Using intelligent agricultural machines in paddy fields has received great attention.\nAn obstacle avoidance system is required with the development of agricultural machines. In order to\nmake the machines more intelligent, detecting and tracking obstacles, especially the moving obstacles\nin paddy fields, is the basis of obstacle avoidance. To achieve this goal, a red, green and blue (RGB)\ncamera and a computer were used to build a machine vision system, mounted on a transplanter.\nA method that combined the improved You Only Look Once version 3 (Yolov3) and deep Simple\nOnline and Realtime Tracking (deep SORT) was used to detect and track typical moving obstacles,\nand figure out the center point positions of the obstacles in paddy fields. The improved Yolov3 has\n23 residual blocks and upsamples only once, and has new loss calculation functions. Results showed\nthat the improved Yolov3 obtained mean intersection over union (mIoU) score of 0.779 and was\n27.3% faster in processing speed than standard Yolov3 on a self-created test dataset of moving\nobstacles (human and water buffalo) in paddy fields. An acceptable performance for detecting and\ntracking could be obtained in a real paddy field test with an average processing speed of 5-7 frames\nper second (FPS), which satisfies actual work demands. In future research, the proposed system could\nsupport the intelligent agriculture machines more flexible in autonomous navigation.
Loading....