A Tobacco Brand Recognition Method Based on HOG and SVM

The appearance inspection system cannot automatically switch the brand, and a person needs to log into the system and manually switch the inspection brand in the process of production and brand change. To address this problem, a deep learning-based automatic cigarette brand recognition method is developed and validated. Experiments show that: the system can automatically detect the production brand, effectively save the time to change the brand, and has important reference significance for the application in the field of brand identification and automatic switching detection of the brand.