Pub Date : 2025-04-28DOI: 10.1016/j.sasc.2025.200257
Mengying Ma , Chengdi Xu , Junfeng Han
In recent years, the rapid modernization and increasing adoption of smart homes have disrupted the traditional balance between electrical design standards, power line systems, and fire monitoring frameworks. This disruption has heightened electrical safety risks in residential buildings, endangering both lives and property. This study introduces an edge response delay calculation model using the Modbus protocol and a fine-grained distributed edge node networking architecture to enhance system efficiency. This research examines the platform's effectiveness in improving electrical safety and accident prevention in civil buildings, focusing on three key aspects: design concept, system architecture, and implementation. The platform seamlessly integrates electrical fire monitoring, IoT technology, and digital building simulation, enabling an intelligent early warning system for pre-disaster detection and an automated post-disaster response mechanism. The results show that the system enhances the pre-disaster early warning capability in the electrical fire business scenario, and provides efficient decision-making support for personnel escape evacuation and fire rescue in the post-disaster stage, which has important practical application value.
{"title":"Application of an intelligent electrical fire monitoring system based on the EC-IOT framework in high-rise residential buildings","authors":"Mengying Ma , Chengdi Xu , Junfeng Han","doi":"10.1016/j.sasc.2025.200257","DOIUrl":"10.1016/j.sasc.2025.200257","url":null,"abstract":"<div><div>In recent years, the rapid modernization and increasing adoption of smart homes have disrupted the traditional balance between electrical design standards, power line systems, and fire monitoring frameworks. This disruption has heightened electrical safety risks in residential buildings, endangering both lives and property. This study introduces an edge response delay calculation model using the Modbus protocol and a fine-grained distributed edge node networking architecture to enhance system efficiency. This research examines the platform's effectiveness in improving electrical safety and accident prevention in civil buildings, focusing on three key aspects: design concept, system architecture, and implementation. The platform seamlessly integrates electrical fire monitoring, IoT technology, and digital building simulation, enabling an intelligent early warning system for pre-disaster detection and an automated post-disaster response mechanism. The results show that the system enhances the pre-disaster early warning capability in the electrical fire business scenario, and provides efficient decision-making support for personnel escape evacuation and fire rescue in the post-disaster stage, which has important practical application value.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200257"},"PeriodicalIF":0.0,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143898738","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 : 2025-04-28DOI: 10.1016/j.sasc.2025.200262
Yan Luo, Zhibin Nie
The emphasis on physical health has grown significantly in recent years. As future contributors to society, students' physical health deserves greater attention. Therefore, it is necessary to strengthen research on managing students’ physical health information, in order to establish a representative, scientific, practical, and operable information management (IM) model. This is highly significant for the scientific assessment and management of students’ physical health in practice. With the rapid growth of information, managing students' physical health now involves handling vast amounts of data, and managing these data relies on applying big data. In view of the problem of low validity of students' physical health information assessment results and difficulty in timely improvement of physical health status, this article constructed a visual, real-time, and comprehensive student physical health IM model using big data. Students' physical health was evaluated using multiple Gaussian distributions, ensuring data reliability, systematic management, comprehensive analysis, and real-time feedback, thereby effectively improving the effectiveness and practical guidance of students’ physical health management results. The experimental results of this article indicated that before the experiment, there were 20 and 19 students in the control group and the experimental group who failed in physical health, and 4 and 5 students in the two groups who had excellent physical health, respectively. After the experiment, there were 15 students in the control group and 8 students in the experimental group who failed in physical health, while 6 and 16 students in excellent physical health. The results showed a significant increase in the number of students with excellent physical health in the experimental group, demonstrating the effectiveness of the proposed big data-based management model. This indicated that by managing student physical health information in the big data environment, students’ physical health can be effectively understood and improved.
{"title":"A Model for Building Student Physical Health Information Management in a Big Data Environment","authors":"Yan Luo, Zhibin Nie","doi":"10.1016/j.sasc.2025.200262","DOIUrl":"10.1016/j.sasc.2025.200262","url":null,"abstract":"<div><div>The emphasis on physical health has grown significantly in recent years. As future contributors to society, students' physical health deserves greater attention. Therefore, it is necessary to strengthen research on managing students’ physical health information, in order to establish a representative, scientific, practical, and operable information management (IM) model. This is highly significant for the scientific assessment and management of students’ physical health in practice. With the rapid growth of information, managing students' physical health now involves handling vast amounts of data, and managing these data relies on applying big data. In view of the problem of low validity of students' physical health information assessment results and difficulty in timely improvement of physical health status, this article constructed a visual, real-time, and comprehensive student physical health IM model using big data. Students' physical health was evaluated using multiple Gaussian distributions, ensuring data reliability, systematic management, comprehensive analysis, and real-time feedback, thereby effectively improving the effectiveness and practical guidance of students’ physical health management results. The experimental results of this article indicated that before the experiment, there were 20 and 19 students in the control group and the experimental group who failed in physical health, and 4 and 5 students in the two groups who had excellent physical health, respectively. After the experiment, there were 15 students in the control group and 8 students in the experimental group who failed in physical health, while 6 and 16 students in excellent physical health. The results showed a significant increase in the number of students with excellent physical health in the experimental group, demonstrating the effectiveness of the proposed big data-based management model. This indicated that by managing student physical health information in the big data environment, students’ physical health can be effectively understood and improved.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200262"},"PeriodicalIF":0.0,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143947679","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 : 2025-04-28DOI: 10.1016/j.sasc.2025.200251
Shu Li
The scheduling process of music courses in education is complex and difficult to optimize. Traditional scheduling systems usually use simple algorithms or manual intervention, resulting in low efficiency and uneven resource allocation. To optimize the resource allocation and course scheduling of music courses, considering the limitations of genetic algorithms, the randomness and traversal characteristics of introducing chaotic systems were studied to optimize population diversity, forming a new scheduling method based on chaotic genetic algorithms. This study used music course data from a particular school, including classroom resources, number of students, course time, etc. The results showed that after 300 iterations, the average running time of the research method decreased by 76.57 %, 66.46 %, 58.39 %, and 48.24 %, respectively. Meanwhile, this research method not only had the fastest convergence speed, but also had the highest fitness function value during the convergence process. In practical applications, this research method significantly improved students' music grades, demonstrating its effectiveness in optimizing the music course scheduling system. This study provides a new research direction for future educational scheduling systems.
{"title":"Optimizing music course scheduling with real number encoding and chaos genetic algorithm","authors":"Shu Li","doi":"10.1016/j.sasc.2025.200251","DOIUrl":"10.1016/j.sasc.2025.200251","url":null,"abstract":"<div><div>The scheduling process of music courses in education is complex and difficult to optimize. Traditional scheduling systems usually use simple algorithms or manual intervention, resulting in low efficiency and uneven resource allocation. To optimize the resource allocation and course scheduling of music courses, considering the limitations of genetic algorithms, the randomness and traversal characteristics of introducing chaotic systems were studied to optimize population diversity, forming a new scheduling method based on chaotic genetic algorithms. This study used music course data from a particular school, including classroom resources, number of students, course time, etc. The results showed that after 300 iterations, the average running time of the research method decreased by 76.57 %, 66.46 %, 58.39 %, and 48.24 %, respectively. Meanwhile, this research method not only had the fastest convergence speed, but also had the highest fitness function value during the convergence process. In practical applications, this research method significantly improved students' music grades, demonstrating its effectiveness in optimizing the music course scheduling system. This study provides a new research direction for future educational scheduling systems.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200251"},"PeriodicalIF":0.0,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143898739","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 : 2025-04-28DOI: 10.1016/j.sasc.2025.200254
Zhijie Peng , Lili Yin
This research presents a novel nonlinear prediction model for Internet of Things (IoT) driven vehicle network traffic management. Current traffic prediction systems use linear models that do not characterize the highly nonlinear urban traffic dynamics. We integrate real-time IoT sensor data with a dual-layer long short-term memory (LSTM) neural network architecture optimised for traffic prediction. System architecture consists of three spatially separated layers: IoT sensor network for data collection, real-time data processing pipeline and the user interface for visualization. The predictive accuracy in terms of Mean Squared Error (0.0842), Mean Absolute Error (0.0623) and the R² score (0.9187) was better on average for 35 strategic urban sites at 6 months. It achieved a 92 % prediction accuracy during morning peak hours and maintained response times <200 ms for 98.5 % of predictions under any load conditions. The system resilience testing involved 99.95 % uptime with robust operation even with 15 % of the sensors failing. Challenges with extreme weather conditions and data gaps still exist; however, this research contributes to theoretical understanding of nonlinear traffic dynamics and practical applications for smart city development. While the system presented here paves the way for more intelligent, adaptive solutions to Urban Mobility to reduce congestion significantly and improve traffic management efficiency, there still exist issues regarding the acquisition of traffic data, the phenomenon of commuting behavior, and only rudimentary efforts to mathematically model passenger exposure.
{"title":"Nonlinear prediction model of vehicle network traffic management based on the internet of things","authors":"Zhijie Peng , Lili Yin","doi":"10.1016/j.sasc.2025.200254","DOIUrl":"10.1016/j.sasc.2025.200254","url":null,"abstract":"<div><div>This research presents a novel nonlinear prediction model for Internet of Things (IoT) driven vehicle network traffic management. Current traffic prediction systems use linear models that do not characterize the highly nonlinear urban traffic dynamics. We integrate real-time IoT sensor data with a dual-layer long short-term memory (LSTM) neural network architecture optimised for traffic prediction. System architecture consists of three spatially separated layers: IoT sensor network for data collection, real-time data processing pipeline and the user interface for visualization. The predictive accuracy in terms of Mean Squared Error (0.0842), Mean Absolute Error (0.0623) and the R² score (0.9187) was better on average for 35 strategic urban sites at 6 months. It achieved a 92 % prediction accuracy during morning peak hours and maintained response times <200 ms for 98.5 % of predictions under any load conditions. The system resilience testing involved 99.95 % uptime with robust operation even with 15 % of the sensors failing. Challenges with extreme weather conditions and data gaps still exist; however, this research contributes to theoretical understanding of nonlinear traffic dynamics and practical applications for smart city development. While the system presented here paves the way for more intelligent, adaptive solutions to Urban Mobility to reduce congestion significantly and improve traffic management efficiency, there still exist issues regarding the acquisition of traffic data, the phenomenon of commuting behavior, and only rudimentary efforts to mathematically model passenger exposure.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200254"},"PeriodicalIF":0.0,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143932196","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 : 2025-04-27DOI: 10.1016/j.sasc.2025.200260
MHM Ali, Mostafa R. A. Atia, Moustafa A. Fouz
Involving human robot collaboration advancements are transforming industrial safety protocols. This paper proposes an enhancement approach to improve robot decision accuracy, in Speed and Separation Monitoring (SSM), according to ISO/TS 15066 safety standard. This approach integrates Machine Learning (ML), Artificial Intelligence (AI), for decision-making using data extracted from an active depth camera, which tracks operators’ hand movements and measures distances on line. The developed algorithm enables the robot to make decisions based on protective separation distance (PSD) and dynamic separation distances (DSDs). A test rig developed to determine separation distances required across four zones for safe pick-and-place application. The result shows that the defined thresholds enhances both safety and operation efficiency. This creates a suitable collaborative environment for the operator, and makes the task easier to perform.
{"title":"Decision enhancement of speed and separating monitoring modes for human-robot collaborative safety","authors":"MHM Ali, Mostafa R. A. Atia, Moustafa A. Fouz","doi":"10.1016/j.sasc.2025.200260","DOIUrl":"10.1016/j.sasc.2025.200260","url":null,"abstract":"<div><div>Involving human robot collaboration advancements are transforming industrial safety protocols. This paper proposes an enhancement approach to improve robot decision accuracy, in Speed and Separation Monitoring (SSM), according to ISO/TS 15066 safety standard. This approach integrates Machine Learning (ML), Artificial Intelligence (AI), for decision-making using data extracted from an active depth camera, which tracks operators’ hand movements and measures distances on line. The developed algorithm enables the robot to make decisions based on protective separation distance (PSD) and dynamic separation distances (DSDs). A test rig developed to determine separation distances required across four zones for safe pick-and-place application. The result shows that the defined thresholds enhances both safety and operation efficiency. This creates a suitable collaborative environment for the operator, and makes the task easier to perform.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200260"},"PeriodicalIF":0.0,"publicationDate":"2025-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143895588","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 : 2025-04-27DOI: 10.1016/j.sasc.2025.200266
Xingxing Fu
With the continuous development of intelligent computer technology, its application in the field of art and design is becoming increasingly widespread. However, the current application of artificial intelligence technology in folk art design is relatively simple and lacks systematic analysis methods. Therefore, this study aims to construct an optimization model through the use of advanced computational science methods and digital media technology to achieve a comprehensive analysis of various indicators of folk art design. This study uses watermark verification theory and digital models to analyze folk art design, and constructs corresponding optimization models through parameter encoded verification curves and calculations of different indicators. This model can conduct targeted analysis for different indicators to obtain corresponding calculation results. The experimental results show that the model calculation results exhibit relatively significant fluctuations in design concepts and public aesthetics, indicating that these two factors have a significant impact on the model. The design strategy presents a linear variation, with a relatively small range of changes in overall aesthetics and color elements, and a relatively small impact on the calculation results. In addition, the validation coefficient has an effect when the independent variable is small, while larger independent variables have an inhibitory effect on the validation coefficient. This study achieved a comprehensive analysis of various indicators of folk art design by constructing an optimization model. The effectiveness and accuracy of the model have been verified through comparison with experimental data.
{"title":"Combined digital media technology and advanced computing science in folk art design application","authors":"Xingxing Fu","doi":"10.1016/j.sasc.2025.200266","DOIUrl":"10.1016/j.sasc.2025.200266","url":null,"abstract":"<div><div>With the continuous development of intelligent computer technology, its application in the field of art and design is becoming increasingly widespread. However, the current application of artificial intelligence technology in folk art design is relatively simple and lacks systematic analysis methods. Therefore, this study aims to construct an optimization model through the use of advanced computational science methods and digital media technology to achieve a comprehensive analysis of various indicators of folk art design. This study uses watermark verification theory and digital models to analyze folk art design, and constructs corresponding optimization models through parameter encoded verification curves and calculations of different indicators. This model can conduct targeted analysis for different indicators to obtain corresponding calculation results. The experimental results show that the model calculation results exhibit relatively significant fluctuations in design concepts and public aesthetics, indicating that these two factors have a significant impact on the model. The design strategy presents a linear variation, with a relatively small range of changes in overall aesthetics and color elements, and a relatively small impact on the calculation results. In addition, the validation coefficient has an effect when the independent variable is small, while larger independent variables have an inhibitory effect on the validation coefficient. This study achieved a comprehensive analysis of various indicators of folk art design by constructing an optimization model. The effectiveness and accuracy of the model have been verified through comparison with experimental data.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200266"},"PeriodicalIF":0.0,"publicationDate":"2025-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143895590","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 : 2025-04-25DOI: 10.1016/j.sasc.2025.200265
Yaru Li
With the extensive application of information technology in education, physical education teaching is gradually optimized and improved using data-driven methods. This paper focuses on constructing and optimizing the physical education teaching analysis platform by using the two-way long and short-term memory network technology. The study collected multi-dimensional physical education data from >500 students, and conducted in-depth analysis through the Bi-LSTM model, aiming to improve the accuracy of teaching evaluation. The results show that the platform has achieved significant progress in the automatic scoring system, and the scoring accuracy has increased to 92 %, a 20 % improvement compared with the traditional methods. The platform can also accurately predict the physical improvement of students, with an accuracy of 85 %, and real-time analysis of skills to master the progress and sports risks, providing strong support for personalized teaching. These results not only enhance the objectivity of physical education evaluation, but also provide teachers with rich data insight and help them to develop more scientific and personalized teaching strategies.
{"title":"Research on the construction and optimization of physical education teaching analysis platform based on Bi-LSTM model","authors":"Yaru Li","doi":"10.1016/j.sasc.2025.200265","DOIUrl":"10.1016/j.sasc.2025.200265","url":null,"abstract":"<div><div>With the extensive application of information technology in education, physical education teaching is gradually optimized and improved using data-driven methods. This paper focuses on constructing and optimizing the physical education teaching analysis platform by using the two-way long and short-term memory network technology. The study collected multi-dimensional physical education data from >500 students, and conducted in-depth analysis through the Bi-LSTM model, aiming to improve the accuracy of teaching evaluation. The results show that the platform has achieved significant progress in the automatic scoring system, and the scoring accuracy has increased to 92 %, a 20 % improvement compared with the traditional methods. The platform can also accurately predict the physical improvement of students, with an accuracy of 85 %, and real-time analysis of skills to master the progress and sports risks, providing strong support for personalized teaching. These results not only enhance the objectivity of physical education evaluation, but also provide teachers with rich data insight and help them to develop more scientific and personalized teaching strategies.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200265"},"PeriodicalIF":0.0,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144072742","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 : 2025-04-25DOI: 10.1016/j.sasc.2025.200252
Jing Zhang, Ningyu Hu
To solve the high cost and low accuracy in facial recognition system, a facial recognition system based on deep learning algorithm is designed in this paper. First, the YOLO model is improved by introducing the EfficientNet to enhance the performance of the facial detection model. Second, a feature extraction model based on the loss function of the improved FaceNet is constructed. In the medium test dataset validation, the proposed facial detection model improved the detection accuracy by an average of 26.30 % compared with the YOLOv3 series models. The LFW dataset validation showed that the model achieved 99.54 % accuracy after 90,000 iterations, which was 1.59 % higher than the average of other models. In the mixed dataset, the proposed facial recognition system improved the accuracy by 4.76 % and 8.64 % compared with the existing mainstream systems, respectively. The system shows strong robustness in diverse scenarios with different skin colors, ages, facial occlusions, and expressions. The designed facial detection method has high detection efficiency, and the feature extraction model has superior recognition results. The system can provide real-time recognition in complex scenes such as facial occlusion, meeting real-time requirements.
{"title":"Accuracy and robustness evaluation of deep learning algorithms in facial recognition systems","authors":"Jing Zhang, Ningyu Hu","doi":"10.1016/j.sasc.2025.200252","DOIUrl":"10.1016/j.sasc.2025.200252","url":null,"abstract":"<div><div>To solve the high cost and low accuracy in facial recognition system, a facial recognition system based on deep learning algorithm is designed in this paper. First, the YOLO model is improved by introducing the EfficientNet to enhance the performance of the facial detection model. Second, a feature extraction model based on the loss function of the improved FaceNet is constructed. In the medium test dataset validation, the proposed facial detection model improved the detection accuracy by an average of 26.30 % compared with the YOLOv3 series models. The LFW dataset validation showed that the model achieved 99.54 % accuracy after 90,000 iterations, which was 1.59 % higher than the average of other models. In the mixed dataset, the proposed facial recognition system improved the accuracy by 4.76 % and 8.64 % compared with the existing mainstream systems, respectively. The system shows strong robustness in diverse scenarios with different skin colors, ages, facial occlusions, and expressions. The designed facial detection method has high detection efficiency, and the feature extraction model has superior recognition results. The system can provide real-time recognition in complex scenes such as facial occlusion, meeting real-time requirements.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200252"},"PeriodicalIF":0.0,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143898816","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 : 2025-04-25DOI: 10.1016/j.sasc.2025.200263
Qiang Liu, Xi Zheng, Qiuhan Zhang, Hongjie Sun, Jun Yan
Maintaining high standards in wire installation for overhead transmission lines is vital for the dependability and safety of power systems. Traditional inspection techniques depend on manual evaluations, which are subjective and entail considerable safety hazards for workers. To tackle these issues, this paper suggests an automated wire defect detection approach utilizing image recognition, incorporated into an intelligent wire installation quality robot. The system uses a Scale-Invariant Feature Transform (SIFT) algorithm to precisely identify defect markers by initially extracting the texture features of standard wires and subsequently identifying variations that indicate faults. This approach improves defect detection by using optical imaging and real-time processing, ensuring resilience against differing environmental conditions. Tests conducted on various datasets demonstrated a missed detection rate of 4.2 %, a misjudgment rate of 3.5 %, and an overall detection accuracy of 92.3 %. These results substantiate the proposed method’s ability to enhance the automation and reliability of wire installation quality evaluation.
{"title":"Defect identification method for overhead transmission lines based on SIFT algorithm","authors":"Qiang Liu, Xi Zheng, Qiuhan Zhang, Hongjie Sun, Jun Yan","doi":"10.1016/j.sasc.2025.200263","DOIUrl":"10.1016/j.sasc.2025.200263","url":null,"abstract":"<div><div>Maintaining high standards in wire installation for overhead transmission lines is vital for the dependability and safety of power systems. Traditional inspection techniques depend on manual evaluations, which are subjective and entail considerable safety hazards for workers. To tackle these issues, this paper suggests an automated wire defect detection approach utilizing image recognition, incorporated into an intelligent wire installation quality robot. The system uses a Scale-Invariant Feature Transform (SIFT) algorithm to precisely identify defect markers by initially extracting the texture features of standard wires and subsequently identifying variations that indicate faults. This approach improves defect detection by using optical imaging and real-time processing, ensuring resilience against differing environmental conditions. Tests conducted on various datasets demonstrated a missed detection rate of 4.2 %, a misjudgment rate of 3.5 %, and an overall detection accuracy of 92.3 %. These results substantiate the proposed method’s ability to enhance the automation and reliability of wire installation quality evaluation.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200263"},"PeriodicalIF":0.0,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143890968","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 : 2025-04-24DOI: 10.1016/j.sasc.2025.200258
Mequanent Degu Belete, Girma Kassa Alitasb
Hate speech has been disseminated more frequently on social media sites like Facebook in recent years. On Facebook, hate speech can proliferate through text, image, or video. We suggested a deep learning approach to identify offensive memes posted on Facebook in case of Amharic language'. The research process commenced by manually gathering memes posted by Facebook users. Next came textual data extraction, annotation, preprocessing, splitting, feature extraction, model development and assessment Amharic OCRs were employed to extract textual data. Character normalization, stop word removal, and unnecessary character removal make up the text-preprocessing step. Using Stratified KFold the textual dataset is split into the train set (80 %), the validation set (10 %) and the test set (10 %). Vectors are created from the preprocessed texts using the Bog of words (BOW), TFIDF and word embeddings. Following that, the vectors are fed into Machine learning algorithms: NB, DT, RF, KNN, LSVM and LR, and deep learning models that are based on Dense, BiGRU, and BiLSTM algorithms. The model with the optimal parameters is chosen after numerous experiments. With an accuracy rate of 94 %, the BiLSTM + Dense model, the suggested technique identified nasty meme posts on Facebook written in Amharic.
近年来，仇恨言论在Facebook等社交媒体网站上的传播更为频繁。在Facebook上，仇恨言论可以通过文字、图片或视频传播。我们建议采用深度学习方法来识别Facebook上发布的冒犯性表情包，以阿姆哈拉语为例。”研究过程始于手动收集Facebook用户发布的表情包。接下来是文本数据提取、标注、预处理、分割、特征提取、模型开发和评价，使用Amharic ocr提取文本数据。字符规范化、停止词删除和不必要的字符删除组成了文本预处理步骤。使用分层KFold将文本数据集分为训练集（80%）、验证集（10%）和测试集（10%）。向量是从使用Bog of words （BOW）、TFIDF和单词嵌入的预处理文本中创建的。然后，将向量输入机器学习算法：NB、DT、RF、KNN、LSVM和LR，以及基于Dense、BiGRU和BiLSTM算法的深度学习模型。经过多次试验，选择了参数最优的模型。BiLSTM + Dense模型的准确率为94%，该技术可以识别出Facebook上用阿姆哈拉语写的令人讨厌的表情包。
{"title":"Identification of hateful amharic language memes on facebook using deep learning algorithms","authors":"Mequanent Degu Belete, Girma Kassa Alitasb","doi":"10.1016/j.sasc.2025.200258","DOIUrl":"10.1016/j.sasc.2025.200258","url":null,"abstract":"<div><div>Hate speech has been disseminated more frequently on social media sites like Facebook in recent years. On Facebook, hate speech can proliferate through text, image, or video. We suggested a deep learning approach to identify offensive memes posted on Facebook in case of Amharic language'. The research process commenced by manually gathering memes posted by Facebook users. Next came textual data extraction, annotation, preprocessing, splitting, feature extraction, model development and assessment Amharic OCRs were employed to extract textual data. Character normalization, stop word removal, and unnecessary character removal make up the text-preprocessing step. Using Stratified KFold the textual dataset is split into the train set (80 %), the validation set (10 %) and the test set (10 %). Vectors are created from the preprocessed texts using the Bog of words (BOW), TFIDF and word embeddings. Following that, the vectors are fed into Machine learning algorithms: NB, DT, RF, KNN, LSVM and LR, and deep learning models that are based on Dense, BiGRU, and BiLSTM algorithms. The model with the optimal parameters is chosen after numerous experiments. With an accuracy rate of 94 %, the BiLSTM + Dense model, the suggested technique identified nasty meme posts on Facebook written in Amharic.</div></div>","PeriodicalId":101205,"journal":{"name":"Systems and Soft Computing","volume":"7 ","pages":"Article 200258"},"PeriodicalIF":0.0,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143895589","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号