information community file
All customer cases - Rongxun Weiye
Rongxun Weiye
Rongxun Weiye is a high-tech enterprise focusing on enabling offline physical stores through AI technology. The AI intelligent cashier equipment developed by him has been put into use in Dalian lock stores in Beijing, creating a precedent for the application of artificial intelligence in physical stores.
Use product
Support and communication
EasyDL helps intelligent settlement and further intelligent retail
Value achievements
Rongxun Weiye trained the fruit and vegetable recognition model through Baidu EasyDL zero threshold AI development platform, and combined with the existing electronic scale equipment, realized the self-service checkout of fruit and vegetable ingredients in physical stores. The accuracy rate of the model is up to 97% or more, and it only takes 2-3 seconds to complete the intelligent checkout, which is 15-20 times more efficient than the traditional manual checkout, which saves each store an average of 120000 yuan/year in labor costs. Now, the smart fruit and vegetable electronic scale deployed with the special SDK for EasyDL model has effectively improved the checkout efficiency and accuracy of supermarket convenience stores, which has been widely praised by merchants and customers, and has been used by many large chain stores.
Case Story
Case Story
Core demands
When people go to the offline supermarket convenience store to buy daily necessities such as fruits and vegetables, they often have to go through the complex checkout process of packaging and pricing, manually pasting barcodes, and queuing to scan the code one by one. This has led to the continuous occurrence of customer crowding, crowded queuing, and crowded checkout areas, resulting in a poor shopping experience for customers. At the same time, the operating costs of traditional offline retail stores are rising year by year, and the labor costs are growing, so it is increasingly difficult to recruit cashiers. Therefore, how to reduce labor costs, improve customer experience, and efficiently and accurately checkout has become a major challenge for new offline retail stores.
Solution
Rongxun Weiye trained about 40000 images of more than 500 kinds of fruits and vegetables collected by offline supermarket stores through Baidu EasyDL customized image recognition professional version, and obtained the "supermarket fruit and vegetable recognition" model. At the same time, after deploying the special SDK for EasyDL model, the intelligent electronic scale developed by the company can accurately identify the varieties of fruits and vegetables and assist in completing the independent cashier settlement. It only takes 1s to obtain the identification results, and the accuracy rate of the model obtained in practical application is more than 97%.
The specific operation process is as follows:

Step 1: When customers settle accounts after shopping in supermarkets, they will place fruits and vegetables in the intelligent electronic scale identification area for taking photos;

Step 2: The electronic scale cashier system will identify the fruit and vegetable commodities purchased by the user through the EasyDL professional training model, and the identification results and the calculated amount of commodities will be displayed in the cashier system;

Step 3: After the user finishes the payment, the system displays a prompt indicating that the payment was successful.
Technical capability
Voice technology
Character recognition
Face and Human Body
Image technology
Language and knowledge
video technique