Pub Date : 2021-12-10DOI: 10.1109/SMART52563.2021.9676219
M. Verma, R. Ranjan, Rakesh Kumar
Uttarakhand state of India was formed in the year 2000 and simultaneously power sector was unbundled from state electricity board to power generation, transmission, and distribution utilities. Previous years Tariff Orders clearly indicate that Uttarakhand is becoming energy surplus state to energy deficit state from its inception. Despite repeated guidelines from state power regulator, state power utilities need to use more smart technologies and accurate short-term electrical load forecasting in consideration with weather and other parameters for predicting system load with a leading time of one hour to 24 hours, which is necessary for adequate scheduling and operation of power systems. It will also help for working of their electrical infrastructure efficiently, securely, and economically. This paper describes 24-hour-ahead load prediction whose results will give day ahead load forecast for the future day. Artificial Neural Networks is used for creating such algorithm. The ANN is a tool that duplicates the idea of the person’s brain. The ANN is designed and skilled to receive past load and climate information like temperature, humidity, wind speed, precipitation, pressure, and irradiance as input and after calculating correlation between load and meteorological parameters and load and days to get optimized inputs which produce load forecast as its output. ANN provides predicted load with minimum error and Mean Absolute Percent Error (MAPE) is calculated. Considering this work to use such type of short-term scheduling, short term power purchase process and its related suggestions may help state to be profit making organization and make state energy surplus again.
{"title":"Analysis of Day Ahead Electrical Load Forecasting for Uttarakhand using Artificial Neural Network","authors":"M. Verma, R. Ranjan, Rakesh Kumar","doi":"10.1109/SMART52563.2021.9676219","DOIUrl":"https://doi.org/10.1109/SMART52563.2021.9676219","url":null,"abstract":"Uttarakhand state of India was formed in the year 2000 and simultaneously power sector was unbundled from state electricity board to power generation, transmission, and distribution utilities. Previous years Tariff Orders clearly indicate that Uttarakhand is becoming energy surplus state to energy deficit state from its inception. Despite repeated guidelines from state power regulator, state power utilities need to use more smart technologies and accurate short-term electrical load forecasting in consideration with weather and other parameters for predicting system load with a leading time of one hour to 24 hours, which is necessary for adequate scheduling and operation of power systems. It will also help for working of their electrical infrastructure efficiently, securely, and economically. This paper describes 24-hour-ahead load prediction whose results will give day ahead load forecast for the future day. Artificial Neural Networks is used for creating such algorithm. The ANN is a tool that duplicates the idea of the person’s brain. The ANN is designed and skilled to receive past load and climate information like temperature, humidity, wind speed, precipitation, pressure, and irradiance as input and after calculating correlation between load and meteorological parameters and load and days to get optimized inputs which produce load forecast as its output. ANN provides predicted load with minimum error and Mean Absolute Percent Error (MAPE) is calculated. Considering this work to use such type of short-term scheduling, short term power purchase process and its related suggestions may help state to be profit making organization and make state energy surplus again.","PeriodicalId":356096,"journal":{"name":"2021 10th International Conference on System Modeling & Advancement in Research Trends (SMART)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129274555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-10DOI: 10.1109/SMART52563.2021.9676334
E. Sivakumar, A. Anand, S. G. Sarate
Breast cancer is one of the most common causes of death worldwide among women, with good survival rates if detected early. In our work, we compared supervised, semi- supervised and unsupervised learning on the biomedical dataset, Wisconsin Breast Cancer Dataset, to establish the model with the best performance and hence apply for computer aided diagnosis. The metrics used for the same includes performance of the network as well as the ease of implementation, As a result, we hope to close the gap between technology innovation and its implementation in healthcare.
{"title":"Analysis of Machine Learning, Deep Learning, and Artificial Neural Network Approaches for Breast Cancer Classification","authors":"E. Sivakumar, A. Anand, S. G. Sarate","doi":"10.1109/SMART52563.2021.9676334","DOIUrl":"https://doi.org/10.1109/SMART52563.2021.9676334","url":null,"abstract":"Breast cancer is one of the most common causes of death worldwide among women, with good survival rates if detected early. In our work, we compared supervised, semi- supervised and unsupervised learning on the biomedical dataset, Wisconsin Breast Cancer Dataset, to establish the model with the best performance and hence apply for computer aided diagnosis. The metrics used for the same includes performance of the network as well as the ease of implementation, As a result, we hope to close the gap between technology innovation and its implementation in healthcare.","PeriodicalId":356096,"journal":{"name":"2021 10th International Conference on System Modeling & Advancement in Research Trends (SMART)","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124598212","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}
The concept of reuse is addressed in software product line engineering (SPLE) by distinguishing between two types of processes for development: application and domain engineering. In domain engineering (DE), the focus is to define and manage all the valid combinations of reusable artefacts participating in the product line (PL) and the relationships between them. Domain analysis (DA) is the activity in DE, which describes the variability and commonalities in a domain. Instead of being applicable to a single software system, DA is applicable to multiple related software systems. Each DA method supports different quality attributes which are preserved while reuse. In this paper, two DA methods namely domain specific software architecture (DSSA) ad feature oriented domain analysis (FODA) are explained to model common and variable requirements of PL(s). A case study on Automated Teller Machine (ATM) is discussed using these two methods to identify quality attributes supported by these methods. This paper also discuss two different DA methods, which allows reusability for the identification, organization and knowledge modelling regarding the domain solution, so as to provide reuse among each element of domain.
{"title":"Identifying Quality Attributes of FODA and DSSA Methods in Domain Analysis using a Case Study","authors":"Megha Bhushan, Ashok Kumar, P. Samant, Sakshi Bansal, Sharad Tiwari, Arun Negi","doi":"10.1109/SMART52563.2021.9676289","DOIUrl":"https://doi.org/10.1109/SMART52563.2021.9676289","url":null,"abstract":"The concept of reuse is addressed in software product line engineering (SPLE) by distinguishing between two types of processes for development: application and domain engineering. In domain engineering (DE), the focus is to define and manage all the valid combinations of reusable artefacts participating in the product line (PL) and the relationships between them. Domain analysis (DA) is the activity in DE, which describes the variability and commonalities in a domain. Instead of being applicable to a single software system, DA is applicable to multiple related software systems. Each DA method supports different quality attributes which are preserved while reuse. In this paper, two DA methods namely domain specific software architecture (DSSA) ad feature oriented domain analysis (FODA) are explained to model common and variable requirements of PL(s). A case study on Automated Teller Machine (ATM) is discussed using these two methods to identify quality attributes supported by these methods. This paper also discuss two different DA methods, which allows reusability for the identification, organization and knowledge modelling regarding the domain solution, so as to provide reuse among each element of domain.","PeriodicalId":356096,"journal":{"name":"2021 10th International Conference on System Modeling & Advancement in Research Trends (SMART)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130775530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-10DOI: 10.1109/SMART52563.2021.9675305
Rakesh Kumar, R. Ranjan, M. Verma
The reliable and continuous power supply is must for the today’s era where most of works in every human’s life is based on electricity. In Uttarakhand due to increasing requirement of electricity load and various Transmission and Distribution losses and other obstructions, the Power Generation and DISCOMs are working very closer to the energy demand and generation. The generated electricity cannot be stored efficiently, due to this reason so the electrical load is managed by power utilities for a small approach. The Forecasting of electricity is essential for Power Generation, Transmission and Distribution companies. This study is based on Long Term Load Forecasting using Artificial Neural Network. Due to long duration of forecast it is difficult to foreseen off-peak load demand and this study is based on Long Term Electricity Load Forecasting in Uttarakhand State. The data of Population, GDP, Historical Load from 2011 to 2020 is used as input layer in three-layer feed forward neural network for training, validation, and testing. As a new approach the data of renewal energy source (solar power plants, biogas) and State Gas Generation Station, Electric Vehicle and Charging Infrastructure for Electrical Vehicle is used as input data. The forecasting of electricity load in Uttarakhand for long terms is calculated from 2021 to 2030. The Government of Uttarakhand has launched Vision 2030 for Uttarakhand where the main aim is to accelerate economic growth in Uttarakhand by inviting investors and promotion of free waiver policies on long term infrastructure setup.
{"title":"New Approach for Long Term Electricity Load Forecasting for Uttarakhand State Power Utilities using Artificial Neural Network","authors":"Rakesh Kumar, R. Ranjan, M. Verma","doi":"10.1109/SMART52563.2021.9675305","DOIUrl":"https://doi.org/10.1109/SMART52563.2021.9675305","url":null,"abstract":"The reliable and continuous power supply is must for the today’s era where most of works in every human’s life is based on electricity. In Uttarakhand due to increasing requirement of electricity load and various Transmission and Distribution losses and other obstructions, the Power Generation and DISCOMs are working very closer to the energy demand and generation. The generated electricity cannot be stored efficiently, due to this reason so the electrical load is managed by power utilities for a small approach. The Forecasting of electricity is essential for Power Generation, Transmission and Distribution companies. This study is based on Long Term Load Forecasting using Artificial Neural Network. Due to long duration of forecast it is difficult to foreseen off-peak load demand and this study is based on Long Term Electricity Load Forecasting in Uttarakhand State. The data of Population, GDP, Historical Load from 2011 to 2020 is used as input layer in three-layer feed forward neural network for training, validation, and testing. As a new approach the data of renewal energy source (solar power plants, biogas) and State Gas Generation Station, Electric Vehicle and Charging Infrastructure for Electrical Vehicle is used as input data. The forecasting of electricity load in Uttarakhand for long terms is calculated from 2021 to 2030. The Government of Uttarakhand has launched Vision 2030 for Uttarakhand where the main aim is to accelerate economic growth in Uttarakhand by inviting investors and promotion of free waiver policies on long term infrastructure setup.","PeriodicalId":356096,"journal":{"name":"2021 10th International Conference on System Modeling & Advancement in Research Trends (SMART)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130230765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-10DOI: 10.1109/SMART52563.2021.9676257
Yusra Siddiqui, Nupur Mittal, Imran Khan
The optimization and comparison of structure of double-gate MOSFETs and gate-all-around (GAA) MOSFETs was carried out. The fin width to gate length ratio and SCE (short channel effects) were discussed and studied. The 3-D simulations affirmed that while gate length was same as fin width, the short channel effects were inhibited. The ratio of the fin width to the gate length was maximized up to 1.2 in cylindrical channel GAA MOSFETs as compared to cubical channel ones.
{"title":"Performance Analysis and Characterization of Double Gate and Gate All Around MOSFET","authors":"Yusra Siddiqui, Nupur Mittal, Imran Khan","doi":"10.1109/SMART52563.2021.9676257","DOIUrl":"https://doi.org/10.1109/SMART52563.2021.9676257","url":null,"abstract":"The optimization and comparison of structure of double-gate MOSFETs and gate-all-around (GAA) MOSFETs was carried out. The fin width to gate length ratio and SCE (short channel effects) were discussed and studied. The 3-D simulations affirmed that while gate length was same as fin width, the short channel effects were inhibited. The ratio of the fin width to the gate length was maximized up to 1.2 in cylindrical channel GAA MOSFETs as compared to cubical channel ones.","PeriodicalId":356096,"journal":{"name":"2021 10th International Conference on System Modeling & Advancement in Research Trends (SMART)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121162570","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-10DOI: 10.1109/SMART52563.2021.9676250
A. Choudhary, G. Raj, A. Agrawal, Hemant Sawhney, P. Nand, Deepak Bhargava
In 20th century, Dental Caries have become a major health issue globally. According to WHO, 2.3 billion adults and 530 million children are suffering from dental caries-related issues. This problem can be controlled by early accurate detection and treatments. There exist many approaches in the literature to classify dental caries. But accuracy of these approaches is still a challenge. This paper proposes an effective approach using convolutional neural networks by adopting VGG16 and VGG19 models. The patient’s X-Ray images have been collected and labeled. The proposed models have been compared on the collected datasets. The results over this dataset indicate the superiority of VGG19 based model with 95% accuracy as compared to VGG16 based model with 91% accuracy.
{"title":"An Effective Approach for Classification of Dental Caries using Convolutional Neural Networks","authors":"A. Choudhary, G. Raj, A. Agrawal, Hemant Sawhney, P. Nand, Deepak Bhargava","doi":"10.1109/SMART52563.2021.9676250","DOIUrl":"https://doi.org/10.1109/SMART52563.2021.9676250","url":null,"abstract":"In 20th century, Dental Caries have become a major health issue globally. According to WHO, 2.3 billion adults and 530 million children are suffering from dental caries-related issues. This problem can be controlled by early accurate detection and treatments. There exist many approaches in the literature to classify dental caries. But accuracy of these approaches is still a challenge. This paper proposes an effective approach using convolutional neural networks by adopting VGG16 and VGG19 models. The patient’s X-Ray images have been collected and labeled. The proposed models have been compared on the collected datasets. The results over this dataset indicate the superiority of VGG19 based model with 95% accuracy as compared to VGG16 based model with 91% accuracy.","PeriodicalId":356096,"journal":{"name":"2021 10th International Conference on System Modeling & Advancement in Research Trends (SMART)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131036575","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-10DOI: 10.1109/SMART52563.2021.9676332
Saloni Manhas
Web technologies are framed for the purpose of catering the need of ubiquitousness. There is no doubt that web applications are providing number of advantages to the masses, but everything comes with certain vulnerabilities. Exploitation of these vulnerabilities can change the game completely by providing fatal results, instead of giving fruitful results. Cross site scripting attack is also a result of mishandling of vulnerabilities located in web applications. In this paper, XENOTIX framework from OWASP has been used for the detection of cross site scripting attack and practices to curb XSS are discussed.
{"title":"Ontology of XSS Vulnerabilities and its Detection using XENOTIX Framework","authors":"Saloni Manhas","doi":"10.1109/SMART52563.2021.9676332","DOIUrl":"https://doi.org/10.1109/SMART52563.2021.9676332","url":null,"abstract":"Web technologies are framed for the purpose of catering the need of ubiquitousness. There is no doubt that web applications are providing number of advantages to the masses, but everything comes with certain vulnerabilities. Exploitation of these vulnerabilities can change the game completely by providing fatal results, instead of giving fruitful results. Cross site scripting attack is also a result of mishandling of vulnerabilities located in web applications. In this paper, XENOTIX framework from OWASP has been used for the detection of cross site scripting attack and practices to curb XSS are discussed.","PeriodicalId":356096,"journal":{"name":"2021 10th International Conference on System Modeling & Advancement in Research Trends (SMART)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128123032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-10DOI: 10.1109/SMART52563.2021.9676261
S. Rani, Swathi Gowroju, Sandeep Kumar
This paper gives a window browsing of the iris technology, its application areas and the spoofing attacks suggested so far from its initiation. Iris recognition and identification algorithms used various patterns and mathematical models to identify humans for various applications i.e., IoT, POS (Point of Sale), passport, health care digital transformation, child trafficking, liveness detection. This paper gives scope to a comparative survey of the literature of different authors who worked on different acquisition methods, localization and normalization methods, and different spoofing attacks presented and registered in the analysis.
{"title":"IRIS based Recognition and Spoofing Attacks: A Review","authors":"S. Rani, Swathi Gowroju, Sandeep Kumar","doi":"10.1109/SMART52563.2021.9676261","DOIUrl":"https://doi.org/10.1109/SMART52563.2021.9676261","url":null,"abstract":"This paper gives a window browsing of the iris technology, its application areas and the spoofing attacks suggested so far from its initiation. Iris recognition and identification algorithms used various patterns and mathematical models to identify humans for various applications i.e., IoT, POS (Point of Sale), passport, health care digital transformation, child trafficking, liveness detection. This paper gives scope to a comparative survey of the literature of different authors who worked on different acquisition methods, localization and normalization methods, and different spoofing attacks presented and registered in the analysis.","PeriodicalId":356096,"journal":{"name":"2021 10th International Conference on System Modeling & Advancement in Research Trends (SMART)","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122498754","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-10DOI: 10.1109/SMART52563.2021.9676256
S. Anusuya, K. Sharmila
Obesity is a perilous consumer of human lives, and is addressed with growing concerns globally. One of the primary reasons for the origination of surgical studies in Bariatrics is the consumption of unhealthy and indolent practices. Multifaceted literary studies are associated with gremlins of the human body, along with food calorie recognition methods. Most commonly many of the health hazards arise with food regimen that individuals choose to consume. Therefore, identification and anatomization of the food and calorie intake is a cardinal aspect which requires meticulous approaches. The whilom approaches relating to food calorie identification and segmentation have been implemented with K-means clustering and color space segmentation approaches. However, this study focuses on the food image enhancement, feature identification and clustering using MSERF and SURF detection parameters. The proposed work also ensures that the implemented work forms a strong pre-processing method to better accuracy of classification for further stages of study. The indagated study is simulated using MATLAB and the results are successfully acquired.
{"title":"K-Means Food Object Clustering and Feature Detection using MSERF and SURF Region Points","authors":"S. Anusuya, K. Sharmila","doi":"10.1109/SMART52563.2021.9676256","DOIUrl":"https://doi.org/10.1109/SMART52563.2021.9676256","url":null,"abstract":"Obesity is a perilous consumer of human lives, and is addressed with growing concerns globally. One of the primary reasons for the origination of surgical studies in Bariatrics is the consumption of unhealthy and indolent practices. Multifaceted literary studies are associated with gremlins of the human body, along with food calorie recognition methods. Most commonly many of the health hazards arise with food regimen that individuals choose to consume. Therefore, identification and anatomization of the food and calorie intake is a cardinal aspect which requires meticulous approaches. The whilom approaches relating to food calorie identification and segmentation have been implemented with K-means clustering and color space segmentation approaches. However, this study focuses on the food image enhancement, feature identification and clustering using MSERF and SURF detection parameters. The proposed work also ensures that the implemented work forms a strong pre-processing method to better accuracy of classification for further stages of study. The indagated study is simulated using MATLAB and the results are successfully acquired.","PeriodicalId":356096,"journal":{"name":"2021 10th International Conference on System Modeling & Advancement in Research Trends (SMART)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127912298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-10DOI: 10.1109/SMART52563.2021.9676263
Jyoti Ranjan Labh, R. Dwivedi
For machine learning applications, digital image production provides for the efficient generation of huge volumes of training data while preserving control over the generation process to ensure the optimal content distribution and variation. Synthetic data has the potential to become an important element of the training pipeline as the demand for deep learning applications grows. Over the last decade, a broad range of strategies for producing training data have been presented. The collecting of these for comparison and categorization is required for future improvement. This study presents a complete list of available visual machine learning image synthesis approaches. In the context of 2D picture production, these are classed as light transfer and colour transfer. The focus is on the computational features of approaches for developing machine learning colour transfer between image-to-image translation in the future. Finally, the learning potential of each approach is assessed based on its reported quality and performance. The study is meant to serve as a complete reference for both data and application developers. This is a comprehensive list of all the methods and approaches discussed in this page.
{"title":"Extensive Study on Color and Light Translation of 2D Images using Machine Learning Approaches","authors":"Jyoti Ranjan Labh, R. Dwivedi","doi":"10.1109/SMART52563.2021.9676263","DOIUrl":"https://doi.org/10.1109/SMART52563.2021.9676263","url":null,"abstract":"For machine learning applications, digital image production provides for the efficient generation of huge volumes of training data while preserving control over the generation process to ensure the optimal content distribution and variation. Synthetic data has the potential to become an important element of the training pipeline as the demand for deep learning applications grows. Over the last decade, a broad range of strategies for producing training data have been presented. The collecting of these for comparison and categorization is required for future improvement. This study presents a complete list of available visual machine learning image synthesis approaches. In the context of 2D picture production, these are classed as light transfer and colour transfer. The focus is on the computational features of approaches for developing machine learning colour transfer between image-to-image translation in the future. Finally, the learning potential of each approach is assessed based on its reported quality and performance. The study is meant to serve as a complete reference for both data and application developers. This is a comprehensive list of all the methods and approaches discussed in this page.","PeriodicalId":356096,"journal":{"name":"2021 10th International Conference on System Modeling & Advancement in Research Trends (SMART)","volume":"543 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128646522","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：