason for the increase in these cases is due to the excessive smoking habit among youngsters and adults. Thus, proper diagnosis of both lung cancer and COPD are important in order to save human life. A fast and effective method to do this is to differentiate accurately among both diseases and provide the required treatment. This paper focuses on efficiently differentiating among chest pathologies in chest X-Ray using different artificial neural networks, machine learning, and deep learning approaches. It shows how an artificial neural network can be used in the prediction of diseases based on the image sets. ResNets help in better feature extraction of the image sets that lead to the correct classification of diseases. The model achieves a better performance in evaluating chest radiograph datasets that depicts the changes caused in a person's lungs when compared to normal lung images such as the formation of small lobes (or) the enlarged arteries in lungs and so on..
{"title":"Identifying Lung Cancer and Chronic Obstructive Pulmonary Diseases using Residual Neural Network","authors":"Asha Sara Thomas, E. Sasikala","doi":"10.1109/ETI4.051663.2021.9619350","DOIUrl":"https://doi.org/10.1109/ETI4.051663.2021.9619350","url":null,"abstract":"In the last ten years, Lung Cancer and Chronic Obstructive Pulmonary Disease (COPD) have become two major diseases in the category of Respiratory Diseases which have lead to a large number of death rates in India and also in other countries. The main reason for the increase in these cases is due to the excessive smoking habit among youngsters and adults. Thus, proper diagnosis of both lung cancer and COPD are important in order to save human life. A fast and effective method to do this is to differentiate accurately among both diseases and provide the required treatment. This paper focuses on efficiently differentiating among chest pathologies in chest X-Ray using different artificial neural networks, machine learning, and deep learning approaches. It shows how an artificial neural network can be used in the prediction of diseases based on the image sets. ResNets help in better feature extraction of the image sets that lead to the correct classification of diseases. The model achieves a better performance in evaluating chest radiograph datasets that depicts the changes caused in a person's lungs when compared to normal lung images such as the formation of small lobes (or) the enlarged arteries in lungs and so on..","PeriodicalId":129682,"journal":{"name":"2021 Emerging Trends in Industry 4.0 (ETI 4.0)","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132189994","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 : 2021-05-19DOI: 10.1109/ETI4.051663.2021.9619207
G. Raviteja, K.S.Rama Praveen, K.Anisha Keerthi, R. Abhishek, V. Sarvari
In this proposed paper, a quad-port C band conformal MIMO antenna is designed. This antenna configuration has four similar CPW-fed elements of size 10x15 mm. It is supported with flexible FR4 epoxy dielectric material with relative permittivity of 4.4 and a loss tangent of 0.02. The proposed antenna achieved an impedance bandwidth in accordance with the -10dB reference from frequency ranges of 4.5 GHz to 7.56 GHz which covers C band satellite applications. Good isolation characteristics are achieved which is less than -15 dB with the help of the orthogonal arrangement of the four MIMO antennas. For the excellent working of MIMO, some of the characteristics like Mean Effective Gain, Total Active Reflection, Envelope Correlation Coefficient are considered as important and they are investigated and found that they are within the standards as MEG < 3dB and ECC < 0.5. The entire work is done with the help of ANSYS High-Frequency Structure Simulator (HFSS) software.
{"title":"A CPW Feed Orthogonal Wideband Quad-Port Conformal MIMO Antenna for Satellite Applications","authors":"G. Raviteja, K.S.Rama Praveen, K.Anisha Keerthi, R. Abhishek, V. Sarvari","doi":"10.1109/ETI4.051663.2021.9619207","DOIUrl":"https://doi.org/10.1109/ETI4.051663.2021.9619207","url":null,"abstract":"In this proposed paper, a quad-port C band conformal MIMO antenna is designed. This antenna configuration has four similar CPW-fed elements of size 10x15 mm. It is supported with flexible FR4 epoxy dielectric material with relative permittivity of 4.4 and a loss tangent of 0.02. The proposed antenna achieved an impedance bandwidth in accordance with the -10dB reference from frequency ranges of 4.5 GHz to 7.56 GHz which covers C band satellite applications. Good isolation characteristics are achieved which is less than -15 dB with the help of the orthogonal arrangement of the four MIMO antennas. For the excellent working of MIMO, some of the characteristics like Mean Effective Gain, Total Active Reflection, Envelope Correlation Coefficient are considered as important and they are investigated and found that they are within the standards as MEG < 3dB and ECC < 0.5. The entire work is done with the help of ANSYS High-Frequency Structure Simulator (HFSS) software.","PeriodicalId":129682,"journal":{"name":"2021 Emerging Trends in Industry 4.0 (ETI 4.0)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132639677","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 : 2021-05-19DOI: 10.1109/ETI4.051663.2021.9619238
Sai Kumar T S, Prabalakshmi A, A. K, S. Alagammal
Recently, many Deep Learning architectures have been employed in the identification and classification of a wide variety of plants. This research mainly focuses on classifying the medicinal plants that are available in rural areas. To do so, six well-known pre-trained Convolutional Neural Networks (CNN) namely Dense121, InceptionV3, VGG16, Xception, VGG19, and MobileNet, that were trained for the ImageNet dataset, were chosen by implementing Transfer Learning concept. These models were examined with their pre-trained weights for the Rural Medicinal Plant (RMP) dataset that was created using 8 different classes of medicinal plants that sum up to a total of 16000 images. The performance of these models was improved by training through two state-of-the-art Deep Learning optimizers namely, Stochastic Gradient Descent (SGD) and Adam. These models were trained using Keras with a TensorFlow backend. A comparative evaluation was made for these models to identify the model that attains the best classification. The research concluded that for RMP dataset, the MobileNet architecture, in which the training performance was improved with the SGD optimizer is the best suited model to classify medicinal plants and thus proves the novelty of this research. Therefore, the proposed model can be used by traditional medicine practitioners for the identification and classification of medicinal plants.
{"title":"A Comparative Study on Plant Classification Performance using Deep Learning Optimizers","authors":"Sai Kumar T S, Prabalakshmi A, A. K, S. Alagammal","doi":"10.1109/ETI4.051663.2021.9619238","DOIUrl":"https://doi.org/10.1109/ETI4.051663.2021.9619238","url":null,"abstract":"Recently, many Deep Learning architectures have been employed in the identification and classification of a wide variety of plants. This research mainly focuses on classifying the medicinal plants that are available in rural areas. To do so, six well-known pre-trained Convolutional Neural Networks (CNN) namely Dense121, InceptionV3, VGG16, Xception, VGG19, and MobileNet, that were trained for the ImageNet dataset, were chosen by implementing Transfer Learning concept. These models were examined with their pre-trained weights for the Rural Medicinal Plant (RMP) dataset that was created using 8 different classes of medicinal plants that sum up to a total of 16000 images. The performance of these models was improved by training through two state-of-the-art Deep Learning optimizers namely, Stochastic Gradient Descent (SGD) and Adam. These models were trained using Keras with a TensorFlow backend. A comparative evaluation was made for these models to identify the model that attains the best classification. The research concluded that for RMP dataset, the MobileNet architecture, in which the training performance was improved with the SGD optimizer is the best suited model to classify medicinal plants and thus proves the novelty of this research. Therefore, the proposed model can be used by traditional medicine practitioners for the identification and classification of medicinal plants.","PeriodicalId":129682,"journal":{"name":"2021 Emerging Trends in Industry 4.0 (ETI 4.0)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116462481","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 : 2021-05-19DOI: 10.1109/ETI4.051663.2021.9619323
D. J. Krishna Kishore, Maher Rashad Mohamed, K. Sudhakar, S. Jewaliddin, K. Peddakapu, P. S. Rao
Photo-voltaic (PV) is one of the most abundant sources on the earth for the generation of electricity. Although, due to the stochastic nature of PV characteristics to sustain constant power, an accurate PV power prediction is needed for a grid-connected PV system. The proposed model of support vector machine (SVM) with improved dragonfly algorithm(IDA) is used to forecast the PV power. Previously, Theexecution can be done by dragonfly algorithm (DA) through adaptive learning factor along with the differential evolution technique. The IDA is used to select the best support vector machine parameters. Eventually, the suggested model provides better performance as compared to the other algorithm such as SVM with dragonfly algorithm(SVM-DA). It is suitable for forecasting ultra-short-term PV power.
{"title":"Ultra-Short-term PV Power Forecasting Based on a Support Vector Machine with Improved Dragonfly Algorithm","authors":"D. J. Krishna Kishore, Maher Rashad Mohamed, K. Sudhakar, S. Jewaliddin, K. Peddakapu, P. S. Rao","doi":"10.1109/ETI4.051663.2021.9619323","DOIUrl":"https://doi.org/10.1109/ETI4.051663.2021.9619323","url":null,"abstract":"Photo-voltaic (PV) is one of the most abundant sources on the earth for the generation of electricity. Although, due to the stochastic nature of PV characteristics to sustain constant power, an accurate PV power prediction is needed for a grid-connected PV system. The proposed model of support vector machine (SVM) with improved dragonfly algorithm(IDA) is used to forecast the PV power. Previously, Theexecution can be done by dragonfly algorithm (DA) through adaptive learning factor along with the differential evolution technique. The IDA is used to select the best support vector machine parameters. Eventually, the suggested model provides better performance as compared to the other algorithm such as SVM with dragonfly algorithm(SVM-DA). It is suitable for forecasting ultra-short-term PV power.","PeriodicalId":129682,"journal":{"name":"2021 Emerging Trends in Industry 4.0 (ETI 4.0)","volume":"38 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115637404","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：