Open AccessA hyprid technique for human footprint recognition
Author(s) -
Yahya Ismail Ibrahim,
Israa Mohammed Alhamdani
Publication year - 2019
Publication title -
international journal of power electronics and drive systems/international journal of electrical and computer engineering
Language(s) - English
Resource type - Journals
eISSN - 2722-2578
pISSN - 2722-256X
DOI - 10.11591/ijece.v9i5.pp4060-4068
Subject(s) - footprint , computer science , biometrics , block (permutation group theory) , artificial intelligence , feature (linguistics) , authentication (law) , memory footprint , feature selection , computer vision , pattern recognition (psychology) , data mining , computer security , geography , mathematics , linguistics , philosophy , geometry , archaeology , operating system
Biometrics has concerned a great care recently due to its important in the life that starts from civil applications to security and recently terrorism. A Footprint recognition is one of the personal identifications based on biometric measurements. The aim of this research is to design a proper and reliable biometric system for human footprint recognition named (FRBS) that stands for Footprint Recognition Biometric System. In addition, to construct a human footprint database which it is very helpful for various use in scientific application e.g. for authentication. There exist many biometrics databases for other identity but very rare for footprint. As well as the existing one are very limited. This paper presents a robust hyprid techniques which merges between Image Processing with Artificial Intelligent technique via Ant Colony Optimization (ACO) to recognize human footprint. (ACO) plays the essential role that rise the performance and the quality of the results in the biometric system via feature selection. The set of the selected features was treated as exploratory information, and selects the optimum feature set in standings of feature set size. Life RGB footprint images from nine persons with ten images per person constructed from life visual dataset. At first, the visual dataset was pre-processed operations. Each resultant image detects footprint that is cropped to portions represented by three blocks. The first block is for fingers, the second block refers to the center of the foot and the last one determines the heel. Then features were extracted from each image and stored in Excel file to be entered to Ant Colony Optimization Algorithm. The experimental outcomes of the system show that the proposed algorithm evaluates optimal results with smaller feature set comparing with other algorithms. Experimental outcomes show that our algorithm obtains an efficient and accurate result about 100% accuracy in comparison with other researches on the same field.