Fingerprint recognition is one of the common and most widely used biometric technologies. The biometric technique provides a high safety and accuracy performance of personal verification and identification system. When a person is first using the fingerprint biometric reader, his or her biometric characteristics will extract and encode as a template by enrollment process. The template either store in reader’s memory database or external memory device.
The fingerprint recognition can be either in verification mode or identification mode, with comparing the user biometric information and template in dataset, the information of person can determine if they are matched. (Bana & Kaur, 2011)Now a day, different type of sensor had been developed to obtain the information of biometric characteristic: capacitor sensor, optical sensor, thermal sensor, ultrasonic sensor. The sensors scan the image of ridges and furrows on the fingerprint, then image enhancement, fingerprint minutiae determination and convert into a template database. Fingerprint recognition bring benefit of fast and easy for apply, even though there are different type of biometric technique have developed, smart phone or bank are select fingerprint recognition as their identification method because of its convenience and lost cost. Figure 3: Fingerprint recognition conceptChapter 2.2: Concept and TheoryThere are three techniques use for fingerprint recognition: ridge feature based technique, correlation technique and minutiae based technique. Base on research, minutiae base technique is most popular technique that people to use.
(Bana & Kaur, 2011)Base on this reason, this project is focus on minutiae base technique to design fingerprint recognition algorithm. In state of minutiae base technique, fingerprint recognition can describe in three level field: sensor, minutiae extractor and minutiae matcher. Figure 4: Fingerprint recognition level field For sensor, optical sensor and capacitance sensor are widely used. These sensors have high accuracy and efficiency performance unless user’s finger is too dirty or dry. (Sondhi & Bansal, 2014) User required put his or her finger on the reader, then sensor will detect ridges and furrows on fingerprint and then enroll and encode to template database.In minutiae extractor, it will separate into three stages: preprocessing, minutiae extraction and post processing. In preprocessing stage, the quality of fingerprint image is enhanced and noise has filter.
The image after enhancement was binarization by threshold method for simplify image processing. Then, the removing unwanted area of binarization on image was done by image segmentation. The main stage for minutiae extractor is minutiae extraction for determine the fingerprint minutiae. Thinning process is to simplify minutiae characteristics of fingerprint.
For minutiae marking, each minutia that used for matching is mark. Ridge ending and ridge bifurcation are common use for minutiae marking. Third stage is post-processing, false marked minutiae will be removed. (Kaur & Ameeta, 2014) Figure 5: Minutiae extractorThe minutiae matcher field main an important role for determine the fingerprint matching result. The position and orientation of minutiae are calculate. The minutia matcher chooses any two minutiae set as a reference minutia pair and then matches their associated ridges. If the ridges match well, the two fingerprint images are aligned and matching is conducted for all the remaining minutiae.
(Kaur & Ameeta, 2014)