Terahertz image target recognition method based on deep learning and RPCA

Abstract

A Terahertz image target recognition method based on deep learning and RPCA comprises the following steps of (1) removing the background noise by using a robust principal component analysis (RPCA) method; (2) using shape priori knowledge to reduce the region of interest; (3) generating terahertz image data sets; (4) training deep learning network Faster-RCNN; (5) using the deep learning network Faster-RCNN for target recognition. The method removes the background noise interference of the terahertz image by using the robust principal component analysis (RPCA) method, reduces region of interestby using shape prior knowledge, uses the depth learning network Faster-RCNN to perform the target recognition on the region of interest, so tha the method of the invention uses the robust principal component analysis RPCA and the shape prior knowledge to overcome the influence of the background noise, and can detect the object in the terahertz image very quickly and accurately.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a terahertz image target recognition method based on deep learning and Robust Principle Component Analysis (RPCA) in the technical field of image recognition. The invention can be used for target detection and recognition on terahertz security images in the field of public security. Background technique [0002] Terahertz waves (THz waves) include electromagnetic waves with a frequency of 0.1 to 10 THz. The term applies to frequencies between the high-frequency edge (300 GHz) of the millimeter-wave band of electromagnetic radiation and the low-frequency far-infrared spectral band edge (3000 GHz), corresponding to wavelengths of radiation in this frequency band ranging from 0.03 mm to 3 mm. Simply put, terahertz waves are a non-contact, non-destructive detection method, and can penetrate high-density molecular structures that many imaging techniques (such as ultras...

AI Technical Summary

Problems solved by technology

The deformable part model DPM splits and transforms the detection problem of the whole target in the traditional target detection method into the detection problem of each part of the model, but the method still has the disadvantage that the matching calculation of the model and the target is very large , resulting in greatly increased recognition time and cannot be used for real-time detection of terahertz images

However, the disadvantage of this method is that it does not use the characteristics of background similarity of terahertz images to remove a large amount of background noise. The region proposal network RPN directly generates candidate regions for the entire terahertz image, and the accuracy rate is not high in the application of real-time detection. , The detection time is long, and it cannot be used for real-time detection of terahertz images

Images