The first is to verify the borrower's identity by checking the borrower's name and ID number, which becomes the two major elements of citizen information in the risk control industry, and then cooperates with the mobile phone number and bank card to achieve the consistency of the four major elements. The advantage of this method is simple and convenient, but the disadvantage is that because there are too many small white data sold on the black market, the risk of being cheated is too high, and it is easy to form a number of fraud attacks.
The second is to double-check the borrower's name, ID number and public security photos through the data of the Citizenship Information Center. Compared with the first method, this method has one more face comparison, which will be more secure. But the shortcomings are also very obvious: first, the service is unstable (affected by policies). Due to the increasing protection of citizenship information by laws and regulations in the past two years, we will resolutely crack down on illegal caching and reselling of citizenship information. In particular, on September 30 this year, the Citizenship Information Center will implement a comprehensive ban on providing screen photos to the outside world, and all portrait comparisons are not allowed at customers to ensure the protection of citizen information. Under such a high-pressure policy, not all authentication vendors can provide services stably. Secondly, the legitimacy of data sources cannot be guaranteed. In the wave of authentication services in recent years, many companies have done data caching and reselling. Without the eyes of insiders, it is easy to use cached data, which will affect the accuracy of verification.
Third, through AI image recognition technology, intelligent image algorithm is used to identify the authenticity of ID cards, and citizen information and mesh photos are combined for multiple verification and protection, which can even prevent the cloning of certificates! The advantage of this method is that only one ID photo is needed to identify the authenticity of the certificate, and because it is AI image recognition technology, the stability of the service is extremely high.
Beijing Situ scenario integrates AI technology into financial business processes, and reconstructs financial anti-fraud business scenarios with advanced AI technology, massive data and financial scenario understanding ability. It is the first ID card original image recognition algorithm in China to realize the ability of distinguishing authenticity of an image.
Situ scene official website introduced.
The in-situ scenario provides three product modules for the Internet financial risk control platform:
First of all, it is a single technical function product: face recognition, image recognition, expression recognition and certificate authentication.
Secondly, the solution for specific business content: "Shentong" intelligent interview. This product provides financial services such as full-process authentication, identity authentication, loan willingness verification, AI quality inspection, etc., and reserves a chain of evidence with judicial effect for the "authentication" of transaction effectiveness.
The third is the image intelligent platform. Through deep learning technology, the product can automatically identify and classify pictures, slice key contents and recognize characters, and can sort and classify a large number of documents with one click, such as on-site identification, authenticity identification, signing or not, and confirmation of different versions of contracts and documents. The platform can reduce a lot of manpower and cost for customers and speed up business processing. The system can be deployed on the local server without data outflow.