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

An Image Recognition-Based Integrity Standardization Method for Catering Receipts

An image recognition and integrity technology, applied in the field of image recognition, can solve the problems of many intermediate links, serious integrity of behavioral integrity data, large loss, etc., to achieve strong flexibility and growth, avoid complexity, and reduce complexity.

Active Publication Date: 2021-07-20
HUNAN UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In this process, there are many intermediate links, large losses, and difficulties in information communication and synchronization, so problems may arise in almost every link, especially behavioral integrity and data integrity.
Behavioral integrity can be manifested in chefs and suppliers stealing food ingredients, which damages the interests of restaurant owners. Data integrity can be manifested in the process of receiving goods, the receiving environment is dirty and messy, and the quality and authenticity of ingredients are recorded through paper records. The accuracy cannot be guaranteed, and the data is manually entered into the system, which has problems such as low degree of informatization and standardization

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
  • An Image Recognition-Based Integrity Standardization Method for Catering Receipts
  • An Image Recognition-Based Integrity Standardization Method for Catering Receipts
  • An Image Recognition-Based Integrity Standardization Method for Catering Receipts

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0083] First, use the Mealcome data set to carry out experiments to evaluate the model proposed by the present invention, then, compare the model of the present invention with existing models such as ResNet and VGG-16, and evaluate respectively from two different aspects;

[0084] 1) Data set and experimental environment,

[0085] The Mealcome dataset (MLC dataset) is provided by China’s large-scale food supply chain platform (www.mealcome.com), which provides services for nearly 1,000 restaurants; the dataset consists of three parts: Integrity Data (MLC-IP) , non-integrity data (MLC-NP) and data containing purchase orders (MLC-PO); the invention obtains the original food material images sorted by date, each folder contains all images generated on the day, and all food material images are on-site Therefore, there are too many images containing overexposure, dark light, plastic bag packaging, etc., so the present invention uses these images to create MLC-NP, and uses all clear ...

Embodiment 2

[0096] Experimental evaluation of the Enhanced Order-CNN model,

[0097] The data set used in this experiment contains the purchase order information of each image; in the training process of the Enhanced Order-CNN model, the input is the order information with images and labels, but the original Caffe framework cannot support this multi-label Input and training, so modify the source code of the Caffe framework to add a new layer "order_weightd_type"; such as Figure 6As shown, the newly added layer "order_weight_type" (Order-weight layer) in Enhanced Order-CNN is shown;

[0098] Considering that weight information is an important feature in the order of ingredients, the present invention adds weight information to the back of each order information, and then compares it with the Order-CNN model that does not involve weight features; Figure 7 As shown, you can see the file list with weight information added, so the Caffe framework will add the obtained weight information to ...

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 method for standardizing the integrity of food and beverage receipt based on image recognition, including an Auto-Integrity CNN model for solving the problem of image recognition in the environment of honest receipt and an Enhanced Order-CNN model for solving the problem of food material image recognition based on combined orders The construction method of the CNN model; the image recognition-based catering receiving integrity specification method of the present invention adopts the Auto-Integrity model to improve the Caffe framework and the CaffeNet model for solving the image recognition problem of the integrity receiving environment; adopts Enhanced Order- The CNN model improves the Caffe framework and the CaffeNet model, and combines the type, weight, order of purchase and other business information in the order to solve the problem of food image recognition based on combined orders.

Description

technical field [0001] The invention relates to an image-recognition-based method for regulating the integrity of food and beverage receipts, and belongs to the technical field of image recognition. Background technique [0002] The traditional food procurement process includes four links: order declaration, stock preparation, goods receipt, and account reconciliation. Specifically: first, the chef needs to count the food material gap, hand it over to the head chef for review, and then call the supplier to prepare, and the supplier has finished stocking After delivering the goods, the store will receive and enter the goods, and finally the financial department will be responsible for the account reconciliation. In this process, there are many intermediate links, large losses, and difficulties in information communication and synchronization. Therefore, problems may arise in almost every link, especially behavioral integrity and data integrity. Behavioral integrity can be ma...

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 Patents(China)
IPC IPC(8): G06K9/62
CPCG06F18/24147
Inventor 肖光意吴淇刘欢刘毅黄宗杰陈浩
Owner HUNAN 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