切割刀具图像边缘破损检测方法仿真

Simulation of edge damage detection method of cutting tool image

采用目前方法对切割刀具边缘进行破损检测时,没有对刀具图像进行增强处理,导致方法存在图像对比度低、适用性能差和检测精度低的问题。提出基于直方图均衡化的切割刀具边缘破损检测方法,利用天然橡胶刀具图像的灰度中值,在分割直方图等面积原则的基础上进行灰度直方图分割处理,均衡化处理分割处理后的各子灰度直方图,完成天然橡胶刀具图像的增强处理。结合轮廓分割算法和Canny算子对天然橡胶刀具图像进行边缘检测和分割处理,构建天然橡胶刀具的轮廓几何模型,在此基础上生成刀具的标准轮廓曲线,通过对比求差处理获得天然橡胶刀具边缘的破损区域,实现切割刀具边缘破损检测。仿真结果表明,所提方法的图像对比度高、适用性能好、检测精度高。

Generally, the traditional cutting tool edge damage detection method has defects, such as ignoring the enhancement of tool image, low image contrast, detection accuracy and poor applicability. Therefore, a cutting tool edge damage detection method based on histogram equalization is presented. Based on the area principle of gray median and segmentation histogram of natural rubber tool image, the gray histogram was segmented. The sub-gray histogram after segmentation was equalized in order to enhance the natural rubber tool image. The contour segmentation algorithm and Canny operator were introduced to detect and segment the natural rubber tool image to establish the contour geometric model of the natural rubber tool, thus generating the standard contour curve of the tool. According to the comparison of the damaged area of the natural rubber tool edge, the edge damage detection of the cutting tool was achieved. The simulation results show that this method has high image contrast, detection accuracy and excellent applicability.