Obstacle detection:improved YOLOX-S based on swin transformer-tiny

作     者：ZHANG Hongying LU Chengjian CHEN Enyao ZHANG Hongying;LU Chengjian;CHEN Enyao

作者机构：College of Electronic Information and AutomationCivil Aviation University of ChinaTianjin300300China 

出 版 物：《Optoelectronics Letters》 (光电子快报（英文版）)

年 卷 期：2023年第19卷第11期

页      面：698-704页

核心收录：

学科分类：0809[工学-电子科学与技术（可授工学、理学学位）] 08[工学] 0702[理学-物理学] 

基　　金：supported by the Graduate Research Innovation Project of Civil Aviation University of China (No.2021YJS086) 

主　　题：obstacle network backbone 

摘      要：Aiming at the accuracy challenge in obstacle detection for autonomous driving,we propose an improved you only look once X-S(YOLOX-S) model based on swin transformer-tiny YOLOX-S(ST-YOLOX-S) for obstacle detection,which could detect multiple targets,including people,cars,bicycles,motorcycles,and *** method mainly comprises two aspects as *** improve the capability of local feature extraction and then obtain more accurate detection for obstacles under real-world vehicle conditions,the existing backbone of YOLOX-S is replaced with the swin transformer-tiny *** reduced the number of channels between the swin transformer and path aggregation-feature pyramid network(PA-FPN) from [96,192,384,768] to [192,384,768],to decrease the computational cost and then make the swin transformer-tiny more compatible with the ***,on the popular COCO dataset,the proposed ST-YOLOX-S improves the detection mean average precision(mAP) by 6.1% when compared with *** the five types of obstacles that appear in simulated actual vehicle conditions,our ST-YOLOX-S also achieves superior performance compared to ***,our method achieves significant performance over the YOLOv3 on obstacle detection,which shows the effectiveness of the proposed algorithm.