Tableware-pattern-based pricing method of dish automatic identification system

Abstract

The invention relates to a tableware-pattern-based pricing method of a dish automatic identification system. According to the tableware-pattern-based pricing method, patterns on dinner plates are regarded as identification objects, specific figures or marks are printed on edges of the tableware, each specific figure or mark represents the type or price of a kind of dish, and a machine vision system can be used for identifying the types or prices of dishes in the tableware through reading the figures or marks printed on the edges of the tableware, thereby performing automated pricing and being capable of improving automation efficiency of a restaurant significantly. Compared with the existing container identification method, the tableware-pattern-based pricing method is not liable to be limited by containers, does not need to use special tableware as the containers, can just adopt the tableware with different figures and patterns, and is low in cost. The tableware-pattern-based pricing method can traverse a plurality of regions, can identify a plurality of dinner plates simultaneously after one-time shooting, and does not need to scan the dinner plates one by one. The tableware-pattern-based pricing method identifies the dinner plates quickly through machine vision, the form of identification is not limited by tableware combination, and the tableware-pattern-based pricing method is more intelligent.

Description

technical field [0001] The present invention relates to the fields of machine vision and machine learning, and more specifically relates to a pricing method for an automatic dish recognition system based on tableware patterns. The present invention can be widely used in canteens, fast food restaurants and other places that provide self-service meal selection and settlement services . Background technique [0002] As the pace of urban life continues to accelerate, people are increasingly using various types of fast food to solve the problem of "food". Then choose and settle by queuing up to swipe your card or cash settlement. And how to calculate the price of the selected dishes, there are two kinds of manual pricing and automatic pricing in the existing technology. Due to the increase in the number of diners, the efficiency of manual meal delivery and price calculation is low. During the peak consumption period, the slow settlement speed often leads to queuing phenomenon, ...

AI Technical Summary

Problems solved by technology

"The image information mentioned in it includes the size, shape and color of the dinner plate, and the acquired image information of the dinner plate is recognized through image recognition technology. However, to recognize the color and shape of the dinner plate itself, it is necessary to restrict the color and shape of the tableware. and limited combinations

On the other hand, this solution uses a handheld device to scan the codes of the dishes one by one, and can only manually identify the tableware in sequence, which reduces the efficiency of the entire process

[0011] The above-mentioned existing settlement methods based on image recognition still do not have a solution that can solve the above-mentioned problems

Images