Automatic detection and identification method for hidden articles

Abstract

The invention discloses an automatic detection and identification method for hidden articles. The method comprises the following steps of: performing millimeter-wave scanning on a detected person to obtain an original image; regulating the original image to obtain a target image; segmenting and locating parts of a human body according to the target image; generating a bar combination model; detecting a non-human body target according to the original image to obtain a non-human body target distribution original image; obtaining position distribution information of the non-human body target distribution original image relative to the human body through the bar combination model; and identifying the type of the non-human body target and displaying the position distribution information of the hidden articles relative to the human body. By the automatic detection and identification method for hidden articles provided by the invention, a manual mode of detection and identification for the hidden articles is changed to an automatic mode, demands on staff is lowered, human error is reduced, and the detection and judging time is shortened.

Description

technical field [0001] The present invention relates to detection technology in the field of security inspection, and more specifically, the present invention relates to an automatic detection and identification method for concealed objects based on millimeter-wave imaging. Background technique [0002] In the field of security inspection, there are many ways to detect the human body and its hidden objects: metal detectors, X-ray perspective, infrared detection, and millimeter wave detection. Metal detectors are realized by electromagnetic induction. They can only judge the presence or absence of metal objects, but cannot image or determine the location of objects. X-ray fluoroscopy equipment has a strong penetrability and is generally used for the detection of luggage items. If it is directly detected on the human body, it will be more dangerous to people, so it is generally rarely used for human detection in security inspections. Infrared detection uses the thermal radiat...

AI Technical Summary

Problems solved by technology

Although image segmentation techniques based on grayscale multi-threshold, boundary extraction, edge detection, region segmentation, wavelet transform, morphology, fuzzy mathematics, genetic algorithm, neural network, information entropy and other methods have been tried and applied in the automatic detection of hidden objects , but because it is out of touch with the human visual mechanism, only using the grayscale and spatial information in the image to segment the image will still cause inconsistencies with human vision

The positioning analysis method based on the prior model of the human body is applied in the motion tracking of the human body, which reduces the complexity of tracking, which mainly includes such as Figure 32 The strip model shown, Figure 33 However, since the strip model only contains the outline information of the human body, such as structure, shape, posture, etc., and the stick model only contains the joint points of the human body, it can only be limited to the detection of the human body, and it cannot directly solve the problem of hidden objects in the human body. automatic detection and identification of issues

[0006] Third, through millimeter-wave scanning imaging, hidden object information on the human body can be detected, but at the same time it will cause the exposure and display of human privacy (such as the face and private parts). How to analyze and process the image after millimeter-wave imaging to display hidden objects Shielding the private information of the human body is also a technical problem that needs to be solved in the security inspection system

Images