Cigarette brand recognition method in complex scene

Abstract

The invention discloses a cigarette brand recognition method in a complex scene. The method comprises steps of carrying out graying processing on the original color image and eliminating noise interference by combining image filtering; and utilizing an improved Sobel edge operator to roughly position the edge of the preprocessed image, obtaining a block-shaped connected candidate region of the binary image through refinement processing such as mathematical morphology operation and the like, and sending the block-shaped connected candidate region to a deep learning neural network Faster RCNN model for accurate positioning and identification. According to the method, candidate areas are intercepted through edge detection, so that the interference of the background on the detection performance is reduced, and meanwhile, an improved Sobel operator focuses on detecting the edge in the vertical direction in combination with the characteristics of shelf cigarette pictures; according to the method, the Faster RCNN detection model modifies the anchor frame size and proportion in the area suggestion network according to the cigarette size and shape characteristics, so that the missed detection probability of small targets is reduced, and the detection and recognition accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a cigarette brand recognition method in complex scenes. Background technique [0002] my country is the largest tobacco producing and consuming country in the world. Cigarettes, as a kind of tobacco products, rank among the top in the world in terms of production, sales and growth rate. According to the different tobacco components and production processes, the quality of cigarettes is quite different. Tobacco companies divide cigarettes into different brands, and use the shell pattern, text logo and other characteristics to distinguish them. At present, the classification of cigarette brands is mainly manual, and some technologies for automatic classification using barcode information have also been developed. However, for a large number of densely arranged cigarettes, such as supermarket cigarette cabinets, some cigarettes have a relatively similar appearance, and the ...

AI Technical Summary

Problems solved by technology

However, due to the complex background, small scale and seamless arrangement of cigarettes, only using the Faster RCNN and R-FCN algorithms cannot accurately and comprehensively obtain the location and brand information of the target

[0006] To sum up, for the brand recognition problem of cigarettes on shelves with complex background, dense arrangement and small size, the existing end-to-end target detection algorithm still has certain defects, and a systematic, high-precision, computational Less complex brand recognition and statistical methods

Images