基于无人机视觉的河道漂浮垃圾分类检测技术研究

Study on Classification and Detection Technology of River Floating Garbage Based on UAV Vision

传统人工河道漂浮垃圾巡检耗时耗力,无人机、无人船河道巡检成为主要方式,目前尚局限于河道漂浮垃圾图像人工判读或简单计算机目标检测,缺乏对漂浮垃圾的自动分类检测。基于无人机航拍影像构建了研究区河道漂浮垃圾数据集,使用深度学习方法对垃圾进行分类识别。顾及河道漂浮垃圾类别不均衡以及在无人机影像中占比较小等情况,在多尺度检测以及数据增强等方面对YOLOv5s目标检测算法进行了针对性改进,经试验验证,改进后算法相较于原始算法,提升了对小目标的检测精度,其类别均衡准确率提高了3.47%。研究表明:将深度学习方法与无人机技术相结合能够高效、准确地对垃圾进行识别和分类,为治理河道漂浮垃圾提供决策依据。

The traditional manual inspection of river floating garbage is time-consuming and labor-consuming. UAVsand unmanned boats have become the main methods of river inspections. At present,it is limited to manual interpretation ofriver floating garbage images or simple computer target detection,and there is a lack of automatic classification and detectionof floating garbage. In this paper,a dataset of floating garbage in the river channel of the study area is constructed based ondrone aerial images,and a deep learning method is used to classify and recognize garbage. Taking into account the unbal-anced types of floating garbage in the river and the relatively small proportion in UAV images,the YOLOv5s target detectionalgorithm has been improved in terms of multi-scale detection and data enhancement. The experimental results show that theimproved algorithm improves the mean average precision by 3.47% compared with the original algorithm. The study resultsshow that the combination of deep learning methods and drone technology can efficiently and accurately identify and classifygarbage,and provide a basis for decision-making in the treatment of floating garbage in the river.