Automated Person Identification using Thumb Impression
The need for automatic person identification is increasing more and more in every part of business, industry, security and defense. Recognition of fingerprint is one of the most popular and powerful biometric techniques used for the identification of individuals. Law and enforcement agencies use this technique for criminal identification. Now a days, finger prints are being used for several applications such as access control for high security installations, credit card usage verification and employee identification.
The Fingerprint Scanner captures the images of the fingerprints. The images are processed to enhance the inherent features useful for subsequent analysis. Descriptors of the images are chosen such that they are invariant to scaling, translation and rotation of the images. The components are recognized by comparing the descriptors calculated from the acquired image with those calculated from the original image. If both matches the Electro magnetic lock will be driven by the micro controller circuit and the door will be opened for 5 seconds and the details of the person who is logged is stored on the Database else no action will be taken and the time and date is inserted in the database as error log for tracing.
THE TASK OF FINGERPRINT IDENTIFICATION The aim of any image indexing system is: Given a test image which may be noisy, distorted or deformed, the image which best matches with the test image must be retrieved by searching the reference image database which may be a few hundred to several million records in size. Basically fingerprint identification is content based retrieval system. The problem of fingerprint identification is to search efficiently and retrieve accurately, from the reference fingerprint image database, a match corresponding to a given query fingerprint image. If no match is found then the query will be rejected. The rejected fingerprint image is identified manually. With rapid growth in population a typical fingerprint image database grows to several million records in size. Efficient searching, through each and every image, and accurate retrieval are thus not often guaranteed in all cases.
MOTIVATION The main objective of this project is to implement some high Level of security in a concern. We are designing this project for our college to implement automatic students and staffs Identification and to maintain students attendance and personal details So we analyzed the pros and cons of this problem and developed a electronic door locking system where the user who needs to enter must have enrolled his finger print into the database. Pros: • • • • Convenience Greatest variety of biometric devices High Accuracy Rate It is difficult to “false or fake” a fingerprint
Why we chose this….! The basic principle of the finger prints is as follows 1. The finger print will remain largely unchanged during an individual’s lifetime. 2. Friction ridges develop in the fetus itself. 3. Friction ridge patterns and their details are always UNIQUE.
4. No two fingers have yet been identified to posses the identical ridge characteristics (Even in identical twins).
(People preferred finger print scans rather than others)
Ease of Use Accuracy User Acceptance Required Security Level Long-term stability
ALGORITHM We have used the following algorithms, 1. Minutiae based 2. Seven invariant Moments.
What is a minutia?
In the Biometric process of fingerprint scan, minutiae are specific points in a finger image. There are two types, known as ridge endings and bifurcations. Sometimes, other details, such as the points at which scars begin or terminate are considered minutiae. The number and locations of the minutiae vary from finger to finger in any particular person, and from person to person for any particular finger (for example, the thumb on the left hand). When a set of finger images is obtained from an individual, the number of minutiae is recorded for each finger. The precise locations of the minutiae are also recorded, in the form of numerical coordinates, for each finger. The result is a function that can be entered and stored in a computer database. A computer can rapidly compare this function with that of anyone else in the world whose finger image has been scanned. In theory, if a complete set of finger images was obtained for every person in the world, and the minutiae analyzed and recorded with sufficient accuracy, it would be possible for a single computer to determine the identity of any individual within seconds. Ridge is a curved line in a finger image. Some ridges are continuous curves, and others terminate at specific points called ridge endings. Sometimes two ridges come together at a point called bifurcation. Bifurcations have the appearance of branch points between curved lines.
x p y p f (x, y) dx dy
p, q 0,1,2,....,
Moments are a descriptive technique with an intuitive basis in the study of the mechanics of bodies. Properties of invariance to R, S and T transformations may be derived using functions of moments. Continuous moment transforms can be considered first and then they can be formulated and ramnificated for discrete implementations. The moment transform of an image function, f(x, y), is given by
Where f(x,y) is a 2-D polynomial basis functions Mpq = ip jq (i,j)
(i i) p ( j j)q f (i, j )
i = (m10 / m00) j = (m01 / m00). The central moments from this equation are still sensitive to R and S transformations. The scale invariance may be obtained by further normalizing µpq. The normalized central moments, denoted by npq, are defined as npq = pq / 00 Where = (p+q)/2+1, p+q = 2,3, Table for Interpretation of moments Central moment Interpretation 20 02 11 12 21 30 03 Horizontal centralness Vertical centralness Diagonality Horizontal divergence Vertical divergence Horizontal imbalance Vertical imbalance
Orientation 0o 90o 180o 270o
φ7 0.115362 0.105735 0.116675
0.010045 0.042431 0.018192 0.040466 0.016662 0.040077
0.061780 0.057371 0.061121 0.050646 0.062762 0.058207
0.117958 0.078586 0.114922 0.070813 0.119729 0.078245
0.013962 0.042463 0.062801 0.057789 0.118483 0.078996 0.114990
The False Acceptance rate for Finger scan using Minutiae algorithm is 1/1000 We have combined the above said Minutiae algorithm and the seven invariant moments algorithm to reduce the False Acceptance Rate and to enhance effective matching. How it works…! 1 2 Capture - Finger print is captured by the fingerprint scanner Extraction - minutiae points and invariant moments are extracted from the Live data and a template is created. 3 4 Comparison - the live data is then compared with the templates stored in Database. Match/Non-Match - the system then decides if the features extracted from the template are a match or a non-match. 5 If there is a match then this output is given to a microcontroller to enable a Electronic lock which is driven by a micro controller circuit. 6 Also the visual basic software will track if there is any error and tracks who and When the person is logging.
IMPLEMENTATION We have decided to implement the “AUTOMATED PERSON IDENTIFICATION USING THUMB IMPRESSION” with the above said methodology in RMK ENGINEERING COLLEGE KAVARAIPETTAI GUMUDIPOONDI THIRUVALLUR DISTRICT. Already We have installed the “AUTOMATED PERSON IDENTIFICATION USING THUMB IMPRESSION” in Government Polytechnic college Purasaiwakkam Chennai.
Finger Print samples Collected from the students and staff of our college.
The default database is available at N I S T, U S A. (National Institute for Soft-computing Technology).
1 2 3 4
The future and destiny of computerized network security and identification lies in biometrics. Biometrics needs to meet the customer and user’s requirements for security. The use of biometrics in networks as an authentication feature is gaining momentum However the widespread use and acceptance of biometrics is, at this current time, still in its infancy.
Your fingerprint may become the key to your house
Bibliography Robotics www.biometricsworld.com www.biometrics.com CSG Lee Gonzalez