Security check system and method fusing visible light image, depth image and terahertz image

Abstract

The invention provides a security check system and method fusing a visible light image, a depth image and a terahertz image, which are used for improving the accuracy of a security check process and optimizing the quality of a security check image (terahertz image). The security check system comprises: a plurality of sensor modules configured to collect a visible light image, a depth image and a terahertz image of a scene including one or more target objects; the image synchronization module that is configured to synchronize the collected visible light image, the depth image and the terahertz image; the image registration module that is configured to perform registration on the visible light image, the depth image and the terahertz image after synchronization; the image fusion module that is configured to fuse the visible light image, the depth image and / or the terahertz image after registration; and an identification module configured to identify a target object carried by the target object based on the fused image.

Description

technical field [0001] The present invention relates to the technical field of security inspection, and more particularly relates to a system and method that integrates information from various sensors, which can improve the accuracy of security inspection schemes and improve display effects. Background technique [0002] Terahertz imaging technology can be used to achieve non-cooperative, fast, and non-inductive security checks for passengers, and has been widely used in customs, civil aviation airports and other scenarios. But this technology still has the following shortcomings: [0003] 1. The current mainstream method for detecting and identifying suspects in terahertz images is the deep learning model, but the effect of the deep learning model is very dependent on the image data used for training. Furthermore, the models used on real devices are all pre-trained. But in fact, the use scenarios of the device are very complex and diverse, especially the background of th...

AI Technical Summary

Problems solved by technology

When performing object recognition directly based on terahertz images of such a large scene, the accuracy of object recognition may drop due to the presence of people out of focus

[0005] 3. Since the terahertz image is a grayscale image and lacks details, it is difficult to accurately divide the body parts based on this image

Therefore, for the detected target, it is impossible to provide accurate information about which specific part of the human body the target is hiding in.

[0006] All of the above problems have brought challenges to the promotion of terahertz security technology.

And solutions that rely solely on terahertz images do not provide great results

Images