David Gerwe (Boeing), Brandoch Calef (Boeing Laser Technical Services), Carlos Luna (Boeing Phantomworks)
Keywords: image enhancement, image quality
Abstract:
Image enhancement algorithms typically require tuning one or more input parameters to get the best results. Skipping this step or poor choice of values can often result in significant decrease in enhancement level or even degrade the image. This paper demonstrates the utility of image quality metrics in automating this tuning process for Space Situational Awareness imagery of resolved Resident Space Objects. The metrics considered in this study compare an original pristine image to the final displayed image, thus only apply directly to simulated images. However it is shown that a training set can be used to determine the best settings as a function of measureable imaging condition (light level, r0, ) to produce a look-up table that can be used for field collected data.
Date of Conference: September 9-12, 2014
Track: Daylight Imaging