Self-adaptive fingerprint image adjusting method

Abstract

The invention provides a self-adaptive fingerprint image adjusting method comprising the following steps: in step S1, before users put fingers in place, histogram information of a current blank background image is saved, and a quantity threshold value and characteristic point information of preliminary background determination can be obtained. In step S2, after the users put the fingers in place,histogram statistics are run on fingerprint images, and the fingerprint images are subjected to partitioning operation. In step S3, the characteristic point information is substituted in, and the whole fingerprint images are subjected to smoothing operation. In step S4, smoothing operation results are saved, and finger types are determined after the fingers are put in place. In step S5, after thefinger type is determined, the fingerprint images are classified according to the finger types, and statistical operation can be completed. The self-adaptive fingerprint image adjusting method does not require ex-factory calibration, and real time matching can be realized; problems of foreign matters on a sensor, unclean fingers and the like can be effectively solved; parameters change along withchanges of the fingers of the users in four seasons as well as dryness and wetness changes of the fingers, and the fingerprint images are uniform in background and high in clearness and image effectsafter the method is adopted.

Description

technical field [0001] The invention relates to the field of fingerprint image processing, in particular to an adaptive fingerprint image adjustment method. Background technique [0002] With the promotion of fingerprint identification applications, high-performance fingerprint identification systems need to adapt to various application environments accurately and quickly and extract fingerprint feature information. However, due to various environmental factors, most existing collectors usually introduce noise points during the fingerprint recording process, resulting in unclean fingerprint backgrounds. Noise points will seriously affect post-image processing, detail point extraction, etc., thereby affecting the application effect of fingerprints. In order to improve the accuracy of fingerprint feature extraction, generally at the factory, the inconsistency of the sensor surface image blocks is detected by AGC, and the appropriate parameters are calculated one by one through...

AI Technical Summary

Problems solved by technology

The existing technology needs to do a separate background correction, which is slow and affects the speed of mass production, and the factory conditions are too ideal to adapt to the particularity of the user's use environment

After leaving the factory, the parameters remain unchanged, so the existing technology cannot well match the situation of large changes in finger dryness and wetness due to changes in the four seasons

Images