Abstract
Aiming at the question of defect image segmentation in quality detection of bullet appearance. Through the analysis of color characteristics of bullet appearance defects and combined with OTSU algorithm and two-peak algorithm, Proposes a hybrid threshold segmentation method based on color model. The method first extracts component map R , using OTSU algorithm to segmentation high light part of the image ,and finally with two-peak algorithm to the further defect segmentation. Application of Matlab simulation results show that, this method can be successfully applied to the bullet appearance defect segmentation. General Introduction The bullet is the most important part of light weapons, whose role it is self-evident. In order to ensure the quality in the production process of bullets, they must to detect bullet surface, to eliminate defective cartridge. At present, China's military enterprises generally adopt manual work to detect the bullets surface, and the manual sorting way after found the defective bullets. This traditional testing method has low efficiency, low accuracy and high labor intensity, reduced the degree of automation of the production of bullet. With the development of science and technology progressing, and the bullet's quality and production efficiency requirements increasing, while the artificial detection method of cartridge surface have been unable to meet the requirements of the automatic production of the era, how to improve the quality, efficiency and automation degree of the bullet appearance inspection, is a very important problems to be solved. Automatic detection method for image processing can achieve defect, improve the efficiency of the detection. In the automatic detection of defects, defect image segmentation is a key for defect identification, the segmentation quality directly affects the subsequent image processing effects, and even determine the success or failure. Because the surface of a metal cylinder bullet is smooth surface, the appearance of the image acquisition process cartridges often occur in high light phenomenon, which makes it unable to complete the bullet appearance defect segmentation by gray level image, while color images provide more information than the gray image, so get more and more attention. In order to accurately, rapidly segment image defect, from the analysis of the color features of defects and color feature comparison bullets body appearance, the method of using RGB color model and hybrid threshold segmentation to determine the defect area of the bullet appearance. Gray threshold segmentation of the defect 1.1 OTSU algorithm There are many kinds of appearance defects of bullets, discuss several bullet appearance defects of the most common here, including green spot, indentation and dirt. As shown in fig 1. International Conference on Advances in Mechanical Engineering and Industrial Informatics (AMEII 2015) © 2015. The authors Published by Atlantis Press 1313 (a)Green spot (b) Indentation (c)Dirt Fig:1 Appearance defects of bullets The most basic gray threshold segmentation method is two-peak algorithm and OTSU algorithm, OTSU algorithm was proposed in 1979 by Otsu in Japan, the segmentation of image is calculated as follow. The pixels in a digital image is divided into 0 1 and C C types according to the gray level in ( ) f x, y with the threshold, i.e. { } { } 0 1 min 1 2 max ( , ) ( , ) ( , ) ( , ) C f x y f f x y T C f x y f f x y T = ≤ ≤ = ≥ ≥ (1) Among them, min max f f 、 were gray minimum and maximum value in the image. Let i N be the number of pixel whos gray value is ( ) min max i f i f ≤ ≤ , then the total pixel image is i i N N =∑ ,therefore, the each probability of the gray levels for ( ) i N P i N = , then the probability of 0 C is ( )
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.