A Compound Noise Reduction Algorithm For Terahertz Imaging

Terahertz transmission imaging has unique advantages such as non-destructive testing compared with other transmission imaging methods. However, due to the low power of terahertz laser and the obvious diffraction effect of imaging, there is usually a lot of white noise in the system in transmission imaging. Since the system uses a high-density polyethylene lens, there will be additive noise such as Newton's rings in the image. In this paper, an improved mean filtering composite denoising algorithm is used to reduce the noise of the image with a high information entropy, which solves the problem of additive noise and white noise in terahertz transmission imaging. To achieve the purpose of improving visual effects, eliminating noise interference and highlighting the imaging subject. And provide research basis for terahertz imaging image analysis, target detection and feature extraction.