Article identification and classification method, storage medium and device

Abstract

The invention discloses an article identification and classification method, a storage medium and a device, and the method comprises the following steps: the infrared image information of a to-be-confirmed article is obtained, the infrared image information is input into a neural network, and the type and number information of the to-be-confirmed article is obtained, and the to-be-confirmed article is an article externally provided with a transparent film package, the neural network is trained to receive input infrared image information of articles with transparent films and output the types of the articles in the transparent films, the confidence coefficient of the types, the number of the articles and the confidence coefficient of the number, and the method further comprises the step of calculating the total amount of bills according to the obtained types and number information of the articles to be confirmed. According to the technical scheme, the transparent film can be intelligently filtered, and the functions of objects in the transparent film are analyzed. And meanwhile, the total amount of the bill can be calculated according to the commodity classification result, so that the manual verification time and various costs are saved.

Description

technical field [0001] The invention relates to the field of neural network design, in particular to a method for identifying and classifying objects. Background technique [0002] Large and medium-sized shopping malls and supermarkets are increasingly deeply influenced by the Internet and intelligent technology, and a large number of new technologies have begun to be applied to large and medium-sized shopping malls and supermarkets, such as: self-service cash register, commodity inspection, commodity tracking, commodity loss prevention, automatic sorting, visual Settlement, smart payment, etc. Business intelligence and automation experience is getting better and better. For shopping malls and supermarkets, the consumption experience / payment mechanism is very important. Self-service cash register system includes: self-service calculation, automatic degaussing, intelligent payment, payment verification. Payment verification can verify the quantity and name of the consumer'...

AI Technical Summary

Problems solved by technology

In the payment verification technology, there are still many technical limitations and cannot play a role. The technical limitations include but are not limited to the following situations: 1. Shopping plastic bags have colors / patterns; 2. Commodity packaging bags blur the texture characteristics of commodities; The transparency of the bag is too low. 4. If you use an infrared camera, you can see through the plastic bag, but only black and white photos, the color features of the product disappear, and the detection error is large

Images