Face recognition-based information push method and system

Abstract

The invention relates to the field of information push, in particular to a face recognition-based information push method and system. The method comprises the following steps of: when a shot picture of a point-of-sale terminal changes, obtaining a shot picture at the current moment as a first image, and carrying out preliminary body silhouette recognition on the first image; if a body silhouette exists in the first image, playing audio / video data; after a preset time interval, obtaining a shot picture again as a second image, and if a body silhouette exists in the second image, carrying out accurate face feature recognition on the second image to obtain a face recognition feature value; and obtaining a corresponding customer type according to the face recognition feature value, and pushing corresponding customized information to the point-of-sale terminal according to the customer type so as to display the customized information. According to the method, preliminary body silhouette recognition is carried out on shot pictures, the attentions of customers are attracted through paying audio / video data, and pointed customized information is provided for the interested customers, so that the user experience is enhanced and the sale success rate is improved at the same time.

Description

technical field [0001] The present invention relates to the field of information push, in particular to an information push method and system based on face recognition. Background technique [0002] As an important branch of biometric recognition technology, face recognition technology has been widely used in various fields such as video surveillance and human-computer interaction due to its advantages of non-contact and easy collection. [0003] With the development of science and technology, sales terminals are equipped in all major physical stores, and the recommended products and promotional information of the store can be displayed to customers entering the store through the customer display screen on the sales terminal. [0004] However, the current point-of-sale terminal has relatively simple functions, and can only play unified product recommendation information, and cannot make customized recommendations for each customer, cannot automatically determine whether the ...

AI Technical Summary

Problems solved by technology

[0004] However, the current point-of-sale terminal has relatively simple functions, and can only play unified product recommendation information, and cannot make customized recommendations for each customer, cannot automatically determine whether the current customer is a new customer, an old customer, or a member customer, and cannot provide Data such as consumption records and consumption habits of old customers or member customers, and the gender, age, race and other characteristics of new customers cannot be obtained

When the product is traded, it is impossible to effectively count the types of buyers corresponding to each product, and it is even impossible to know the type of the main customer group entering the store.

Images