In the previous article I described the memory of wave neurons, in this article we will look at the image classifier.
What is a classifier
A classifier (Latin classis — category and facere — to do) is a systematized list of named objects, each of which is assigned a unique code.
For the purposes of this article, I will call a classifier a device that gives a pair of images a unique code.
Theory
, , , 1., , 3, . , , . DEFMINCOLOR.
( , ) , DEFMINCOLOR. , , DEFMINCOLOR , , , . , . . , 100. . DEFMINCOLOR .
, .
. , 144p.
gif , .
255- gif .
,
( DEFMINCOLOR ) 255
1.9685%
1.9685/100*255=5
( DEFMINCOLOR ) 255
1.9685%
1.9685/100*255=5
( DEFMINCOLOR ) 255
2.3622%
2.3622/100*255=6 .
, , - , - ( ), . , . , , . .
, . 255, png , 7 15 . .
,
( DEFMINCOLOR ) 700
29.4286%
29.4286/100*700=206
( DEFMINCOLOR ) 700
11.8571%
11.8571/100*700=83
( DEFMINCOLOR ) 700
12.1429%
12.1429/100*700=85 .
, . DEFMINCOLOR
,
( DEFMINCOLOR ) 2000
18.3%
18.3/100*2000=366
( DEFMINCOLOR ) 2000
14.1%
14.1/100*2000=282
( DEFMINCOLOR ) 2000
11.75%
11.75/100*2000=235 .
DEFMINCOLOR 255, , gif . .
,
( DEFMINCOLOR ) 255
45.8824%
45.8824/100*255=117
( DEFMINCOLOR ) 255
15.2941%
15.2941/100*255=39
( DEFMINCOLOR ) 255
14.76%
14.76/100*255=36 .
Image of the classifier
Conclusions: It turns out that the classifier works, but the GIF images are not suitable for classification due to the strong compression.
Thanks for reading the article.