为了提高汽车座椅零件在生产车间分拣的效率,将数字图像处理技术应用于该类产品的检测,设计了基于机器视觉的座椅零件双目检测系统。首先采用简单可行的方法对两个相机拍摄的图像进行拼接,克服传统图像拼接特征点难找的缺陷;接着采用改进最大类间方差法获取动态阈值对图像进行分割,使用形态学处理方法解决纺布产品边缘毛刺干扰的难题;然后将处理后的二值图像分别向X轴Y轴投影取像素直方图方法准确定位目标区域;最后基于轮廓检测、圆检测以及角点检测的参数模板匹配方法可以实现产品的精确测量。实验结果表明,该系统具有识别精度高、可靠性强,分拣速率可达120个/分钟,已经取代传统的人工检测方法 ,成功应用于座椅零件产品生产流水线中。
In order to improve the efficiency of car seat parts sorting in the production workshop,this paper designs a binocular detection system based on machine vision by applying the digital image processing technology to the products detecting.The experiment results show that the system has a high recognition accuracy,reliable performance,and the sorting rate can reach 120/min.The system has replaced the traditional manual detection method,and it is applied to the industrial production lines successfully.