High signal-to-noise ratio and high contrast dark-field confocal imaging of subsurface defects

Abstract

The weak scattering characteristic of subsurface defects (SSD) and the scattering noise from substrate can result in a low SNR and low contrast DFCM image. By applying the proposed high signal-to-noise ratio imaging and details preserving noise reduction algorithm based on TSVD and PCA and an image-adjustment-function, the high signal-to-ratio (SNR) and high contrast dark-field confocal microscopy (DFCM) imaging of SSD is realized. The SNR of the DFCM image is demonstrated a meaningful enhancement of 8.71 times. The proposed method is demonstrated on microscopic images with micron sized SSD structures. At last, we use the processed images for 3D reconstruction. The reconstructed results have higher SNR and contrast and can display more details of SSD.