Abstract
Abstract In flame and smoke image processing scenarios, real-time detection of translucent flames and smoke targets is a difficult task. In the process of optimizing accuracy and speed, this study proposed an SDS-YOLOv8n lightweight detection network by using YOLOv8n as a benchmark. First, SimAM was integrated into the SPPF feature pyramid pooling layer to improve the weight attention of the 3D feature map. Secondly, Slim-Neck is used to replace the neck network layer to reduce the parameters and calculations of the model. Finally, the dynamic upsampling operator DySample was used to improve Slim-Neck, integrated into the YOLOv8n lightweight network. Experimental results on the constructed flames and smoke dataset show that the precision rates increased to 81.7%, recall rates increased to 79.2%, and parameters and calculations were reduced by 8% and 18%, respectively. While reducing computing costs, accuracy is improved, and all indicators surpass comparison methods.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
