Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Convolutional neural network-based self-service checkout system and convolutional neural network-based self-service checkout method

A convolutional neural network and self-checkout technology, which is applied in the field of self-checkout systems based on convolutional neural networks, can solve the problems of heavy workload of waiters, slow settlement time, and disrupted dining environment, so as to enhance information transparency and reduce waiting. time, the effect of optimizing the user experience

Inactive Publication Date: 2019-07-16
EAST CHINA NORMAL UNIV
View PDF4 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Among the three traditional modes, the settlement method is mainly manual calculation, and the settlement time is relatively slow, causing diners to wait in line for a long time, especially in corporate canteens and school student canteens, where there is a dense flow of people and concentrated dining time, manual settlement It often leads to "long queues" in the queue, occupying the precious lunch time of diners and delaying their plans; manual settlement relies on the "paperless calculation" of waiters in just tens of seconds, which is easy to make mistakes, causing organizations or individuals Property loss; in addition, the workload of the waiters is heavy, and the restaurant relies too much on manpower, so there is a large demand for waiters, resulting in high labor costs
From this point of view, the traditional three models are facing problems that need to be solved urgently under the checkout method of manual settlement.
[0005] At present, a self-service settlement system developed based on RFID radio frequency technology, represented by the "Smart Disk" system, has emerged. Compared with the traditional manual settlement method, it saves the waiting time of diners, simplifies the dining process, and reduces the burden on canteen workers. Some progress and breakthroughs have been made; however, the self-service settlement system designed based on RFID radio frequency technology is not optimistic. Problems, such as: high error rate and lack of accuracy; slow operation speed, there will still be long queues; high manufacturing costs, high maintenance and repair costs, cannot effectively reduce canteen expenses; the workload of restaurant staff has not decreased but increased ; Jamming occurs when the card is read after settlement; based on the color of the price plate, it is difficult for diners to judge the price; there is noise during the operation of the self-service settlement system, disturbing the dining environment, and poor user experience; therefore, the existing The operation mode and principle of the self-service settlement system developed based on RFID radio frequency technology, represented by the "Zhipan" system, still needs to be improved.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Convolutional neural network-based self-service checkout system and convolutional neural network-based self-service checkout method
  • Convolutional neural network-based self-service checkout system and convolutional neural network-based self-service checkout method
  • Convolutional neural network-based self-service checkout system and convolutional neural network-based self-service checkout method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] Embodiment 1 neural network model algorithm

[0058] In recent years, the neural network algorithm has achieved great success in the field of image recognition, especially the deep learning algorithm evolved from the neural network, which can finally enable the entire recognition algorithm to obtain an accuracy rate of more than 90%. Moreover, this scheme can discard the training set after completing the learning, and complete the classification of a picture within one millisecond, and its efficiency can greatly meet the recognition requirements of multiple fields.

[0059] The field of neural network algorithms was originally inspired by the goal of modeling biological nervous systems, but then diverged from it as an engineering problem, with good results in the field of machine learning. The basic unit of computation in the brain is the neuron. There are about 86 billion neurons in the human nervous system, and they are connected by about 10^14-10^15 synapses. Such ...

Embodiment 2

[0064] Self-checkout algorithm of embodiment 2 convolutional neural network

[0065] Convolutional neural networks are very similar to regular neural networks: they are made of neurons with learning-capable weights and biases. Each neuron gets some input data, and after the inner product operation, the activation function operation is performed. The entire network is still a derivable scoring function: the input of the function is the original image pixels, and the output is the score of different categories. In the last layer (often a fully connected layer), the network still has a loss function (such as SVM or Softmax), and various tricks and points implemented in neural networks still apply to convolutional neural networks.

[0066] A simple convolutional neural network consists of various layers arranged in sequence, and each layer in the network uses a differentiable function to pass activation data from one layer to another. Convolutional neural networks are mainly com...

Embodiment 3

[0072] Embodiment 3 Self-checkout method based on the above-mentioned self-checkout algorithm

[0073] In this embodiment, the self-checkout method based on convolutional neural network includes the following steps:

[0074] Step a. Establish a self-checkout system based on the neural convolutional network;

[0075] Step b. photographing the selected dish image information, and identifying the actual selected dish quantity and type in the dish image information through the self-checkout system;

[0076] Step c. Calculate the price and complete the checkout according to the quantity and type of dishes actually selected.

[0077] The specific implementation process is as follows:

[0078] First, collect photos of common dishes in the canteen, identify the corresponding types, and build a dish database. After having the dish database, it is necessary to train the convolutional neural network of the self-checkout system so that the convolution kernel has an appropriate weight f...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a convolutional neural network-based self-service checkout method. The method comprises the following steps of establishing a self-service checkout system based on a neural convolutional network; shooting the selected dish image information, and identifying the number and types of actually selected dishes in the dish image information through the self-service checkout system; and according to the actually selected dish quantity and type, calculating the price and completing the checkout. According to the self-service checkout system, the dining process can be simplified, the ordering system can be optimized, inconvenience and complexity caused by manual settlement can be overcome, and the cost can be reduced. The invention also discloses a self-service checkout system based on the convolutional neural network.

Description

technical field [0001] The invention belongs to the technical field of network automatic checkout, and relates to a self-checkout system and method based on a convolutional neural network. Background technique [0002] In the current enterprise or institution employees, student canteens, and profit-oriented restaurants, self-selection, set meals, and window are the three more common traditional ordering modes. Self-selection mode means that the waiter orders dishes by portion and puts them at the window. Customers take dishes according to their preferences and put them on the tray. ;The set meal mode means that the dishes are selected in the prescribed set meal, and the waiter prepares the dishes, and the cashier settles according to the price of different set meals; the window mode means that diners line up at multiple windows to choose dishes, and the waiter After ordering dishes, the waiter will calculate the total amount after selecting the dishes, and remind the diners...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06Q30/02G07G1/12
CPCG06Q30/0283G07G1/12G06V20/10G06V20/68G06F18/24147G06F18/241
Inventor 陆观浦剑
Owner EAST CHINA NORMAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com