欢迎访问光电子·激光编辑部网站！

您是第14631037位访问者！

姚艺莲,裴东,蒲向荣.基于YOLOv5的轻量级火焰视频流实时检测算法[J].光电子激光,2023,(11):1150~1157

基于YOLOv5的轻量级火焰视频流实时检测算法

Lightweight flame video stream real-time detection algorithm based on YOLOv5

投稿时间：2022-11-04 修订日期：2022-12-20

DOI：

中文关键词: 火焰检测 DGC_YOLOv5算法注意力机制 Ghost模块深度可分离卷积(DS_Conv)

英文关键词:flame detection DGC_YOLOv5 algorithm attention module Ghost module depthwise separable convolution (DS_Conv)

基金项目:国家自然科学基金(61961037)资助项目

作者	单位
姚艺莲	西北师范大学物理与电子工程学院，甘肃兰州 730070 甘肃省智能信息技术与应用工程研究中心，甘肃兰州 730070
裴东	西北师范大学物理与电子工程学院，甘肃兰州 730070 甘肃省智能信息技术与应用工程研究中心，甘肃兰州 730070
蒲向荣	西北师范大学物理与电子工程学院，甘肃兰州 730070

摘要点击次数: 86

全文下载次数: 0

中文摘要:

针对火焰检测模型小目标检测能力差、模型体积大、计算复杂、难以部署到移动端设备的问题,提出了一种轻量化的DGC_YOLOv5 (you only look once v5)算法。本文首先调用k-means计算函数,计算出适合本文数据集的锚框尺寸；其次引入卷积块注意力机制(convolutional block attention module,CBAM),提高算法对小目标的检测能力；然后利用轻量型的Ghost模块对主干网络中的C3模块进行改进；最后利用深度可分离卷积(depthwise separable convolution,DS_Conv) ,用简单的线性计算代替复杂计算,降低模型复杂度,减小模型体积。实验表明,相比原始的 YOLOv5算法,本文算法在测试集上的平均精度均值(mean average precision,mAP)可达到94.4%,比原始算法提高1.7个百分点,在视频测试集上平均检测速度可达到71 FPS,可以满足实时检测的要求,参数量和计算量分别减少为原来的41.2%和34.8%,模型大小减少8.4 M,便于后续移动设备端的部署。

英文摘要:

A lightweight DGC_YOLOv5 (you only look once v5) algorithm is proposed to solve the problems of poor detection capability of small targets,large size of the model,complex calculation,and difficult deployment on mobile devices for flame detection model.Firstly,the k-means calculation function is used to calculate the anchor size for this data set.Secondly,the convolutional block attention module (CBAM) is introduced to improve the detection ability of this algorithm to small target.Then the lightweight Ghost module is adopted to improve the C3 modules in backbone network.Finally,the depthwise separable convolution (DS_Conv) which uses simple linear calculation instead of complicated calculation is used to reduce model complexity and size.Experiments show that compared with the original YOLOv5 algorithm,the mean average precision (mAP) of the proposed algorithm can reach 94.4% on the test set,1.7% higher than the original algorithm.The average detection speed of the proposed algorithm can reach 71 FPS on the video test set,which can meet the requirements of real-time detection.Parameters and the floating-point operations (FLOPs) calculating amount are respectively reduced to 41.2% and 34.8% of the original algorithm,and the model size is reduced by 8.4 M,which facilitates the subsequent deployment on mobile devices.

查看全文下载PDF阅读器

关闭