A prohibited goods non-box-opening detection method for the postal industry

Abstract

The invention belongs to the field of goods package security check in the postal industry, and discloses a prohibited goods non-box-opening detection method for the postal industry. The method comprises the following steps: acquiring an X-ray image of a to-be-identified package by a security inspection machine; inputting the X-ray image of the to-be-identified package into a preset deep learning model, and extracting corresponding article information data in the X-ray image; detecting and identifying the received article information data by using a deep learning model; and generating a corresponding identification result according to the information judged as the suspicious prohibited article, wherein a training sample of the deep learning model is obtained through three steps of image matting, augmentation and fusion of prohibited goods and simulated scene images. According to the method, an effective training sample is provided for prohibited goods detection, and the security check efficiency and accuracy of prohibited goods in the postal industry by utilizing a deep learning method are further improved.

Description

technical field [0001] The invention relates to the field of postal goods package security inspection, in particular to a detection method for unopened contraband used in the postal industry. Background technique [0002] In recent years, with the rapid development of my country's postal industry, various illegal and criminal acts using logistics delivery channels have also shown a high incidence. Criminals take advantage of the management loopholes of "not real name" and "non-verification" of logistics delivery channels to use logistics delivery channels Engaged in trafficking in drugs, guns and ammunition, controlled knives, dangerous chemicals, explosives, regional sales of goods, items that need to be investigated by the public security, and vicious incidents that endanger public safety are increasing. Logistics delivery has become a crime for criminals to engage in trafficking in contraband one of the main channels. For this reason, it is particularly necessary to carry ...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a method for detecting contraband in the postal industry without opening the box, so as to solve the problem of security inspection of postal goods and parcels proposed in the above background technology, and improve the efficiency of security inspection of contraband in the postal industry using deep learning methods and accuracy