Self-learning security check method, security check system and security check door based on artificial intelligence

Abstract

The invention discloses a self-learning security check method, a security check system and a security check door based on artificial intelligence. The method comprises the following steps: performing signal acquisition on a security check object; amplifying, shaping and filtering the collected signal, recording the signal waveform of the security check object, carrying out data storage and algorithm processing, establishing a data model, carrying out point tracing restoration on the signal, generating a signal oscillogram, and forming a security check object sample database; repeatedly collecting the same security check object in the same area, and correcting and calibrating the sensitivity of the security check object sample database; all signals generated by the articles entering the security check area are compared with the big data sample library, and the articles are excluded or reversely excluded. According to the invention, a data model when an article passes through the security door can be restored, whether the passing article is consistent with a database sample graph is judged, the security door can adapt to the change of a detection environment through intelligent detection, real-time accurate detection is realized, and the security door is more intelligent and more accurate.

Description

technical field [0001] The invention relates to a self-learning security inspection method, a security inspection system and a security inspection door, in particular to an artificial intelligence-based self-learning security inspection method, a security inspection system and a security inspection door, and belongs to the technical field of intelligent security inspection equipment. Background technique [0002] The security gate of the prior art has a single function and can only detect metal, but cannot specifically distinguish the type and size of the metal. If the user needs to filter out certain small metals (such as belts, key chains, etc.) or daily necessities (such as mobile phones, etc.), Only by reducing the sensitivity (increasing the threshold) can certain small metals, belts, keychains and other metal objects of the same size be excluded. And it needs to manually set parameters, etc., which is a cumbersome operation. [0003] The security gates in the prior ar...

AI Technical Summary

Problems solved by technology

[0004] 1. The types of excluded items are limited, and the types of items need to be almost the same size;

[0005] 2. Do not have item classification or it is difficult to classify items;

[0006] 3. In the process of use, there is no basis for collecting, learning, storing and using sample data for equipment testing, and it does not have the conditions for intelligence and digitization;

[0007] 4. Does not have the anti-exclusion function, that is, the detection function for specific items;

[0008] 5. Intelligent calculation and calibration without equipment parameters

[0009] The above-mentioned problems have not been effectively solved, and people's requirements cannot be met.

Images