Fingerprint Image Classification using Singular Points and Orientation Information

Abstract

The fingerprint classification process is an essential task that reduces fingerprint matching time of an Automatic fingerprint identification system, where a large database is used. It is still a challenging task. The proposed work classifies the fingerprints by using singular points and orientation information below the core point. As the first step of proposed work, features like orientation field and singular points are extracted. Orientation field is estimated using multi-scale principal component analysis and singular points are detected using shape analysis of binary candidate region image. To speed up the classification process, rules based on location of singular points and orientation information below upper core point are used for classification. The proposed work is tested on most popular public NIST special database 4 and experimental results show that classification accuracy is 92.2 % for five-class problem and 97.35% for four-class problem without rejection. Also, the proposed work classifies more accurately the ambiguous fingerprint images into its primary as well as secondary class.

Authors and Affiliations

K. S. Jeyalakshmi, T. Kathirvalava kumar

Keywords

Related Articles

Synthesis and Characterization of Cadmium Substituted Copper Nano – Ferrites

A series of copper substituted cadmium nano ferrites with the compositional formula Cu 1-x CdxO4(where x=0.1, 0.3. 0.5 and 0.7) were prepared by the Citrate gel auto combustion technique and sintered at 873 K for 4 hr. T...

SR-Code: Smart Relay Network Coding for Data Collection for Wireless Sensor Networks

Reliability in data collection for wireless sensor networks is one of the major problems in IoT applications. Sensor nodes are usually placed in harsh conditions where data communication is at risk of losing packets. Ret...

Studies on machining characteristics of titanium alloy Ti-6Al-4V using abrasive water jet cutting

Titanium alloy Ti-6Al-4V has achieved the significance in various applications because of its high modulus to density ratio, biocompatibility, high corrosion resistance, high temperature resistance etc. Therefore machini...

Direct current magnetron sputtered ultrathin Ag/Ta/glass films

In this study, ultrathin Ta, Ag and Ag/Ta thin films deposited onto glass substrates using direct current magnetron sputtering. The physical properties of as-deposited films were investigated by scanning electron microsc...

“ Study of Sesmic Analysis of Masonry Wall Structure”

Earthquakes are natural trouble under which disasters are mainly caused by damage or collapse of the structure and other man-made structures. When an earthquake occurs natural period of vibration is more on heavy loaded...

Download PDF file
  • EP ID EP392134
  • DOI 10.9790/9622-0709023342.
  • Views 140
  • Downloads 0

How To Cite

K. S. Jeyalakshmi, T. Kathirvalava kumar (2017). Fingerprint Image Classification using Singular Points and Orientation Information. International Journal of engineering Research and Applications, 7(9), 33-42. https://europub.co.uk/articles/-A-392134