LAB MISSION: To advance the field of machine learning, deep learning, biometrics, data collection, and data analytics through cutting-edge research and development. Our team strives to enhance our understanding of these disciplines by exploring new and innovative approaches to data collection, analysis, and interpretation. Our goal is to create the next generation of students and researchers that build academic, government and industry focused solutions with real-world impact. Throughout the journey of research and development, our published work and solutions strive to improve the level of understanding of novel technologies by the research community. Our lab believes in the power of collaboration and interdisciplinary teamwork, and we are dedicated to fostering a supportive and inclusive environment for all members of our research community.

LAB OVERVIEW: The Multispectral imagery lab (MILAB) conducts research, development, evaluation and testing of multi-modal and multi-spectral data processing technologies to address national security challenges. Our core research competences are data collection activities, computer vision, machine and deep learning, data analytics, computational modeling, data visualization, image restoration, signal and image processing, and computer science. MILAB’s activities involve first-class interdisciplinary research with other research groups, collaborations with private industry and academic partners across the nation, and mission-specific support for national security sponsors. MILAB thrives in a culture of respect, diversity, curiosity, and discovery.

LAB EXPERTISE

Biometrics: Data Collection: Extensive data collection activities using any modality and at any condition. Algorithmic Development: Design and development of within-spectral and cross-spectral matching techniques while using visible, SWIR and thermal images of different biometric traits.We are also working on developing face, eye, nose, ear detection techniques that can operate when using images captured at any band. See details on our publications page.

Forensics: We have been working on the problem of restoring severely degraded images such as images scanned from passport photos or images subjected to fax compression, down-scaling and printing. Some of our expertise are image de-noising, in-painting, de-blurring and super-resolution.

Computational Physiology: we have expertise on two main areas: (i) extracting, processing and analyzing human-based physiological signals (including ECG) and (ii) bio-imaging data collection, processing and analysis using computer vision and machine learning algorithms.

Funding: Since 2010, MILab has participated in biometrics- and deception detection-related research projects funded by DoD-DAC, DoD-ONR, DoD-DTRA, DoD-ARL, US Department of Justice, DARPA, DHS and CITeR, i.e. a National Science Foundation Industry/University Cooperative Research Center.

Students: Currently, MILab has several graduate and undergraduate students. Many GRA MILAB students have graduated and found excellent jobs.

Impact: MILab has published several books, book chapters, journal articles and magazines, conference papers and patents. MILab’s founder and director, Thirimachos Bourlai, has been invited to present, chair and/or moderate more than 200 national and international events, a growing number that illustrates MILab’s reputation and a recognition of its research impact.