Defect detection of cans using Gabor wavelet

Defect detection of cans using Gabor wavelet

Salar SeyedMohammadGhavami1 Elham Hashemizadeh2

1) Department of Computer, Karaj Branch, Islamic Azad University, Karaj, Iran
2) Department of Mathematics, Karaj Branch, Islamic Azad University, Karaj, Iran

IOI Code : XYDK-AHCCK
Publication : International Conference On Research Science And Technology(restconf.com)
Abstract :
    There is possibility of destruction on the cans, while making canned foods. Regardless of the cause of the malfunction, an accurate diagnosis can help them to faster improve the quality and production process. Soundness of cans has an important role in product quality and market competitiveness. In this paper, the image processing techniques to detect damages on canned foods have been dealt. Using the wavelet transform the failures transform into 4 sub-band images and we focus on the low sub-band frequency. Application of Gabor wavelet filter to the lower sub-band images failures shows significant properties including various directions and frequencies, effective in better extracting textual features. The usage of low-frequency sub-bands instead of the original image, before applying wavelet transform, is to detect the possible noises. In fact, the results are to recognize the true failures that comparing similar methods have lesser computational complexity. System simulation is done by MATLAB and provided method has a better recognition rate.

Keywords : image processing quality control gabor wavelet lower sub band bottom automated review