3D segmentation denoising technology of millimeter wave human body security imaging

Abstract

AbstractTo meet the demand of noise reduction in millimeter‐wave human body security imaging, this paper proposes a new method for 3D segmentation denoising in millimeter‐wave images. The test results indicate that with the incorporation of 3D segmentation denoising technology, the noise in the background area of millimeter wave images has decreased by approximately 20–40 dB, significantly improving image quality. The detection rate has increased from 90% to 95%, while the false positive rate has decreased from 13% to 5%. This has important practical significance for real‐world applications.