Equalized image in SourceAFIS transparency data

SourceAFIS » Algorithm » Transparency » Equalized image

Equalized image, a part of algorithm transparency, is computed from scaled image during feature extraction. Every primary block is equalized separately using data from smoothed histogram to produce high-contrast image. Equalized image is used to compute pixelwise orientation. It is later processed using parallel smoothing and orthogonal smoothing.

KeyMIMEFilename in ZIP
equalized-imageapplication/cbor012-equalized-image.cbor

Visualization

The following fingerprint image can be reconstructed from this stage's binary data. Visualization itself is not part of transparency data.

Equalized fingerprint image. Areas outside mask are erased.

Format

Equalized image has the same format as decoded image except that pixels are in range from -1 (white) to 1 (black). Areas outside the mask are set to -1.

Example: 012-equalized-image.cbor

{
  "width": 388,
  "height": 374,
  "cells": [
    -1.0,
    -1.0,
    "... skipped 145,108 items in range -1.00 to 0.998 ...",
    -1.0,
    -1.0
  ]
}