Biometrics week 1.pptx

FACULTY OF AI & DS DEPARTMENT COMPUTER SCIENCE & Semester 5 /Level 3 ARTIFICIAL INTELLIGENCE AI 301: BIOMETRICS 2024 - 2025  INSTRUCTOR: LECTURER: DR. LAMIAA ALI SAID...

FACULTY OF AI & DS DEPARTMENT COMPUTER SCIENCE & Semester 5 /Level 3 ARTIFICIAL INTELLIGENCE AI 301: BIOMETRICS 2024 - 2025  INSTRUCTOR: LECTURER: DR. LAMIAA ALI SAID EMAIL [email protected] HALL G312 - F322 OFFICE HOURS MONDAY 10:30 – 11:30 WEDNESDAY 8:30 – 10:30 2  TIMETABLE LECTURE: GROUP(C6,C7&C8) MONDAY 8:30  10:20 GROUP(C2&C2) THURSDAY 10:30  12:20 2 CLASS NOTES & SHEETS: AVAILABLE ON THE UNIVERSITY POWER CAMPUS AFTER EACH LECTURE AND LAB. Percenta GRADING: ge Title Week 20% Mid-Term Exam Week 8 10% Interactive Learning Weekly Lab Work (Weekly) – 10% Assignments (min. 1 every 2 Weeks) – Weeks [2 – 14] Quizzes (min. TWO) 20% Practical exam and/or Project Week 15 Discussion 40% Final Exam Weeks [16 – 17] 4 100% Total  COURSE DESCRIPTION BIOMETRICS APPLICATIONS: FINGERPRINT RECOGNITION, FACE RECOGNITION, IRIS RECOGNITION, BIOMETRIC TRAITS, MULTIBIOMETRICS SYSTEMS. PERFORMING PROJECTS ON DIFFERENT BIOMETRICS SYSTEMS. 5  COURSE OBJECTIVES TO INTRODUCE STUDENTS TO BIOMETRICS TO INTRODUCE STUDENTS APPLICATIONS FOR BIOMETRICS SUCH AS FINGERPRINT RECOGNITION, FACE RECOGNITION ,IRIS RECOGNITION, BIOMETRIC TRAITS. TO DEVELOP STUDENTS’ ABILITY TO PERFORM MULTIBIOMETRICS SYSTEMS. TO DEVELOP STUDENTS’ ABILITY TO PRACTICE BIOMETRICS APPLICATIONS BY DOING PROJECTS. 6  BIOMETRICS Humans typically use body characteristics such as face, voice, and gait along with other contextual information (e.G., Location and clothing) to recognize one another. Identity management plays a critical role in a number of application. A biometric system measures one or more physical or behavioral including fingerprint, palmprint, face, iris, retina, ear, voice, signature, gait, hand vein, or 7 the DNA informations.  PERSON RECOGNITION Basic approaches to person recognition. - Traditional schemes use passwords (“what you remember”) (“what he knows”) - id cards or Lockers to validate individuals and ensure that system resources are accessed only by a legitimately enrolled individual. 8  PERSON RECOGNITION With the advent of biometrics, it is now possible to establish an identity based on “who you are intrinsically”. 9  PERSON RECOGNITION Since biometric systems require the user to be present at the time of authentication, they can also deter users from making false repudiation. Biometric recognition is being increasingly adopted in a number of government and civilian identity management applications either to replace or to complement existing knowledge-based and token based mechanisms 10 HOW DOES A BIOMETRIC SYSTEM IDENTIFY A USER BASED ON HIS PHYSICAL AND/OR BEHAVIORAL TRAITS? 11  HOW DOES A BIOMETRIC SYSTEM IDENTIFY A USER BASED ON HIS PHYSICAL AND/OR BEHAVIORAL TRAITS? This process consists of two main phases, namely, enrollment and recognition During the enrollment phase, the biometric data is acquired from the individual and stored in a database 12 along with the person’s identity.  Enrollment and recognition phases typically, the acquired biometric data is processed to extract salient and distinctive features. In many cases, only the extracted feature set gets stored, while the raw biometric data is discarded. During the recognition phase, the biometric data is reacquired from the individual and compared against the stored data to determine the user identity 13  BIOMETRIC SYSTEM A biometric system is essentially a pattern recognition (or a pattern matching) system consisting of four basic building blocks, namely, (a) sensor, (b) feature extractor, (c) database, and (d) matcher. 14  SENSOR MODULE A suitable user interface incorporating the biometric sensor or reader is needed to measure or record the raw biometric data of the user. The design of a good user (or human-machine) interface is critical for the successful implementation of a biometric system. The quality of the raw biometric samples also depends on the characteristics of the sensor used Fingerprints scanned at (a) 1000 points per inch (ppi) and (b) 500 points per inch using different fingerprint sensors. Furthermore, factors like cost, size, and durability 15 also impact the sensor design.  FEATURE EXTRACTION MODULE Feature extraction refers to the process of generating a compact but expressive digital representation of the underlying biometric trait, called a template the template is expected to contain only the salient discriminatory information that is essential for recognizing the person. 16  FINGERPRINT FEATURE EXTRACTION FINGERPRINT IS COMMONLY REPRESENTED AS A SET OF POINTS DEPICTING THE MINUTIAE 17  FEATURE EXTRACTION FOR IRIS Is represented as a binary vector depicting the binarized response of an input image to Gabor filters 18  FACE FEATURE EXTRACTION Face is commonly represented as a vector of real numbers depicting, say, the coefficients of linear discriminant analysis (LDA). 19  Face segmentation and enhancement. A. A face image of a person as captured by the camera and B. the processed face image obtained after segmentation (removal of the background and other non-face regions such as hair and regions below the chin) and contrast enhancement based on histogram equalization. 20  FEATURE EXTRACTION MODULE DURING ENROLLMENT The template gets stored either in the central database of the biometric system or is recorded on a token (e.g., Smart card) issued to the individual based on the nature of the application At the time of recognition, the template is retrieved from the database, and matched against the feature set extracted from the new biometric sample acquired from the user 21  FEATURE EXTRACTION MODULE DURING ENROLLMENT The template of a user can be extracted from a single biometric sample, or generated by processing multiple samples acquired during enrollment Some systems store multiple templates in order to account for the large variations that may be observed in the biometric data of a user. 22  Database module During the enrollment process, the feature set extracted from the raw biometric sample (i.e., The template) is stored in the database along with some personal identity information. Storing all the templates in a central database may be beneficial from a system security perspective, because the data can be secured through physical isolation and by having strict access control mechanisms 23  Matching module The purpose of a biometric matcher is to compare the query features against the stored templates to generate match scores. The match score is a measure of the similarity between the template and the query. Hence, a larger match score indicates greater similarity between the template and the query. 24 THANK YOU 25