基于机器视觉的小型光伏电站鸟粪监测系统  

BIRD DROPPINGS MONITORING SYSTEM FOR SMALL PHOTOVOLTAIC POWER STATION BASED ON MACHINE VISION

作　　者：王松 顾翔[1] 王强[1] Wang Song;Gu Xiang;Wang Qiang(School of Information Science and Technology,Nantong University,Nantong 226000,Jiangsu,China)

机构地区：[1]南通大学信息科学技术学院,江苏南通226000

出　　处：《计算机应用与软件》2025年第4期57-62,共6页Computer Applications and Software

基　　金：江苏省高等学校自然科学研究重大项目(19KJA320004)。

摘　　要：为了准确、高效地识别和定位小型光伏电站上的鸟粪,将改进后的YOLOv5模型搭载到树莓派开发板上构成一套光伏电站鸟粪检测系统。调低置信度阈值识别所有可疑鸟粪,识别并划分出单块光伏板,调高置信度阈值对有可疑鸟粪的光伏板进行精准鸟粪识别。为了使YOLOv5算法更适用于鸟粪目标的检测,在原YOLOv5算法中融合金字塔分割注意力模块,增加小目标检测层,用SoftPool替换原有池化操作。在测试集上,针对光伏板识别的PV-YOLOv5模型的mAP_0.5为96.78%,比Faster-RCNN高2.35百分点;针对鸟粪识别的NF-YOLOv5的mAP_0.5为94.12%,较原YOLOv5模型提升5.8百分点。In order to accurately and efficiently identify and locate the bird droppings on small photovoltaic power station,the improved YOLOv5 model is carried on the Raspberry Pi to form a bird droppings detection system of photovoltaic power plants.The system reduced the threshold of confidence to identify all suspicious bird droppings,identified and partitioned single photovoltaic panels,and increased the confidence threshold to accurately detect suspicious bird droppings in photovoltaic panels.In order to make the YOLOv5 algorithm more suitable for detection,the pyramid split attention was integrated in the algorithm.The small target detection layer was added and the original pooling operation was replaced by SoftPool.In the test set,the mAP_0.5 of PV-YOLOv5 model identified for photovoltaic panels was 96.78%,which was 2.35 percentage points higher than that of Faster-RCNN.The mAP_0.5 of NF-YOLOv5 for bird droppings recognition was 94.12%,which was 5.8 percentage points higher than the original YOLOv5 model.

关 键 词：鸟粪 光伏电站 机器视觉 YOLOv5 树莓派 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TM615[自动化与计算机技术—计算机科学与技术]