Automatic recognition and evaluation of micro-contaminant particles on ultra-smooth optical substrates using image analysis method

Micro-contaminant particles on surface of optical substrate have unfavorable influence on light characteristics and instability of optical devices. Proper recognition and evaluation of micro-contaminant particles on post-cleaning substrate are important in fabricating high-performance optical substrates. Based on image analysis and difference of gray scale threshold in image zones, this paper presents a technique for automatically recognizing micro-contamination, and the relevant software developed for on-line evaluation of the level of cleanliness on ultra-smooth optical substrate. Using the self-developed software incorporating high-resolution microscope, optical yttrium iron garnet (YIG) and K8 substrate surfaces of post-chemical mechanical polishing (CMP) were investigated. Results from analyzing the YIG and K8 optical substrates before-and-after laser cleaning illustrated the success of the developed automatic recognition and on-line evaluation system in identifying the micro-contaminant particles on the ultra-smooth substrate surface.