Non-costly, non-invasive, safe, and reliable electronic vision enhancement systems (EVES) and their methods have presented a huge medical and industrial demand in the early 21st century. Two vision compensation and enhancement algorithms are first presented, qualitatively optimizing the view of a truncated image, the “convex”, and the “cartoon superimposition” techniques. The author compares these to a novel technique, motivated by the characterization of quality vision parameters in an attempt to account for and compensate reported viewing difficulties and low image quality measures associated with these two existing methods.
This “partial cartoon” technique is based on introducing the invisible image to the immediate left and right of the truncated image as a superimposed cartoon into respective sides of the truncated image, yet only on a partial basis as not to distract the central view of the image. Warped images are quantitatively compared by evaluating the Root-Mean-Square Error (RMSE) and the Universal Image Quality Index (UIQI), both representing image distortion and quality measures of warped, as compared to original images for five different scenes. It is concluded that the presented partial cartoon method exhibits superior image quality for all objective measures.
|Partial Cartoon Method ICMCS 2014 (AMS)||124.58 كيلوبايت|