CSCI 0454A: Biometrics (Fall 2017)
|When and Where||TuTh 1:30 – 2:45PM in MBH 505|
|Office Hours||M: 10:30AM – 12:15 PM
W: 10:30AM – 12:15 PM
Biometric recognition, or simply biometrics, is the science of establishing the identity of a person based on physical or behavioral attributes. In this course we will cover the three primary modalities of biometric recognition, namely fingerprint, face, and iris. We will also introduce other emerging technologies such as recognition of gait, hand geometry, and ear. Other topics will include the security of biometrics, statistics for biometric evaluation, spoofing, ethical issues related to biometric technology, the relation to forensic science, and the impact biometric recognition has had on the judicial system.
- To develop a fundamental knowledge in the phases of of biometric system for identification and verification tasks.
- To quantitatively and qualitatively evaluate the strength and weaknesses of several biometric modalities from measures, such as error metrics, usability, and public perception, and apply these skills to emerging biometric technologies.
- To assess the boundaries between privacy, security, and ethics and the impacts this has on large scale implementations of biometric systems.
- Required Text: Introduction to Biometrics . Jain, Anil, Ross, Arun A., Nandakumar, Karthik. Springer, 2011.
- Statistical Programming Tool: R, Matlab, Python, etc.
- History of Biometrics and Biometrics in the Media
- Biometric Systems
- Privacy and Ethics
- Fingerprint Recognition
- Face Recognition
- Iris Recognition
- Multimodal Biometrics
- Soft Biometrics and Other Modalities
- Biometric Spoofing
- Human vs. Computer Performance
- Biometric Deployment Case Studies
- Additional Topics TBD
|45%||Homework and Assignments|
Each week, a short quiz (10-15 minutes to complete) will be given at the start of the class on Tuesday. We will review the quiz in class on Thursday. There are ten (10) quizzes, and the lowest grade will be dropped.
Four (4) homework assignments will involve a significant amount of coding to generate and/or manipulate data from biometric systems. Additionally, there will be a writing assignment will be on the topic of privacy, security, and ethics. There will be a film shown, tentatively scheduled for the third week of class, which will aid in developing your writing assignment.
Questions on the final exam will be in the same format as the weekly quizzes, so it would be wise to review your quizzes each week. Information regarding the time and location will be posted when it becomes available.
Assignments may be turned in one day late with a single letter grade deduction (10%). Except under extenuating circumstances or when otherwise specified, assignments will not be accepted after 24 hours of the due date.
Grading is based on a percentage scale. Grades between 90 – 100% are an A; less than 90% and above 80% will be given a B; less than 80% and above 70% a C; less than 70% and above 60% a D, and less than 60% an F. High and low grades within each range are given a + or -, except for the A range, where — in keeping with college policy — an A+ is not offered.
Students with documented disabilities who believe that they may need accommodations in this class are encouraged to contact me as early in the semester as possible to ensure that such accommodations are implemented in a timely fashion. Assistance is available to eligible students through Student Accessibility Services. Please contact Jodi Litchfield or Courtney Cioffredi, the ADA Coordinators, for more information: Courtney Cioffredi can be reached at email@example.com or 802-443-2169 and Jodi Litchfield can be reached at firstname.lastname@example.org 802-443-5936. All discussions will remain confidential.
Middlebury College supports an inclusive learning environment where diversity and individual differences are understood, respected, appreciated, and recognized as a source of strength. It is expected that students in this class will respect differences and demonstrate diligence in understanding how other people’s perspectives, behaviors, and world views may be different from their own.
Collaboration Policy and Honor Code
Students are expected to uphold the Honor Codeoutlined by Middlebury College, and should have a good understanding of what constitutes as cheating or plagiarism. For a refresher, please review Middlebury’s Writing and Plagiarism Guides.
The computer science faculty believes that collaboration fosters a healthy and enjoyable educational environment. For this reason, you are encouraged to talk with other students about the course, to form study groups, and to use the online discussion forum to post questions and help each other.
Unless otherwise instructed, feel free to discuss problem sets with other students and exchange ideas about how to solve them. However, there is a thin line between collaboration and plagiarizing the work of others. Therefore, it is required that you must compose your own solution to each assignment. In particular, while you may discuss strategies for approaching the programming assignments with your classmates and may receive debugging help from them, you are required to write all of your own code. It is unacceptable(1) to write a program together and turn in two copies of the same program or (2) to copy code written by your classmates. This implies that you should never have in your possession a copy of all or part of another student’s work. It is your own responsibility to protect your work from unauthorized access. The exams must be entirely your own work.
In keeping with the standards of the scientific community, you must give credit where credit is due. If you make use of an idea that was developed by (or jointly with) others, please reference them appropriately in your work. E.g., if you get a key idea for solving a problem from person X, your solution should begin with a note that says “I worked with Xon this problem” and should say “The main idea (due to X) is …” in the appropriate places. It is unacceptablefor students to work together but not to acknowledge each other in their write-ups.
When working on homework problems, it is perfectly reasonable to consult public literature (books, articles, etc.) for hints and techniques. However, you must reference any sources that contribute to your solution. It is also OK to borrow code from the textbook, from materials discussed in class, and from other sources as long as you give proper credit. Assignments and solutions from previous terms are notconsidered to be part of the “public” literature, and consulting problem set solutions from previous terms constitutes a violation of the Honor Code.
If you are uncertain how the Honor Code applies to a particular assignment, please ask me. The Department of Computer Science takes the Honor Code seriously. Violations are easy to identify and will be dealt with promptly.
|Date||Topic/Paper||Reading Assignments and Info Re:Quizzes|
|9/12||Course Overview, Biometrics in Media, History of Biometrics|
|9/14||Introduction to Biometric Systems|
|9/19||Errors and Performance Measures||Quiz 1: Biometric Systems|
|9/21||Design Cycle of Biometric Systems|
|9/26||Privacy and Security||Quiz 2: Biometric Systems|
|9/28||Fingerprint Recognition Logistics|
|10/3||Fingerprint Alignment and Matching||Quiz 3: Fingerprint Recognition|
|10/5||Face Recognition Logistics|
|10/10||Face Detection (Integral Images, Haar-like features, Viola-Jone)||Quiz 4: Face Recognition|
|10/12||Appearance-based Face Recognition (PCA, ICA, and LDA)|
|10/17||Appearance-based Pt.2||Quiz 5: Face Recognition and Integral Images|
|10/19||Texture-based Face Recognition|
|10/26||Advanced Topics in Face Recognition||Please read pp. 129-138 in the Course Textbook and “DeepFace: Closing the Gap to Human-Level Performance in Face Recognition“|
|10/31||Introduction to Iris Recognition||Please read pp. 141-159 in the Course Textbook
Quiz 6: Appearance-based, model-based, and texture-based face recognition
|11/2||Iris Recognition (continued)||Daugman, John. “How iris recognition works.” IEEE Transactions on circuits and systems for video technology 14.1 (2004): 21-30.|
|11/7||Multibiometrics||Please read pp. 209-224 in the Course Textbook
Quiz 7 (Section 4.1 – 4.5 ) and Daugman’s Paper “How Iris Recognition Works.”
|11/9||Biometric Fusion||Please read pp. 224-256 in the Course Textbook|
|11/14||Applications of Biometrics||Quiz 8: Multibiometrics pp. 209-256|
|11/21||Spoofing and the Biometric Menagerie|
|11/28||Quiz 9: Biometric Security|
|11/30||Biometrics as a Forensic|
|12/5||In-Class Assignment||Quiz 10: Biometric Security|