Fingerprint template based on fuzzy feature point information and fingerprint identification method

1. A fingerprint identification method based on fuzzy feature point information, comprising the following steps: S 1 , obtaining a fingerprint template based on the fuzzy feature point information and a scene fingerprint; S 4 , comparing the scene fingerprint with the fingerprint template based on the fuzzy feature point information;

the fingerprint template based on the fuzzy feature point information comprises: multiple pieces of fuzzy feature point information established based on a part of feature point information of a template fingerprint, and the multiple pieces of fuzzy feature point information are insufficient to restore the template fingerprint; and

the fuzzy feature point information comprises registration information and comparison information, registration information comprises a horizontal coordinate, a vertical coordinate and an attribute of the fuzzy feature point information, the comparison information includes the value of a polynomial P(X i )=C 0 +C 1 X i +C 2 X i 2 + . . . +C n X i n , n is an order of the polynomial, X i is a numerical value obtained by quantifying one or multiple pieces of feature point information except the registration information in the feature point information, and C 0 -C n are coefficients having a preset rule; and

the step S 4 comprises: substituting the corresponding feature point information of the scene fingerprint into the polynomial, calculating the values of C 0 -C n , judging whether the calculated C 0 -C n meet the preset rule, if the judgment result is yes, the scene fingerprint is matched with the fingerprint template based on the fuzzy feature point information, and if the judgment result is no, the scene fingerprint is not matched with the fingerprint template based on the fuzzy feature point information.

2. The fingerprint identification method based on the fuzzy feature point information of claim 1 , further comprising the following step between the step S 1 and the step S 4 :

S 2 , carrying out fingerprint registration on the scene fingerprint and the fingerprint template based on the fuzzy feature point information; and

the step S 2 comprises the following steps:

a, selecting three horizontal coordinates and vertical coordinates of the fuzzy feature point information of the fingerprint template based on the fuzzy feature point information to serve as vertexes, in order to establish a first non-isosceles triangle, and recording the geometric data of the first non-isosceles triangle;

b, selecting three horizontal coordinates and vertical coordinates of the feature point information of the scene fingerprint to serve as vertexes, in order to establish a second non-isosceles triangle, and recording the geometric data of the second non-isosceles triangle;

c, matching the first non-isosceles triangle with the second non-isosceles triangle;

d, repeating the step a to the step c, if matched non-isosceles triangles larger than a first threshold pair can be obtained, executing e; if the matched non-isosceles triangles larger than the first threshold pair cannot be obtained, the scene fingerprint is not matched with the fingerprint template based on the fuzzy feature point information; and

e, calculating a rotation angle and a translation distance of the scene fingerprint relative to the fingerprint template based on the fuzzy feature point information, and rotating and/or translating the scene fingerprint according to a calculation result to register the scene fingerprint with the fingerprint template based on the fuzzy feature point information.

3. The fingerprint identification method based on the fuzzy feature point information of claim 2 , wherein the geometric data comprise: a largest side length, a second largest side length, a smallest side length, a largest angle, a second largest angle, a smallest angle, a largest angle vertex horizontal coordinate, a largest angle vertex vertical coordinate, a vector from the largest angle vertex to the second largest angle, a vector from the largest angle vertex to the smallest angle vertex, an attribute of the largest angle vertex, an attribute of the second largest angle vertex and an attribute of the smallest angle vertex.

4. The fingerprint identification method based on the fuzzy feature point information of claim 2 , wherein the step e of calculating the rotation angle and the translation distance of the scene fingerprint relative to the fingerprint template based on the fuzzy feature point information, and rotating and/or translating the scene fingerprint according to the calculation result, comprises: with the largest angle vertex of the scene fingerprint as a fiducial point, respectively calculating the rotation angle and the translation distance of the non-isosceles triangle matched with the first threshold pair, and using the calculated numerical values with the highest frequency as a rotation coefficient and a translation coefficient of the scene fingerprint for rotating and translating the scene fingerprint.

5. The fingerprint identification method based on the fuzzy feature point information of claim 2 , wherein matching success in the step c comprises the following conditions:

the attributes of three pairs of vertexes corresponding to the scene fingerprint and the fingerprint template based on the fuzzy feature point information are consistent;

the length differences of three pairs of sides corresponding to the scene fingerprint and the fingerprint template based on the fuzzy feature point information are within a second threshold; and

the angle differences of three pairs of angles corresponding to the scene fingerprint and the fingerprint template based on the fuzzy feature point information are within a third threshold.

6. The fingerprint identification method based on the fuzzy feature point information of claim 5 , further comprising the following step between the step S 2 and the step S 4 :

S 3 , calculating the number of the horizontal coordinate, the vertical coordinate and the attribute of the rotated and translated scene fingerprint matched with the feature points of the registration information of the fingerprint template based on the fuzzy feature point information, if the calculation result is not larger than a fourth threshold, the scene fingerprint is not matched with the fingerprint template based on the fuzzy feature point information, and if the calculation result is larger than a fourth threshold, executing the step S 4 .

7. The fingerprint identification method based on the fuzzy feature point information of claim 1 , wherein C 0 -C n comprise a first portion and a second portion, the value of the first portion is randomly generated, and the value of the second portion is a value obtained by calculating the value of the first portion according to the preset rule.

8. The fingerprint identification method based on the fuzzy feature point information of claim 7 , wherein the step S 4 of substituting the feature point information of the scene fingerprint into the polynomial, calculating the values of C 0 -C n and judging whether the calculated C 0 -C n meet the preset rule, comprises:

substituting the feature point information of the scene fingerprint into the polynomial, calculating the values of C 0 -C n , obtaining the value of the first portion in C 0 -C n , calculating the value of the first portion according to the preset rule to obtain the value of the second portion, and judging whether the calculated value of the second portion is the same as the value of the second portion in the template fingerprint.