| CPC G06V 40/10 (2022.01) [G06V 10/426 (2022.01); G06V 10/457 (2022.01); G06V 10/48 (2022.01)] | 6 Claims |
|
1. A computer implemented method of tracking and tracing reptile skins having a skin surface with scales for surface identification, comprising the steps of
acquiring at least one image of a surface portion to be identified,
an edge feature detection step detecting edge features of borders of scales in said image by scanning the acquired image along scan lines over an area supposedly comprising a plurality of scales acquiring intensity or color profiles,
an edge identification step, which analyses the acquired scan line image profiles for determining scale edge candidates of one or more scales,
a scale construction voting step based on said scale edge candidates determining adequate scale edges being part of a specific scale,
a scale vote accumulation step determining a data set representative for the identified scale comprising one or more data taken from the group comprising data relating to an ellipse, the major axis and minor axis of an ellipse, and the center position of an ellipse,
building for each of the identified scales a graph of the repetitive pattern scales positions of the detected scales using the data sets for each of the identified scales,
introducing a recomputing scales step identifying further scales where the built graph of scales presents gaps, with the steps of:
representing a grid of scales as graph with nodes corresponding to scale centers and graph links connecting centers of scales that share a common border,
calculating a feature value for each graph link comprising the link midpoint as 2D-coordinates and a link property as z-coordinate,
wherein the link property comprises interscale distance and related sum of scale radii,
wherein missing scales are identified through determining that feature values of the area are not in accordance with neighbor values,
determining the outline of each of the detected scales and creating a representative data set for each detected scale comprising said outline data,
determining recognition features data from the plurality of representative data sets of detected scales of the surface comprising scales,
storing said recognition features data for said surface comprising scales in a database,
comparing said acquired recognition features data of said animal skin specimen with previously acquired and stored sets of recognition features data of surfaces comprising scales from reptile skins for an identification of the surface portion of the animal skin specimen within said stored recognition features data, and
in the case of a match of the acquired recognition features data of the animal skin specimen with a stored set of recognition features, updating said database with the comparison result and the updated acquired recognition features.
|