Intelligent identification method for prohibited goods based on X-ray security inspection machine image

Abstract

The invention discloses an intelligent identification method for a prohibited goods based on an X-ray security inspection machine image. The method comprises steps: a data set for training is built; aunified prohibited goods identification model is trained, wherein the unified prohibited goods identification model comprises a shape classification model and a color classification model; the shapeclassification model is used to detect a to-be-detected image to output a shape detection result; the detection result is calculated and is inputted to the color classification model, and a color detection result is outputted; by using the shape detection result and the color detection result, the possibility of existence of a dangerous substance and the position are calculated; the model capableof detecting a dangerous substance and the to-be-detected image are compared, and the kind of the dangerous substance is marked and outputted. Intelligent machine identification replaces manual dangerous substance identification, the manual cost is saved, and the profit maximization is realized.

Description

technical field [0001] The invention relates to an intelligent identification method for contraband based on an image of an X-ray security inspection machine. Background technique [0002] X-ray security inspection device, also known as security inspection machine and luggage security inspection device, is an electronic device that completes the inspection by sending the checked luggage into the X-ray inspection channel by means of a conveyor belt. It is mainly installed in subways, airports, museums, government agencies, etc. that require security checks. place. Its detection principle: when luggage enters the X-ray inspection channel, it will block the package detection sensor, and the detection signal is sent to the system control part to generate an X-ray trigger signal, which triggers the X-ray source to emit X-ray beams. A fan-shaped X-ray beam passing through the collimator passes through the inspected items on the conveyor belt, the X-rays are absorbed by the inspec...

AI Technical Summary

Problems solved by technology

[0006] Each security inspection machine needs to be equipped with 1-2 security inspectors to view the images, which is time-consuming and laborious;

[0007] Most of the security inspection machines are set up in 24-hour open places, which requires the security inspectors to check the images of the security inspection machines all the time, which will inevitably cause fatigue. When people are tired, their attention and judgment will be weakened, which will affect the identification of contraband, and even will affect public safety

[0008] To sum up, the current model based on convolutional neural network training has certain limitations, contraband is diverse, and the scene is also complex and changeable, which requires the design of the convolutional neural network model to have enough layers, the training set The sample size must be large enough to ensure accuracy

Images