摘要
在建筑业、工地等场景下,由于受到天气、人数、光照强度、拍摄角度和距离等因素的影响,导致在安全帽智能检测时容易出现准确度低、漏检率大、错检率高的问题.为了解决该问题,提出了一种基于YOLO-ST的安全帽佩戴检测算法.该算法具有两个优势:首先,使用更容易捕获图像全局信息的Swin Transformer作为网络的特征提取器,增强网络对安全帽特征的提取能力;其次,设计密集的空间金字塔池化模块并引入到YOLO-ST中,以获取目标中更加丰富的细节信息.实验结果表明,在公开的SHWD数据集上,YOLO-ST的平均识别精度达到了91.3%.与其它方法相比,YOLO-ST取得了更精确的检测结果.
In the construction industry scenarios,due to the influence of factors such as weather,number of people,light intensity,imaging angle and distance,it is easy to lead to the problem of helmet detection with low accuracy,high missed detection and false detection ratio.To solve this problem,an accurate helmet detection algorithm based on YOLO-ST is proposed in this paper.YOLO-ST has two advantages:First,Swin Transformer,which is easier to capture the global information of the image,is employed as the feature extractor of YOLO-ST to improve the ability of extracting the features of the safety helmet.Secondly,a dense spatial pyramid pooling module is designed and introduced into YOLO-ST,which can capture richer details about the objects.The experimental results demonstrate that the average detection accuracy of YOLO-ST can achieve 91.3%on the public SHWD dataset,which is more accurate than other popular methods.
作者
杜晓刚
王玉琪
晏润冰
古东鑫
张学军
雷涛
DU Xiao-gang;WANG Yu-qi;YAN Run-bing;GU Dong-xin;ZHANG Xue-jun;LEI Tao(Shaanxi Joint Laboratory of Artificial Intelligence,School of Electronic Information and Artificial Intelligence,Shaanxi University of Science&Technology,Xi′an 710021,China;College of Mechanical and Electrical Engineering,Shaanxi University of Science&Technology,Xi′an 710021,China;School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处
《陕西科技大学学报》
北大核心
2022年第6期177-183,191,共8页
Journal of Shaanxi University of Science & Technology
基金
国家自然科学基金项目(61861024,61871259,62271296)
甘肃省自然科学基金项目(20JR5RA404,21JR7RA282)
陕西省杰出青年科学基金项目(2021JC-47)
陕西省重点研发计划项目(2021ZDLGY08-07,2022GY-436)
陕西省创新能力支撑计划项目(2020SS-03)
陕西省自然科学基础研究计划项目(2022JQ-018,2022JQ-175)。
