Shelf commodity layering method based on deep learning

Abstract

The invention discloses a shelf commodity layering method based on deep learning, and relates to the technical field of artificial intelligence and machine vision deep learning. The method comprises the following steps: S1, obtaining shelf pictures with different angles, different illumination and different resolutions; S2, training a commodity detection model; S3, training a shelf layer segmentation model; S4, detecting the shelf image to obtain a position box and category information of the commodity; S5, carrying out shelf layer segmentation on the shelf image to obtain a shelf layer area;S6, obtaining a minimum enclosing rectangle of the shelf layer, and then obtaining a shelf line; and S7, obtaining a central point of the commodity position box, and then judging a shelf layer where the commodity position box is located. According to the method, commodities on the shelf can be automatically layered to obtain the number of shelf layers where the commodities are located, so that whether the commodities are displayed at gold positions can be judged. The method is efficient and accurate, and replaces a low-efficiency and error-prone manual mode.

Description

technical field [0001] The invention belongs to the technical field of deep learning of artificial intelligence and machine vision, and in particular relates to a method for layering shelf goods based on deep learning. Background technique [0002] In the new retail era, in order to better control the sales of offline products, retail manufacturers usually send sales agents to supermarkets to take pictures and inspect, and then review the display of products. Among them, whether the goods are displayed in the "golden position" is very important. "Golden position" usually refers to the shelf position that consumers can easily see and get. For example, the golden position of the 5-tier shelf is usually the second and third floors (counting from top to bottom). [0003] To judge whether the product is displayed in the golden position, the shelf needs to be layered first, and then the number of layers of the shelf where the product is located is obtained. At present, this info...

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a shelf product layering method based on deep learning, which solves the problem of low efficiency and error-prone manual review

Images