A millimeter wave imaging dangerous goods detection method based on FPGA and depth learning

Abstract

The invention discloses a millimeter wave imaging dangerous goods detection method based on FPGA and depth learning, which comprises the steps of obtaining the millimeter wave imaging pictures and identifying dangerous goods in the pictures; constructing the depth learning network model of hazardous materials detection, and using the labeled millimeter-wave imaging images to train the model to getthe trained detection model; locating the trained detection model into the detection system of FPGA platform; using the FPGA to detect the millimeter wave image to be measured; using the trained detection model to detect the collected millimeter-wave images; if the image to be measured contains hazardous materials, marking the hazardous materials category, alarming, and marking the coordinates ofthe irregular outline position of the hazardous materials at the same time. Based on the FPGA-based platform, using the depth neural network model trained by Mask R-CNN object detection framework forthe real-time foreign body detection, so that compared with ASIC chips, the method has better customizability.

Description

technical field [0001] The invention relates to a millimeter wave imaging dangerous object detection technology, in particular to a millimeter wave imaging dangerous object detection method based on FPGA and deep learning. Background technique [0002] Millimeter waves generally refer to electromagnetic waves with frequencies between infrared and microwaves in the 26.5-300 GHz band, which are in the transition stage from macro electronics to micro photonics. Compared with microwaves, millimeter waves have higher bandwidth characteristics and resolution capabilities; compared with light waves, millimeter waves have higher energy conversion efficiency and can image the surface of certain substances; compared with X-rays, their electron energy low, will not cause harm to human tissue. Active millimeter-wave imaging uses millimeter-wave transmitting devices to irradiate the inspected person and the items they are carrying, and then uses a receiving antenna to collect reflected ...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is that the existing technology cannot accurately extract the irregular contours of dangerous goods, and a method for detecting dangerous goods based on millimeter wave imaging based on FPGA and deep learning is provided.

Images