An Effective Extraction Algorithm for Ridge Information Minutia Information and DWT from Fingerprint Image

A fingerprint image is a detailed representation of the unique spatial arrangement of ridges and valleys on the fingertip skin.This intricate pattern serves as a distinctive biometric signature, utilized in various applications such as forensic science, security systems,
and access control. The complexity and individuality of each fingerprint make it a reliable and secure method for personal identification and verification. The work has been motivated by studies in anthropometry, biometric characteristic [3], and pattern recognition suggesting that it is possible to extract more detailed information from ridge, minutia and DWT information from fingerprints. The detailed studies of feature extractions in classification like gender, age, blood etc. using fingerprint only, is essential for its easiness, economical and less complex model to design as compared to other techniques as the fingerprint size results in small storage space. An automated fingerprint classification system compares the features of a test fingerprint with stored data on ridges and valleys in a database. It involves a detailed analysis of spatial patterns, minutiae points, and unique attributes for precise identification. Utilizing advanced algorithms, the system matches the test fingerprint with stored data, facilitating effective recognition in applications like law enforcement, security, and biometric authentication. The result of a fingerprint image is titled as “matching” if both the produced features of the testing image are
matched with features of the fingerprints in database, regardless of the time and method by which each image is collected. In most of the existing fingerprint based gender identification systems, the features used are the fingerprint minutiae, mainly ridge bifurcation, ridge count,
ridge ending, ridge thickness, valley thickness, ridge thickness to valley thickness ratio (RTVTR), Discrete wavelength transform . etc. However, the fingerprint based features on ridge or minutiae based are developed so far works necessarily, still there are several other
characteristics that can also be extracted on ridge and minutiae and utilize it in the classification process. The paper highly emphases on implementation of different algorithms on new features based on first discrete wavelet transform, second ridge length - i.e minimum
maximum and average ridge length, and third minutiae information-Ridge bifurcation count(RBC), Ridge end count (REC), Minutia count (µC), those can be extracted from fingerprint images