2 ms·

> you could definitely learn a filter that would be very precise in reconstructing the image when zoomed Yes, your result would be a very clear image of one po

by TillE 11y ago

> you could definitely learn a filter that would be very precise in reconstructing the image when zoomed Yes, your result would be a very clear image of one possible license plate. An algorithm may be able to do slightly better than a squinting human, but ultimately you can't retrieve destroyed information.

skimpycompiler 11y ago
Of course you can't retrieve destroyed information. But information is not destroyed by perfectly unpredictable (uniform) distributions of noise. > Yes, your result would be a very clear image of one possible license plate. Fortunately there are methods of evaluating how well your statistical filter works, if it's precise enough you'd be fine with indeterministic nature of your filter. Or even better, you could generate all of the highly probable licence plates, instead of having only one - given by your deterministic algorithm.