Multiscale enhancement combined with local maxima for infrared filter defect detection

Due to the problems that some defects on the surface are very weak and the low contrast of the background, adaptive local contrast enhancement was adopted. Then multiscale gaussian side window filter was proposed to enhance the defect details. By looking for appropriate local maxima to locate the de...

Full description

Saved in:
Bibliographic Details
Published in2020 IEEE 9th Joint International Information Technology and Artificial Intelligence Conference (ITAIC) Vol. 9; pp. 1491 - 1496
Main Authors Liu, Qinxiao, Li, Zemin, Wang, Fang, Ding, Chaoyuan, Hu, Dongxia
Format Conference Proceeding
LanguageEnglish
Published IEEE 11.12.2020
Subjects
Classification algorithms
defect detection
Feature extraction
Filtering algorithms
image processing
Image segmentation
Information technology
local maxima
multiscale detail enhancement
Surface morphology
Surface treatment
Online AccessGet full text

Cover

More Information
Summary:Due to the problems that some defects on the surface are very weak and the low contrast of the background, adaptive local contrast enhancement was adopted. Then multiscale gaussian side window filter was proposed to enhance the defect details. By looking for appropriate local maxima to locate the defect area for segmentation. The broken scratches were connected by the morphological method, then scratches and stains were classified by the geometric features parameters. By comparing and analyzing various methods of image detail enhancement and defect segmentation, the results show that the accuracy of the algorithm proposed in this paper is 91.3% in defect detection, faster than the traditional algorithms, and has higher detection efficiency and precision.
DOI:10.1109/ITAIC49862.2020.9338797