基于机器视觉的高杆绣球检测计数系统

High-rod Hydrangea Detection and Counting System Based on Machine Vision

针对光照强度在目标轮廓提取上的问题,提出一种自适应阈值分割与连通域筛选的目标轮廓提取算法。通过自适应阈值分割结合形态学梯度进行边缘检测,并利用漫水填充算法提取投球内圈及以外待检测区域,对提取的待检测区域采用高斯混合模型进行目标及运动方向检测,最终实现高杆抛绣球检测与计数。实验结果表明,该方法能实时有效地提取目标轮廓并统计投进投球圈的绣球数。

Aiming at the problem of illumination intensity extraction on target contour,a target contour extraction algorithm based on adaptive threshold segmentation and connected domain filtering is proposed.The edge detection is performed by adaptive threshold segmentation combined with morphological gradient,and the water-filled filling algorithm is used to extract the area to be detected outside the pitch inner ring and the inner circle.The Gaussian mixture model is used to detect the target and motion direction.Achieve high rod throwing hydrangea detection and counting.The experimental results show that the method can effectively extract the target contour in real time and count the number of hydrangea thrown into the pitch circle.