An intelligent settlement method for dish recognition based on machine vision and neural network

Abstract

The invention discloses an intelligent settlement system for dish recognition based on machine vision and neural network, comprising the following steps: S1, entry of dish characteristics; S2, detection of dish characteristics; S3, recognition of dish objects; S4, according to The recognition result calculates the actual price. Based on machine vision and neural network, the present invention obtains multi-angle feature information of the image, and then through the means of mapping, speeds up the speed of feature matching. Chip cutlery reduces costs.

Description

technical field [0001] The invention relates to digital image processing technology and deep learning technology, in particular to an intelligent settlement system that realizes dish recognition based on machine vision and neural network. Background technique [0002] With the development of society and the popularity of self-service, more and more restaurants have given up the inefficient traditional manual checkout and adopted the high-efficiency and low-cost self-checkout service. [0003] The Chinese invention patent with application number CN2017102765124 records a settlement system for smart restaurants based on machine vision. It reads the real-time video stream through the camera, the background program extracts key frames, and removes noise from the key frames, transforms them in grayscale, and blurs them. After Hough transform, the fuzzy neural network algorithm recognizes tableware with specific shapes and colors. The recognition rate of this method is greatly af...

AI Technical Summary

Problems solved by technology

[0006] The invention provides an intelligent settlement method for dish recognition based on machine vision and neural network, aiming to solve the problems of slow self-checkout speed, low recognition rate and high cost

Images