基于机器视觉的采茶机割刀控制方法

Cutter control method of tea-picking machine based on machine vision

摘要:

针对现有乘坐式采茶机切割的茶叶质量偏低的问题,提出了一种基于机器视觉的采茶机割刀控制方法。首先对相机参数进行低复杂度的标定,然后采用动态阈值分割方法和颜色分类器,提取茶叶图像中的嫩芽区域,并设计间接定位法定位弧形割刀位置；接着通过计算两个指定区域的嫩芽面积及它们的和与差异,得到弧形割刀左右两侧的动作参数；最后将动作参数传给下位机,控制割刀到达预期位置。本方法分别对单株茶树和实地茶园进行了实验,结果表明该方法能够准确定位弧形割刀位置和识别嫩芽区域,实现割刀位置的自适应调整,具有较好的茶叶切割效果,切割得到的嫩芽比例可达70%以上。

Abstract:

To improve the low picking quality of traditional tea-picking machine, a cutter control method based on machine vision is proposed. First, a low complexity calibration method for camera parameters is designed. Second, the cutter is located by an indirect location method. Then, R、G and B components are obtained from the tea image, and a dynamic threshold and a color classification are used for recognizing the tea bud. In addition, tea bud areas of two designated areas are calculated, and the parameters of cutter action is acquired by the sum and differ of the two areas. Finally, the parameters are sent to the inferior machine and the inferior machine adjusts the cutter to the desired position. Experimental results show that the proposed method can locate the cutter and recognize the tea bud accurately, and achieve adjusting the cutter adaptively and the bud rate is more than 70%.