基于表面评估的工件轮廓识别方法研究

现有加工工件材料和加工方式的多样性使得工件的表面情况多样，导致视觉系统难以准确识别工件轮廓，因此提出了一种适用于不同工件表面的轮廓识别方法。根据纹理区域面积与凸包面积的比值对工件表面进行评估和分类。对于表面质量较好的工件，采用高通线性滤波器对图像进行滤波处理，通过工件表面信息与边缘信息的差异实现工件图像边缘提取；对于表面质量较差的工件，采用一种自适应轮廓提取方法来识别图像边缘。实验表明，与传统的Canny边缘检测算法相比，所提出的方法能够更好地去除噪声干扰，其识别轮廓的精度更高。所提出的轮廓识别方法对不同工件表面有较好的适应性，具有一定的应用价值。

Research on workpiece contour recognition method based on surface evaluation

The diversity of existing workpiece materials and processing methods makes the surface of the workpiece diverse, which makes it difficult for the vision system to accurately recognize the workpiece contour. Therefore, a contour recognition method suitable for different workpiece surfaces was proposed. The workpiece surface was evaluated and classified according to the ratio of texture area to convex hull area. For the workpiece with good surface quality, the high-pass linear filter was used to filter the image, and the edge of the workpiece image was extracted by the difference between the workpiece surface information and the edge information; for the workpiece with poor surface quality, an adaptive contour extraction method was adopted to recognize the image edge. Experimental results showed that the proposed method could better remove noise interference than the traditional Canny edge detection algorithm, and its contour recognition accuracy was higher. The proposed contour recognition method has good adaptability to different workpiece surfaces and has certain application value.