篇名 | Research on Improving Canny Edge Detection Algorithm in the Binocular Visual Recognition for Hazardous Chemical Warehouse |
---|---|
卷期 | 31:3 |
作者 | Yong-Yan 、 Hong-chao Ma 、 Ze-Chen Huang 、 Kai-Li Li 、 Bo-Tao Wang 、 Yang-Li 、 Jian-Shi |
頁次 | 337-350 |
關鍵字 | Canny algorithm 、 hazardous chemicals storage 、 iterative mean 、 median filtering 、 Otsu 、 EI 、 MEDLINE 、 Scopus |
出刊日期 | 202006 |
DOI | 10.3966/199115992020063103027 |
The safety distance monitoring in the hazardous chemical warehouse is the key to ensure the safe storage of hazardous chemicals. Binocular vision is one of the promising technical means. The Canny edge detection algorithm in the binocular vision of the warehouse is an effective algorithm for the boundary extraction of the dangerous goods used in warehouse, but the algorithm faces the salt and pepper noise interference in the warehouse where has a complex environment. Firstly, adaptive median filtering is used to denoising. Then iterative mean method and Otsu method is used to adaptively set threshold and extract edges. Under normal lighting, night vision, night vision and light interference, carry on experimental research on storage stacks such as warehouse model boxes and cylinders. The results show that the improved algorithm significantly improves denoising effect and edge extraction accuracy.