TensorFlow-based security check contraband identification system and method

Abstract

The invention provides a TensorFlow-based security check contraband identification system and method, and the system comprises an X-ray three-dimensional imaging module which is used for collecting anX-ray image of an object carried by a passenger; the image processing module which is used for constructing a neural network model through a TensorFlow framework so as to carry out feature extractionprocessing on the X-ray image to obtain a preprocessed image; the detection module which is used for carrying out dangerous goods identification and judgment training on the preprocessed images through a neural network model, and is used for constructing a dangerous goods sample library through sample training so as to carry out matching; the debugging module which is used for transmitting the X-ray image to the manual end equipment for debugging when the detection module cannot judge the danger of the detected object, and memorizing a debugging result to reinforce the dangerous object samplelibrary; and the dangerous goods prompting module which is used for outputting dangerous goods related information to remind related personnel when the detection module detects the dangerous goods.

Description

technical field [0001] The invention relates to the technical field of machine vision recognition, in particular to a TensorFlow-based security inspection contraband recognition system and method. Background technique [0002] As a means of transportation for our daily travel, the subway has the characteristics of convenience, speed and comfort. However, problems such as large passenger flow and complex passenger groups will bring certain safety hazards to the society. For example: (1) Because the subway is generally built in an underground place, there is little space for people to move around. In the event of fire, explosion and other accidents, it will be difficult for passengers to escape; , these criminals may carry prohibited items (knives, flammable items, explosives, etc.), and then appear to destroy public equipment, injure others, etc., seriously endangering the safety of passengers; (3) Although there are clear legal provisions, vigorously publicize and call on ci...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a security inspection contraband identification system based on TensorFlow to solve the problems of low identification efficiency, false detection and missed detection in existing X-ray security inspection machines

Images