Fingerprint feature evaluation method based on combination of fuzzy mathematics and probability theory

Abstract

The invention relates to a fingerprint feature evaluation method based on the combination of fuzzy mathematics and the probability theory, which includes the following steps: after a field fingerprintand a ten-finger sample fingerprint are given, determining whether the pattern features of the field fingerprint and the ten-finger sample fingerprint are consistent through comparison, and determining the occurrence probability of the fingerprint pattern features; determining whether the field fingerprint and the ten-finger sample fingerprint are consistent in feature point type, fingerprint line direction and number of interval lines between feature points through analysis, marking the detail features possessed by both the field fingerprint and the ten-finger sample fingerprint on a fingerprint image, and calculating the combination probability that the fingerprint detail features occur in a fingerprint line area at the same time; analyzing and judging the stability of the marked fingerprint detail features, and obtaining the value of each marked fingerprint detail feature in a fingerprint feature stability judgment domain; listing the fuzzy membership degree corresponding to each marked fingerprint detail feature; and calculating the feature matching probability of the field fingerprint and the ten-finger sample fingerprint.

Description

technical field [0001] The invention belongs to the technical fields of public security criminal technology, judicial identification and public security, and in particular relates to a fingerprint feature evaluation method based on the combination of fuzzy mathematics and probability theory. Background technique [0002] Fingerprints are considered to be the first in physical evidence, and fingerprint feature matching methods, as the basis for identifying criminal suspects, play an important role in the fields of criminal technology, judicial identification, and public security in public security organs. Since the Argentine police first identified the murderer with on-site fingerprint features in a homicide case in 1892, fingerprint feature matching has a history of more than 100 years. Has not stopped. At present, countries around the world do not have a uniform standard for the minimum number of minutiae feature points required for the same identification of fingerprint f...

AI Technical Summary

Problems solved by technology

However, the current fingerprint feature matching method does not organically combine the ambiguity of fingerprint features judged by the fingerprint feature comparison staff with the probabilistic expression of fingerprint feature matching. The practice of various fingerprint feature matching shows that the current fingerprint feature matching method is in a certain There is a lack of scientific basis to a certain extent, and the fingerprint feature matching conclusions also lack scientificity to a certain extent.

Images