Thinned skeleton (SourceAFIS transparency data)

SourceAFIS » Algorithm » Transparency » Thinned skeleton

Thinned skeleton, a part of algorithm transparency, is computed from binarized skeleton by eroding all ridges down to single pixel thick lines and then applying some filters. Two thinned skeletons are constructed during feature extraction, one for ridges and one for valleys. Thinned skeleton is still a 2D image. It is processed into higher level data structure in traced skeleton stage.



Visualization of thinned skeleton was constructed from this stage's binary data with original fingerprint in the background. Visualization itself is not part of transparency data.

Thinned fingerprint image. Besides errors inherited from binarization, thinning introduces new errors especially in the form of tiny ridges or "tails" growing out of the real ridges. These errors will be filtered later.

Binary data

Binary data has the same structure as for binarized image except that ridges/valleys are now thinned.



Json data describes basic properties of the binary data, most importantly dimensions of the image.


  "axes": [
  "dimensions": [
  "scalar": "boolean",
  "bitness": 8,
  "format": "false as 0, true as 1"