Large dynamic range dark-field imaging based on microscopic images fusion

Abstract

High-quality dark-field confocal microscopic images are vital to the detection of optical components subsurface defects (SSD). The SSD images obtained by dark-field confocal microscopy (DFCM) has low dynamic range and low image brightness, which results in a loss of image details and a loss of weak scattered signals from micro-scale SSD. This paper presents a microscopic image fusion technique to extend the dynamic range, enhance the root-mean-square contrast and improve the SNR of DFCM images. The proposed method is demonstrated on the samples with micro/nano subsurface structures. The best obtained quantitative indicators are shown as follows, the dynamic range is improved by 3.30 times, the root-mean-square contrast is increased by 2.06 times, the signal-to-ratio (SNR) is increased by 13.55 times and the image entropy is increased by 3.45 times, respectively.