Pub Date : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110789
P. Grakhov, A. Fedin, V. Yasoveev
This publication presents the results of studies of the static characteristics of actuators for displacements (by means of an example of angular displacements) with a bending type of deformation of the active element – a magnetostrictive bilayer plate with one active layer of 49K2F alloy under the total action of prestresses caused by external forces, moments of internal forces that are induced and arise during a temperature change of a plate made of materials with different values of the temperature coefficients of linear expansion. The article shows the influence of ratios of magnitudes and signs of preliminary moments on the nature of the distribution of elastic stresses across the thickness of the active layers. Relationships have been refined to eliminate zones of an undesirable sign in the distribution of elastic stresses. The prospects of using the action of preliminary moments to achieve greater stability of characteristics with changes in the temperature of the magnetostrictive plate are shown.
{"title":"Characteristics of Angular Displacement Actuators Based on Magnetostrictive Plates under the Action of Prestressing","authors":"P. Grakhov, A. Fedin, V. Yasoveev","doi":"10.1109/SmartIndustryCon57312.2023.10110789","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110789","url":null,"abstract":"This publication presents the results of studies of the static characteristics of actuators for displacements (by means of an example of angular displacements) with a bending type of deformation of the active element – a magnetostrictive bilayer plate with one active layer of 49K2F alloy under the total action of prestresses caused by external forces, moments of internal forces that are induced and arise during a temperature change of a plate made of materials with different values of the temperature coefficients of linear expansion. The article shows the influence of ratios of magnitudes and signs of preliminary moments on the nature of the distribution of elastic stresses across the thickness of the active layers. Relationships have been refined to eliminate zones of an undesirable sign in the distribution of elastic stresses. The prospects of using the action of preliminary moments to achieve greater stability of characteristics with changes in the temperature of the magnetostrictive plate are shown.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"16 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127088915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110833
V. Latypova
The design documentation agreement procedure is one of the production process stages. The procedure is usually rather long because it demands the participation of a large group of staff, conducting the complex documentation control execution. During the agreement process errors in design documentation are identified and a review with relevant remarks to finalize it is formed, typically in manual mode. This operation is repeated until all remarks are finally eliminated. In the latter case, the design documentation agreement procedure ends. In existing research attention has been paid to automation and information support in different stages of the agreement procedure. But the issue of automatization of approving person review formation has not been given enough attention. A method of automated review formation, when agreeing on the design documentation, using text mining of approving person remarks and an automated system, implementing this method, is proposed to solve the problem. Approbation of the automated system was carried out in the agreement of design documentation on the example of design models of printed circuit boards of electronic devices for CNC machines for plasma metal cutting. The results of approbation show that the usage of an automated system allows to reduce the time of review preparation by approving persons. This in turn allows to decrease the time of the design documentation agreement process execution and the time of the manufacturing process in general.
{"title":"Design Documentation Agreement Automated System with Approving Person Intelligent Support","authors":"V. Latypova","doi":"10.1109/SmartIndustryCon57312.2023.10110833","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110833","url":null,"abstract":"The design documentation agreement procedure is one of the production process stages. The procedure is usually rather long because it demands the participation of a large group of staff, conducting the complex documentation control execution. During the agreement process errors in design documentation are identified and a review with relevant remarks to finalize it is formed, typically in manual mode. This operation is repeated until all remarks are finally eliminated. In the latter case, the design documentation agreement procedure ends. In existing research attention has been paid to automation and information support in different stages of the agreement procedure. But the issue of automatization of approving person review formation has not been given enough attention. A method of automated review formation, when agreeing on the design documentation, using text mining of approving person remarks and an automated system, implementing this method, is proposed to solve the problem. Approbation of the automated system was carried out in the agreement of design documentation on the example of design models of printed circuit boards of electronic devices for CNC machines for plasma metal cutting. The results of approbation show that the usage of an automated system allows to reduce the time of review preparation by approving persons. This in turn allows to decrease the time of the design documentation agreement process execution and the time of the manufacturing process in general.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127143616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110783
Alexey Poletikin, Nadyr Mengazetdinov, K. Semenkov
The paper presents the results of research on the possibility of using single-board computers in instrumentation and control systems. It is shown on the example of the OPERATOR software platform it is possible to create nature-like structural nodal components for software and hardware complexes on the base of sinlge-board computers. The authors called these node elements as "veurons". The advantages (the reduction of manufacturing costs, complete hardware unification, easy repair and replacement of units) and limitations (need of parallel computational algorithms, moderate network and data flows) of veurons are given. The examples of veuron- based configurations of the top-level of instrumentation and control systems are presented.
{"title":"Nature-Like Structural Node Components for Software and Hardware Complexes on the Basis of Single-Board Computers","authors":"Alexey Poletikin, Nadyr Mengazetdinov, K. Semenkov","doi":"10.1109/SmartIndustryCon57312.2023.10110783","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110783","url":null,"abstract":"The paper presents the results of research on the possibility of using single-board computers in instrumentation and control systems. It is shown on the example of the OPERATOR software platform it is possible to create nature-like structural nodal components for software and hardware complexes on the base of sinlge-board computers. The authors called these node elements as \"veurons\". The advantages (the reduction of manufacturing costs, complete hardware unification, easy repair and replacement of units) and limitations (need of parallel computational algorithms, moderate network and data flows) of veurons are given. The examples of veuron- based configurations of the top-level of instrumentation and control systems are presented.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127213729","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110723
I. R. Abdulveleev, A. Karandaev, E. Khramshina, I. V. Liubimov, S. A. Evdokimov, V. Gasiyarov
The introduction of systems for technical condition monitoring is relevant for high-voltage vacuum circuit breakers (VCB) that switch furnace transformers. Among other parameters, these systems have to perform online monitoring of the temperature of vacuum chambers (modules) integrated into transformer phases. VCBs switch over 100 times a day due to the specifics of the production modes of electric-arc steel-making furnace and ladle-furnace units. This results in the overheating of the VCB contacts and modules and the reduction of the reliability of the unit. This article presents a method and a system of continuous VCB technical condition control based on the measurement of the vacuum chamber temperature and data transfer over the radio channel. The authors detail the functional diagram and the hardware structure of the system. The hardware set based on contact temperature sensors is justified. The system was deployed on the VCB of a 110 kV closed switchgear installed in the arc furnace shop of a metal plant. The authors reviewed the system interface and provided the temperature trends for the modules included in one of the circuit breaker poles. It is concluded that the VCB is in a satisfactory state. The results of the long-term operation of the developed system are analyzed and its technical efficiency is confirmed.
{"title":"Construction Smart System for Monitoring Technical Condition of Furnace Transformer Vacuum Circuit Breaker","authors":"I. R. Abdulveleev, A. Karandaev, E. Khramshina, I. V. Liubimov, S. A. Evdokimov, V. Gasiyarov","doi":"10.1109/SmartIndustryCon57312.2023.10110723","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110723","url":null,"abstract":"The introduction of systems for technical condition monitoring is relevant for high-voltage vacuum circuit breakers (VCB) that switch furnace transformers. Among other parameters, these systems have to perform online monitoring of the temperature of vacuum chambers (modules) integrated into transformer phases. VCBs switch over 100 times a day due to the specifics of the production modes of electric-arc steel-making furnace and ladle-furnace units. This results in the overheating of the VCB contacts and modules and the reduction of the reliability of the unit. This article presents a method and a system of continuous VCB technical condition control based on the measurement of the vacuum chamber temperature and data transfer over the radio channel. The authors detail the functional diagram and the hardware structure of the system. The hardware set based on contact temperature sensors is justified. The system was deployed on the VCB of a 110 kV closed switchgear installed in the arc furnace shop of a metal plant. The authors reviewed the system interface and provided the temperature trends for the modules included in one of the circuit breaker poles. It is concluded that the VCB is in a satisfactory state. The results of the long-term operation of the developed system are analyzed and its technical efficiency is confirmed.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121436650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110815
A. Shishkov, M. Dudkin, V. K. Le, N. Eremin
The article proposes a new algorithm for space vector PWM (SVPWM) with a five-stage switching sequence (SS) providing a neutral point (NP) voltage balance for a three-level neutral point clamped voltage source inverter (3L NPC VSI). The proposed algorithm uses a five-stage SS to provide the lowest level of switching losses in the inverter power switches. The paper considers four combinations for building five-stage SS with different impacts on the NP voltage. In this case, the NP voltage control is done by switching between various five-stage SS combinations. The effectiveness of the proposed algorithm is confirmed on the basis of computer simulation in the MatLab & Simulink environment. Experimental dependences of the relative NP voltage error to the modulation coefficient and cosφ of the load for the proposed and well-known SVPWM algorithms with five-stage switching sequences are provided. The results showed that the proposed control algorithm 2.4 times reduces the maximum level of the relative NP voltage error. This makes it possible to balance the voltage of the three-level power supply and, as a result, improve the operational reliability of the 3L NPC VSI.
{"title":"Neutral Point Voltage Balance Based on Space-Vector PWM with Five-Stage Sequence for Three-Level Voltage Inverter","authors":"A. Shishkov, M. Dudkin, V. K. Le, N. Eremin","doi":"10.1109/SmartIndustryCon57312.2023.10110815","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110815","url":null,"abstract":"The article proposes a new algorithm for space vector PWM (SVPWM) with a five-stage switching sequence (SS) providing a neutral point (NP) voltage balance for a three-level neutral point clamped voltage source inverter (3L NPC VSI). The proposed algorithm uses a five-stage SS to provide the lowest level of switching losses in the inverter power switches. The paper considers four combinations for building five-stage SS with different impacts on the NP voltage. In this case, the NP voltage control is done by switching between various five-stage SS combinations. The effectiveness of the proposed algorithm is confirmed on the basis of computer simulation in the MatLab & Simulink environment. Experimental dependences of the relative NP voltage error to the modulation coefficient and cosφ of the load for the proposed and well-known SVPWM algorithms with five-stage switching sequences are provided. The results showed that the proposed control algorithm 2.4 times reduces the maximum level of the relative NP voltage error. This makes it possible to balance the voltage of the three-level power supply and, as a result, improve the operational reliability of the 3L NPC VSI.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"44 14","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120961760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110816
D. Goncharov, O. Ivashchuk, N. G. Reznikov
The digital transformation of crop production is part of a federal strategy aimed at adapting agricultural areas to climate change and obtaining a consistently high yield through the introduction of Industry 4.0 digital technologies and platform services. The article explores the use of augmented reality technology in agriculture to visualize the results of growth and development modeling of plants, to evaluate and compare adaptation scripts for managing agricultural areas in the terms of the greenhouse effect dynamics. The article highlights efficiency of the use of virtual and augmented reality technologies in the agro-industrial complex as an important tool for information support of decision makers. The article offers a prototype of the devised toolkit to improve the planning of sowing works and achieve high plant yields due to scientifically based zoning of territories.
{"title":"Method for Modeling and Visualization of Agricultural Crops Growth Based on Augmented Reality Technology in Terms of the Greenhouse Effect Dynamics","authors":"D. Goncharov, O. Ivashchuk, N. G. Reznikov","doi":"10.1109/SmartIndustryCon57312.2023.10110816","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110816","url":null,"abstract":"The digital transformation of crop production is part of a federal strategy aimed at adapting agricultural areas to climate change and obtaining a consistently high yield through the introduction of Industry 4.0 digital technologies and platform services. The article explores the use of augmented reality technology in agriculture to visualize the results of growth and development modeling of plants, to evaluate and compare adaptation scripts for managing agricultural areas in the terms of the greenhouse effect dynamics. The article highlights efficiency of the use of virtual and augmented reality technologies in the agro-industrial complex as an important tool for information support of decision makers. The article offers a prototype of the devised toolkit to improve the planning of sowing works and achieve high plant yields due to scientifically based zoning of territories.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127384652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110732
M. Knyazeva, A. Tselykh, Eduard Kotov
Locally integrated energy systems and networks presuppose locally and regionally available energy sources, infrastructure, production and consumption characteristics as well as consumer structures from different sectors, including the transportation system. Scheduling the optimal commissioning process and further repair of such energy stations are among actual problems for locally integrated energy systems. Equipment downtime and commissioning time of both entire regions and individual settlements are important parameters for decision making process. The paper introduces the problem of optimal scheduling of commissioning of energy stations or hubs with respect to temporal variables, downtimes and fines. The proposed model focuses on finding a quasi-Hamiltonian path for a weighted graph, on whose arcs temporal weights and estimates are given while minimizing the total fine or loss under downtimes.
{"title":"A Model for the Optimal Ordering of Weighted Vertices on a Graph for Local Integrated Energy Systems","authors":"M. Knyazeva, A. Tselykh, Eduard Kotov","doi":"10.1109/SmartIndustryCon57312.2023.10110732","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110732","url":null,"abstract":"Locally integrated energy systems and networks presuppose locally and regionally available energy sources, infrastructure, production and consumption characteristics as well as consumer structures from different sectors, including the transportation system. Scheduling the optimal commissioning process and further repair of such energy stations are among actual problems for locally integrated energy systems. Equipment downtime and commissioning time of both entire regions and individual settlements are important parameters for decision making process. The paper introduces the problem of optimal scheduling of commissioning of energy stations or hubs with respect to temporal variables, downtimes and fines. The proposed model focuses on finding a quasi-Hamiltonian path for a weighted graph, on whose arcs temporal weights and estimates are given while minimizing the total fine or loss under downtimes.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130357438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110745
Mikhail P. Sokolov, Pavel A. Manatin, N. Zyulyarkina
This work researches the possibility of applying the concept of a Q-determinant for the parallelization of calculations in the implementation of algorithms to ensure the security of information in systems of connected vehicles. The increasing complexity of connected vehicle systems increases the computational complexity of algorithms used to identify vulnerable nodes in such systems. This requires finding an effective approach to the parallelization of calculations. As an example, Dijkstra’s algorithm is used to search for vulnerable nodes in a network of connected vehicles. The method of designing effective programs based on a Q-determinant concept is described. The results of an experimental study of the developed programs conducted on the "Tornado SUSU" supercomputer are presented. The research makes it possible to conclude that a Q-determinant concept is an effective approach to parallelizing Dijkstra’s algorithm.
{"title":"The Parallelization of Computations for Ensuring Information Security in Connected Vehicle Systems Using Q-Effective Programming: The Example of Dijkstra’s Algorithm","authors":"Mikhail P. Sokolov, Pavel A. Manatin, N. Zyulyarkina","doi":"10.1109/SmartIndustryCon57312.2023.10110745","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110745","url":null,"abstract":"This work researches the possibility of applying the concept of a Q-determinant for the parallelization of calculations in the implementation of algorithms to ensure the security of information in systems of connected vehicles. The increasing complexity of connected vehicle systems increases the computational complexity of algorithms used to identify vulnerable nodes in such systems. This requires finding an effective approach to the parallelization of calculations. As an example, Dijkstra’s algorithm is used to search for vulnerable nodes in a network of connected vehicles. The method of designing effective programs based on a Q-determinant concept is described. The results of an experimental study of the developed programs conducted on the \"Tornado SUSU\" supercomputer are presented. The research makes it possible to conclude that a Q-determinant concept is an effective approach to parallelizing Dijkstra’s algorithm.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130744944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110794
Vitaly Stepanyants, Mantsa Andzhusheva, A. Romanov
Many traffic accidents happen on the roads every day and a lot of them are captured on traffic or dashboard cameras. This data could be used to train machine learning models to predict dangerous situations so that they can be prevented. For that, it should be organized into datasets. Nowadays a limited amount of traffic accident datasets is available and those are not as well annotated as, for example, driving datasets with no accidents, used for training automated vehicles. Following this, our paper presents a review of existing traffic accident datasets. The search was carried out to provide a list of video datasets relevant to the analysis of dangerous situations involving vehicles. For each dataset under consideration, a brief description of the process of data collection and annotation is presented, and the structure and format of videos are analyzed. In addition, the sources of the video, their amount, and the method of splitting the video into fragments are indicated. Where possible, software tools used for video processing are listed. Further, the paper explores existing solutions for dataset development and annotation. Based on the performed analysis, we propose a pipeline for traffic accident dataset development.
{"title":"A Pipeline for Traffic Accident Dataset Development","authors":"Vitaly Stepanyants, Mantsa Andzhusheva, A. Romanov","doi":"10.1109/SmartIndustryCon57312.2023.10110794","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110794","url":null,"abstract":"Many traffic accidents happen on the roads every day and a lot of them are captured on traffic or dashboard cameras. This data could be used to train machine learning models to predict dangerous situations so that they can be prevented. For that, it should be organized into datasets. Nowadays a limited amount of traffic accident datasets is available and those are not as well annotated as, for example, driving datasets with no accidents, used for training automated vehicles. Following this, our paper presents a review of existing traffic accident datasets. The search was carried out to provide a list of video datasets relevant to the analysis of dangerous situations involving vehicles. For each dataset under consideration, a brief description of the process of data collection and annotation is presented, and the structure and format of videos are analyzed. In addition, the sources of the video, their amount, and the method of splitting the video into fragments are indicated. Where possible, software tools used for video processing are listed. Further, the paper explores existing solutions for dataset development and annotation. Based on the performed analysis, we propose a pipeline for traffic accident dataset development.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132468868","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-27DOI: 10.1109/SmartIndustryCon57312.2023.10110736
I. Shishkin, A. N. Grekov
The authors of the work proposed a method for detecting and classifying microplastics and microorganisms in the marine environment using the YOLOv5 deep learning model. The model is trained on a dataset of 300 images collected from the marine environment, which includes microplastics and microorganisms. The images were marked using Label-Studio and the marked data was used to train the YOLOv5 model. The model was then tested on 60 control images to validate its accuracy. The results of the experiment showed that the YOLOv5 model is capable of accurately detecting and classifying microplastics and microorganisms in the marine environment. The YOLOv5 model has the advantage of having a small memory requirement, ability to work in real time, and better background area distinction compared to other models.
{"title":"Implementation of YOLOv5 for Detection and Classification of Microplastics and Microorganisms in Marine Environment","authors":"I. Shishkin, A. N. Grekov","doi":"10.1109/SmartIndustryCon57312.2023.10110736","DOIUrl":"https://doi.org/10.1109/SmartIndustryCon57312.2023.10110736","url":null,"abstract":"The authors of the work proposed a method for detecting and classifying microplastics and microorganisms in the marine environment using the YOLOv5 deep learning model. The model is trained on a dataset of 300 images collected from the marine environment, which includes microplastics and microorganisms. The images were marked using Label-Studio and the marked data was used to train the YOLOv5 model. The model was then tested on 60 control images to validate its accuracy. The results of the experiment showed that the YOLOv5 model is capable of accurately detecting and classifying microplastics and microorganisms in the marine environment. The YOLOv5 model has the advantage of having a small memory requirement, ability to work in real time, and better background area distinction compared to other models.","PeriodicalId":157877,"journal":{"name":"2023 International Russian Smart Industry Conference (SmartIndustryCon)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128779803","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：