Jak rozpozna google stejny obrazek ?

igh frequencies give you detail, while low frequencies show you structure.

The fastest way to remove high frequencies and detail is to shrink the image. In this case, shrink it to 8x8 so that there are 64 total pixels. Don't bother keeping the aspect ratio, just crush it down to fit an 8x8 square.

Reduce color. Average the colors. Compute the bits. Construct the hash.

The resulting hash won't change if the image is scaled or the aspect ratio changes. Increasing or decreasing the brightness or contrast, or even altering the colors won't dramatically change the hash value.

Napsal Jules Winnfield;804920

s timto si lamu hlavu uz chvili. vi tady nekdo ?

je tam naky elektronicky podpis ? jakto ze kdyz obrazek upravim jako olejomalbu a pouziju hledani podle obrazku najde to zase puvodni verzi ??????

Napsal FSP;804944

Zrejme podľa EXIF vlastností/informácií/dát?

Napsal Jules Winnfield;804920

s timto si lamu hlavu uz chvili. vi tady nekdo ?

je tam naky elektronicky podpis ? jakto ze kdyz obrazek upravim jako olejomalbu a pouziju hledani podle obrazku najde to zase puvodni verzi ??????