Smoothed orientation

Smoothed orientation, a part of algorithm transparency, is computed from block orientation during feature extraction by adding orientation vectors of four edge-neighboring blocks to every primary block on the block map that is within mask. Smoothed orientation is used in parallel smoothing and orthogonal smoothing.

KeywordSuffixZip
smoothed-orientation.json026-smoothed-orientation.json
smoothed-orientation.dat027-smoothed-orientation.dat

Visualization

Visualization of smoothed block orientations was constructed from this stage's binary data combined with block map, filtered mask, and the original fingerprint image. Visualization itself is not part of transparency data.

Block orientations on a fingerprint visualized as red sticks oriented parallel to ridges
Smoothed block-level ridge orientation.

Binary data

Structure of binary data is identical to that of block orientation.

Example: 027-smoothed-orientation.dat (1,300 scalars, range -3,868 to 3,164, average -439)

Json

Json data describes basic properties of the binary data, most importantly dimensions of the orientation map, which is the same as the number of primary blocks.

Example: 026-smoothed-orientation.json

{
  "axes": [
    "y",
    "x",
    "axis"
  ],
  "dimensions": [
    25,
    26,
    2
  ],
  "scalar": "double",
  "bitness": 64,
  "endianness": "big",
  "format": "IEEE754"
}