Terahertz time-domain spectroscopy hidden dangerous goods classification method based on fusion of ResNet and LSTM

Abstract

The invention discloses a terahertz time-domain spectroscopy hidden dangerous goods classification method based on fusion of an ResNet and LSTM. The method comprises the following steps: collecting terahertz time-domain spectroscopy data for a dangerous goods sample to construct a data set, and carrying out preprocessing on the data in the data set; and constructing a ResNet-LSTM network model, then training, testing and evaluating the network model by using the preprocessed data set, and finally obtaining a trained network model for real-time classification of dangerous goods. According to the invention, with the combination of passive terahertz human body imaging security inspection equipment and a terahertz time-domain spectroscopy technology, the security defense capability of public places in cities can be obviously enhanced, the occurrence rate of public security incidents is effectively reduced, and the economic loss caused by the incidents is greatly reduced; and the invention is of great significance to maintain social security and stability.

Description

Technical field [0001] The present invention relates to the field of dangerous goods detection, and more particularly to the classification method of tachz time domain spectrum hidden dangerous goods based on RESNET and LSTM fusion, mainly available on passive terahertz human security equipment. Background technique [0002] X-ray imaging technology is the most conventional method of exploding the explosives and drugs in the package and luggage, but special explosives such as sheets and liquids can be susceptible to missed inspections. The type of dangerous goods is also difficult to confirm, and due to the human body There is a certain radiation damage and cannot be used for the examination of the person. Others such as police dogs and tracer methods are limited to well-packaged dangerous goods. Therefore, in order to find accurate, convenient, safe and economical dangerous goods detection technology is imminent. [0003] With the development of Tharaz Technology, Taihazbo is wi...

AI Technical Summary

Problems solved by technology

[0006] The present invention proposes a terahertz time-domain spectral hidden dangerous goods classification method based on the fusion of ResNet and LSTM, focusing on solving the accuracy and speed of human body hidden goods classification in subway security inspection, so that the classification accuracy and classification speed of dangerous goods match terahertz The speed of imaging human body security inspection equipment meets the needs of practical applications

Images