Wear Detection of Drill Bit by Image-based Technique

3rd International Conference on Mechanical, Manufacturing and Process Plant Engineering

机械制造;工业技术

Sukeri, Maziyah^1 ; Paiz Ismadi, Mohd Zulhilmi^1 ; Othman, Abdul Rahim^1 ; Kamaruddin, Shahrul^1

Mechanical Engineering Department, Universiti Teknologi PETRONAS, Seri Iskandar, Perak

32610, Malaysia^1

关键词: Cross correlation methods;    Direct measurement method;    Edge detection methods;    Image-based analysis;    Image-based techniques;    Manufacturing industries;    Thresholding techniques;    Tool condition monitoring;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/328/1/012011/pdf
DOI  :  10.1088/1757-899X/328/1/012011

学科分类：工业工程学

来源: IOP

【 摘 要 】

Image processing for computer vision function plays an essential aspect in the manufacturing industries for the tool condition monitoring. This study proposes a dependable direct measurement method to measure the tool wear using image-based analysis. Segmentation and thresholding technique were used as the means to filter and convert the colour image to binary datasets. Then, the edge detection method was applied to characterize the edge of the drill bit. By using cross-correlation method, the edges of original and worn drill bits were correlated to each other. Cross-correlation graphs were able to detect the difference of the worn edge despite small difference between the graphs. Future development will focus on quantifying the worn profile as well as enhancing the sensitivity of the technique.