Improvement of SIFT algorithm and its application in visual inspection system for PET bottle capping defects. Issue 1 (1st December 2022)

• Record Type: Journal Article
• Authors: Li, Dan; Yang, Mengqin
• Is Part Of: Journal of physics. Volume 2390 Issue 1(2022)
• Publication Date: 2022-12-01
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/2390/1/012097
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Aiming at the defects of PET beverage bottles in the process of capping and filling, an improved algorithm based on visual inspection is proposed. First, an improved SIFT algorithm is proposed to improve the running speed, and OPENCV, C++, and Virtual Studio are used to build the interface and implement the algorithm. The preprocessed images are analyzed by the improved SIFT algorithm, and experiments and data analysis are carried out. The research results show that compared with the traditional SIFT algorithm, the improved SIFT algorithm greatly saves the algorithm time when detecting PET bottle capping defects and retains the anti-interference ability of SIFT feature points. The detection of insufficient liquid levels provides a new method and makes the algorithm more practical and valuable.