Xinyun Feng, Tao Peng, Ningguo Qiao, Haitao Li, Qiang Chen, Rui Zhang, Tingting Duan, JinFeng Gong
{"title":"ADWNet:基于 YOLOv8 的改进型检测器,用于恶劣天气下的自动驾驶应用","authors":"Xinyun Feng, Tao Peng, Ningguo Qiao, Haitao Li, Qiang Chen, Rui Zhang, Tingting Duan, JinFeng Gong","doi":"10.1049/itr2.12566","DOIUrl":null,"url":null,"abstract":"<p>Drawing inspiration from the state-of-the-art object detection framework YOLOv8, a new model termed adverse weather net (ADWNet) is proposed. To enhance the model's feature extraction capabilities, the efficient multi-scale attention (EMA) module has been integrated into the backbone. To address the problem of information loss in fused features, Neck has been replaced with RepGDNeck. Simultaneously, to expedite the model's convergence, the bounding box's loss function has been optimized to SIoU loss. To elucidate the advantages of ADWNet in the context of adverse weather conditions, ablation studies and comparative experiments were conducted. The results indicate that although the model's parameter count increased by 18.4%, the accuracy for detecting rain, snow, and fog in adverse weather conditions improved by 22%, while the FLOPs (floating point operations) decreased by 5%. The results of the comparison experiments conducted on the WEDGE dataset show that ADWNet outperforms other object detection models in adverse weather in terms of accuracy, model parameters and FLOPs. To validate ADWNet's real-world efficacy, data was extracted from a car recorder under adverse conditions on highways, visual inference was conducted, and its accuracy was demonstrated in interpreting real-world scenarios. The config files are available at https://github.com/Xinyun-Feng/ADWNet.</p>","PeriodicalId":50381,"journal":{"name":"IET Intelligent Transport Systems","volume":null,"pages":null},"PeriodicalIF":2.3000,"publicationDate":"2024-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/itr2.12566","citationCount":"0","resultStr":"{\"title\":\"ADWNet: An improved detector based on YOLOv8 for application in adverse weather for autonomous driving\",\"authors\":\"Xinyun Feng, Tao Peng, Ningguo Qiao, Haitao Li, Qiang Chen, Rui Zhang, Tingting Duan, JinFeng Gong\",\"doi\":\"10.1049/itr2.12566\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>Drawing inspiration from the state-of-the-art object detection framework YOLOv8, a new model termed adverse weather net (ADWNet) is proposed. To enhance the model's feature extraction capabilities, the efficient multi-scale attention (EMA) module has been integrated into the backbone. To address the problem of information loss in fused features, Neck has been replaced with RepGDNeck. Simultaneously, to expedite the model's convergence, the bounding box's loss function has been optimized to SIoU loss. To elucidate the advantages of ADWNet in the context of adverse weather conditions, ablation studies and comparative experiments were conducted. The results indicate that although the model's parameter count increased by 18.4%, the accuracy for detecting rain, snow, and fog in adverse weather conditions improved by 22%, while the FLOPs (floating point operations) decreased by 5%. The results of the comparison experiments conducted on the WEDGE dataset show that ADWNet outperforms other object detection models in adverse weather in terms of accuracy, model parameters and FLOPs. To validate ADWNet's real-world efficacy, data was extracted from a car recorder under adverse conditions on highways, visual inference was conducted, and its accuracy was demonstrated in interpreting real-world scenarios. The config files are available at https://github.com/Xinyun-Feng/ADWNet.</p>\",\"PeriodicalId\":50381,\"journal\":{\"name\":\"IET Intelligent Transport Systems\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":2.3000,\"publicationDate\":\"2024-09-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://onlinelibrary.wiley.com/doi/epdf/10.1049/itr2.12566\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IET Intelligent Transport Systems\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1049/itr2.12566\",\"RegionNum\":4,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ENGINEERING, ELECTRICAL & ELECTRONIC\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IET Intelligent Transport Systems","FirstCategoryId":"5","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1049/itr2.12566","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
ADWNet: An improved detector based on YOLOv8 for application in adverse weather for autonomous driving
Drawing inspiration from the state-of-the-art object detection framework YOLOv8, a new model termed adverse weather net (ADWNet) is proposed. To enhance the model's feature extraction capabilities, the efficient multi-scale attention (EMA) module has been integrated into the backbone. To address the problem of information loss in fused features, Neck has been replaced with RepGDNeck. Simultaneously, to expedite the model's convergence, the bounding box's loss function has been optimized to SIoU loss. To elucidate the advantages of ADWNet in the context of adverse weather conditions, ablation studies and comparative experiments were conducted. The results indicate that although the model's parameter count increased by 18.4%, the accuracy for detecting rain, snow, and fog in adverse weather conditions improved by 22%, while the FLOPs (floating point operations) decreased by 5%. The results of the comparison experiments conducted on the WEDGE dataset show that ADWNet outperforms other object detection models in adverse weather in terms of accuracy, model parameters and FLOPs. To validate ADWNet's real-world efficacy, data was extracted from a car recorder under adverse conditions on highways, visual inference was conducted, and its accuracy was demonstrated in interpreting real-world scenarios. The config files are available at https://github.com/Xinyun-Feng/ADWNet.
期刊介绍:
IET Intelligent Transport Systems is an interdisciplinary journal devoted to research into the practical applications of ITS and infrastructures. The scope of the journal includes the following:
Sustainable traffic solutions
Deployments with enabling technologies
Pervasive monitoring
Applications; demonstrations and evaluation
Economic and behavioural analyses of ITS services and scenario
Data Integration and analytics
Information collection and processing; image processing applications in ITS
ITS aspects of electric vehicles
Autonomous vehicles; connected vehicle systems;
In-vehicle ITS, safety and vulnerable road user aspects
Mobility as a service systems
Traffic management and control
Public transport systems technologies
Fleet and public transport logistics
Emergency and incident management
Demand management and electronic payment systems
Traffic related air pollution management
Policy and institutional issues
Interoperability, standards and architectures
Funding scenarios
Enforcement
Human machine interaction
Education, training and outreach
Current Special Issue Call for papers:
Intelligent Transportation Systems in Smart Cities for Sustainable Environment - https://digital-library.theiet.org/files/IET_ITS_CFP_ITSSCSE.pdf
Sustainably Intelligent Mobility (SIM) - https://digital-library.theiet.org/files/IET_ITS_CFP_SIM.pdf
Traffic Theory and Modelling in the Era of Artificial Intelligence and Big Data (in collaboration with World Congress for Transport Research, WCTR 2019) - https://digital-library.theiet.org/files/IET_ITS_CFP_WCTR.pdf