中国激光, 2012, 39 (2): 0203007, 网络出版: 2012-01-11   

基于视觉激光切割厚板切割状态监控方法

Method of Monitoring Cutting Point State in Laser Cutting Plank Process Based on Vision
作者单位
1 燕山大学机械工程学院, 河北 秦皇岛 066004
2 燕山大学车辆与能源学院, 河北 秦皇岛 066004
摘要
利用光纤镜头和彩色工业摄像机实时采集激光切割厚板中切割点的图像,从彩色图像中分别选取蓝色、绿色和红色通道图像,分析各通道图像的特点和切割点的几何形状特征。首先以激光焦点在图像中的位置为中心建立坐标系,以x轴方向为起始,45°为间隔向8个方向搜索激光切割区域的边缘点,根据边缘点到坐标原点的距离信息确定激光切割方向和切割顶点;建立边缘识别用抛物线模型,根据边缘处存在灰度特征、梯度特征和方向特征设计识别目标函数,识别切割顶点两侧边缘,进而识别整个切割点处的几何形状。实验表明识别方法具有良好的适应性、准确性和实时性。
Abstract
Fiber of optical lens and color industrial video camera are used to capture image of cutting point in laser cutting process. Blue channel, green channel and red channel images are adopted to analyze cutting point geometry shape from color image. The position of laser beam focal point is used as origin of coordinates and the coordinate system is set up. The x axis is original direction and interval is 45 degree, the edge points of laser cutting region are searched from eight directions. According to distances from origin of coordinates to edge points, the laser cutting direction and cutting peak are fixed. Parabola model is set up, and the objective function is designed according to gray feature, gradient feature and direction feature, and the objective function is used to identify edge at bilateral cutting peak, and then cutting point geometry shape is identified. Test shows that the method has preferable adaptability, accuracy and real-time quality.

高士友, 马雷, 张润生. 基于视觉激光切割厚板切割状态监控方法[J]. 中国激光, 2012, 39(2): 0203007. Gao Shiyou, Ma Lei, Zhang Runsheng. Method of Monitoring Cutting Point State in Laser Cutting Plank Process Based on Vision[J]. Chinese Journal of Lasers, 2012, 39(2): 0203007.

本文已被 7 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!