Tool wear image segmentation algorithm based on K-means clustering

Abstract

Abstract The existing tool wear region segmentation and detection methods make it difficult to achieve accurate detection of tool wear images under the conditions of noise interference. To further improve the detection efficiency and detection accuracy, combined with the characteristics of tool wear, this paper proposes a tool wear image segmentation algorithm based on K-means clustering to achieve accurate detection of tool wear images. Firstly, the acquired image is pre-processed to reduce the image processing computation and optimize the image, filter out the noise, and reduce the interference of the image. Then K-means clustering is used to cluster the surface features of the image, and then a threshold segmentation algorithm is used to locate the tool wear area and segment it out. The final experimental results show that the segmentation accuracy of this method is high and can be applied to tool wear area detection.