Main Content

Md Shafaeat Hossain

Morrill Hall 117AMd Shafaeat Hossain
Phone: 203 - 392 - 5868
Email: HossainM3@SouthernCT.edu

Spring 2017

Office Hours

Monday 1:45 p.m. - 3:15 p.m.
Tuesday None
Wednesday 2:15 p.m. - 3:15 p.m.
Thursday 11:00 a.m. - 1:30 p.m.

CURRENT COURSES

  • CSC 581 Computational Intelligence
  • CSC 555 Principles of Information Security
  • CSC 481 Artificial Intelligence
  • CSC 591 Thesis II: Thesis Completion
  • CSC 590 Thesis I: Proposal Development
  • HON 495 CSC Departmental Honors

Recent Courses

  • CSC545 Advanced Database Systems
  • CSC540 Database Systems
  • CSC335 Database Manaagement Systems
  • CSC212 Data Structures
  • CSC200 Information Management and Productivity Software
  • CSC153 Computer Programming II
  • CSC499 Independent Study and Research
  • CSC494 CSC Honors Prospectus

Education

PhD in Computational Analysis and Modeling
Louisiana Tech University
Ruston, Louisiana.

MS in Computer Science
Louisiana Tech University
Ruston, Louisiana.

MS in Mathematics
Louisiana Tech University
Ruston, Louisiana.

MS in Computer Science and Engineering
University of Dhaka
Dhaka, Bangladesh

B.Sc. (Honors) in Computer Science and Engineering
University of Dhaka
Dhaka, Bangladesh.

Research

Smartphone Security, Cyber-behavioral authentication, Multi-biometric verification, Machine learning, Computer Vision, Data Analytics 

Grant

Md S. Hossain (PI), "Verifier Order Selection in Serial Fusion Based Multi-biometric Verification System", Sponsored by SCSU Minority Recruitment & Retention Committee Grant, November 2016 – August 2017, $2000.00.

Md S. Hossain (PI), "Serial Fusion based Biometric Verification," Sponsored by SCSU Faculty Creative Activity Research Grant, June 2016 - May 2017, $2,500.00.

Md S. Hossain (PI), Lisa Lancor (Co-PI), "Continuous User Authentication on Smartphones," Sponsored by SCSU Faculty Creative Activity Research Grant, June 2015 - May 2016,  $2,500.00.

Md S. Hossain (PI), "On Enhancing Biometric Security Systems," Sponsored by SCSU Research/Creative Activity Reassigned Time, January 2016 - June 2016.

Publications

Peer Reviewed Journal Papers

Md S. Hossain, Khandaker A. Rahman, Md. Hasanuzzaman, Md A. Bhuyian, and H. Ueno. Video Image Clustering Based on Human Face and Shirt Color. In International Journal of Image and Graphics (IJIG), 10(1), page 1-19,  2010. 

Khandaker A. Rahman, Md S. Hossain, Md A. Bhuyian, Tao Zhang, Md Hasanuzzaman, and H. Ueno, Optimized Mask Selection for Person Identification and Camera Distance Measurement based on Interocular Distance. In Journal of Computer Engineering Research, 1(2), page 29-42, 2010.

Peer Reviewed Conference Papers

Mohamed A. Rilvan, Kolby I. Lacy, Md S. Hossain, Bing Wang.  User Authentication and Identification on Smartphones by Incorporating Capacitive Touchscreen. In Proceedings of IEEE International Performance Computing and Communications Conference (IPCCC), Las Vegas, NV, December 2016.

Tudor Boran, John Voss, Md S. Hossain. Authorship Categorization of Public Domain Literature. In Proceedings of IEEE Ubiquitous Computing, Electronics, and Mobile Communication Conference (UEMCON), New York City, NY, October 2016.

Md S. Hossain. On Finding Appropriate Reject Region in Serial Fusion based Biometric Verification. In Proceedings of IEEE International Multi-Disciplinary Conference on Cognitive Methods in Situation Awareness and Decision Support (CogSIMA), San Diego, CA, March 2016.

Khandaker A. Rahman, Ryan Moormann, Danielle Dierich, Md S. Hossain, Continuous User Verification via Mouse Activities . In Proceedings of 8th IEEE International Conference on Multimedia Communications, Services and Security, Poland, November 2015. 

Md S. Hossain, Kiran S. Balagani, and Vir V. Phoha. On Controlling the Genuine Reject Rate in Multi-stage Biometric Verification System. In Computer Vision and Pattern Recognition Workshop (CVPRW), Portland, OR, June 2013.

Md S. Hossain, Kiran S. Balagani, and Vir V. Phoha. New Impostor Score Based Rejection Methods for Continuous Keystroke Verification with Weak Templates. In Proceedings of IEEE Fifth International Conference on Biometrics: Theory, Applications, and Systems (BTAS 2012), Washington DC, 2012.

Md S. Hossain, Khandaker A. Rahman, Md Hasanuzzaman, and Vir V. Phoha. A Simple and Efficient Video Image Clustering Algorithm for Person Specific Query and Image Retrieval. In Proc. of International Conference on Internet Multimedia Computing and Service, China, 2009.

Khandaker A. Rahman, Md S. Hossain, Md. A. Bhuyian, Tao Zhang, Md Hasanuzzaman, and H. Ueno. Eye-Distance Based Mask Selection for Person Identification. In Proc. of International Conference on Multimedia and Ubiquitous Engineering (MUE 2009), China, 2009.

Khandaker Abir Rahman, Md S. Hossain, Md. A. Bhuyian, Tao Zhang, Md Hasanuzzaman, and H. Ueno. Person to Camera Distance Measurement Based on Eye-Distance. In Proc. of International Conference on Multimedia and Ubiquitous Engineering (MUE 2009), China, 2009. 

Research in Progress

Continuous User Authentication on Smartphones

To ensure high security on smartphones, it is necessary to increase the level of authentication beyond the standard point-of-entry technique. To this end, we plan to develop a user authentication scheme based on behavioral biometrics, which continuously authenticates users during phone usage rather than simply at the beginning of a session. Specifically, we will be using keystroke dynamics (how a user types on a keypad), touch screen dynamics (how a user interacts with the touch screen (finger pressure, finger tip size, etc.), physics of the device (how a user carries/holds a phone), and linguistic profiling (how a user writes vocabulary and text) to authenticate a user.

Smartphone Security using Body Parts and Voice

To strengthen the user authentication in Smartphone, we plan to develop a biometric authentication system which uses the capacitive sensor of touchscreen  and voice recorder of a smartphone. While capacitive fingerprint scanners which are integrated in recent mobile phones require high quality and expensive sensors, our methodology focus on the touchscreen as the image sensor. Using the touchscreen we can scan users’ body parts, such as ears, palms, and fists. We also incorporate voice recorders to improve the user identification when the user places the touchscreen on his ear to accept an incoming phone call.

Enhancing Multi-biometric Security Systems

Multi-biometric verification systems can be classified into two categories: (1) parallel fusion based and (2) serial fusion based. Though a parallel fusion based system gives high accuracy, its applicability is questionable in many real world applications because of its longer verification time. In contrast, serial fusion can take less time to give a verification decision and thus can provide user convenience. Hence, a serial fusion based system can be a good trade-off between user convenience and accuracy. However, these systems have been poorly investigated. Hence, we plan to investigate these systems and develop a new serial fusion based verification scheme.

Grad STUDENTS

  • Carl Haberfeld
  • Richard Wandell
  • Philip Barello 

undergrad Students

  • Mohamed A. Rilvan
  • John C. Dogan