基于无人机影像技术的柑橘园产量估算方法

Yield Estimation Method of Citrus Orchard Based on UAV Image Technology

为了实现对柑橘果园产量的准确自动估算,设计了一套基于无人机影像技术的柑橘产量估算系统,包括无人机、摄像机、图像接收器和计算机等主要模块。首先,通过无人机航拍获取果园中的柑橘果树视频数据,接收器将获得的数据传输到计算机平台,使用改进的YOLOv5目标检测算法获得视频中柑橘果实的位置信息;然后,通过卡尔曼滤波和卷积神经网络实现不同帧之间相同水果的匹配关联,为每个水果标记独立的ID,以ID总数作为视频中的柑橘果实总数的估计值。试验结果表明:果园柑橘产量的估算最高获得88.6%的准确率,可有效帮助果农掌握果园中的柑橘数量,为果园的智慧管理提供了科学决策依据。

In order to realize the accurate and automatic estimation of citrus orchard yield,a citrus yield estimation system based on UAV image technology was designed.The system includes UAV,camera,image receiver,computer and other main modules.Firstly,the video data of citrus fruit trees in the orchard is obtained by UAV aerial photography,the receiver transmits the obtained data to the computer platform,and the position information of citrus fruit in the video is obtained by using the improved yolov5 target detection algorithm;Then,the matching Association of the same fruit between different frames is realized by Kalman filter and convolutional neural network.Each fruit is marked with an independent ID,and the total number of IDS is used as the estimated value of the total number of citrus fruits in the video.The system is tested on the citrus orchard video collected on the spot to obtain the highest yield estimation accuracy of 88.6%,which can effectively help fruit farmers master the number of citrus in the orchard and provide scientific decision-making basis for the intelligent management of the orchard.