Auxiliary method, device and storage medium for micro-expression interview

Abstract

The present invention provides a micro-expression interview assistance method, device and computer-readable storage medium. The method comprises the following steps: generating stress questions and standard answers thereof; cutting the first video of the customer answering the stress questions to obtain the first video segment corresponding to each stress question; extracting the information contained in the customer's face in each first video segment The first and second action features; the feature subset formed by the first and second action features, and the fraudulent labeling of the video clip training support vector machine model to obtain the scoring model; collecting the second video of the customer answering the stressful question, cutting The second video obtains the second video segment corresponding to each stress problem; extracts the feature subset formed by the first and second action features contained in the customer's face from each second video segment; inputs the feature subset into the scoring model, Output the fraud scoring results for each stressor question answered by the customer. With the invention, it is possible to effectively identify whether the customer is suspected of fraud.

Description

technical field [0001] The invention relates to the technical field of video processing, in particular to a system and method for micro-expression analysis based on video. Background technique [0002] When a financial institution receives a large amount of credit loan applied by a customer, it usually requires an offline or remote interview by risk control personnel. The risk control personnel checks whether the relevant information of the borrower is fraudulent by asking questions. At present, checking whether a borrower is fraudulent relies heavily on the experience and judgment of risk control personnel. There is no tool on the market that can objectively and effectively identify whether a borrower is fraudulent. Contents of the invention [0003] In view of the above reasons, the present invention provides a micro-expression interview auxiliary method, device and storage medium, which can identify and analyze the micro-expressions of loan customers during the intervie...

AI Technical Summary

Problems solved by technology

At present, checking whether a borrower is fraudulent relies heavily on the experience and judgment of risk control personnel. There is no tool on the market that can objectively and effectively identify whether a borrower is fraudulent

Images