Study on quality metrics for restored simulated range images of coherent ladar

Image quality metrics is an important aspect in image processing of ladar. Range images of coherent ladar denoised by median filter are evaluated by structural similarity (SSIM) index and root mean square error (RMSE) in this paper. The performances of the metrics are compared, and whether real range values are kept or not in restored images should be paid more attention to. The results show that the quality metrics, RMSE, and error of target mean should be synthetically considered in evaluating range images. When RMSE of range images are not significantly different, keeping real range values should be considered first.