Title: Detection method of interface defects of titanium nitride thin film coating materials based on image processing

Authors: Shengan Zhou; Yuan Wang; Gengsheng Huang; Yiyun Zhang

Addresses: Guangdong Vocational Institute of Public Administration Electronic Information System, Guangzhou 510800, China ' Guangdong Vocational Institute of Public Administration Electronic Information System, Guangzhou 510800, China ' Guangdong Vocational Institute of Public Administration Electronic Information System, Guangzhou 510800, China ' Guangdong Vocational Institute of Public Administration Electronic Information System, Guangzhou 510800, China

Abstract: In order to improve the accuracy of defect location and defect feature extraction of coating materials, an interface defect detection method based on image processing is designed in this paper. After scanning the interface image of TiN film coating material, denoising was carried out to improve the accuracy of defect location. After the preprocessing image is segmented and the defect area is located, the optical signal of the defect area is collected to achieve the purpose of improving the accuracy of defect feature extraction, and then the defect location is detected by extracting the details of grey image. Experimental results show that the accuracy of the proposed method is between 92% and 97%, and the accuracy of defect feature extraction is between 93.6% and 95.7%, which indicates that the proposed method is more effective.

Keywords: image scanning; image denoising; image preprocessing; titanium nitride film coating; defect detection; defect location; feature extraction.

DOI: 10.1504/IJMPT.2022.124256

International Journal of Materials and Product Technology, 2022 Vol.65 No.1, pp.41 - 51

Received: 11 Jun 2021
Accepted: 28 Sep 2021
Published online: 20 Jul 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article