News & Events
Awards and Achievements
ALERT Research Highlight: The Largest and Most Systematic Re-id Benchmark to Date May 30, 2018
ALERT researchers Professor Octavia Camps (Northeastern University, Project R4-A.1) and Professor Rich Radke (Rensselaer Polytechnic Institute, Project R4-A.3) and their students, Srikrishna Karanam, Mengran Gou, Ziyan Wu, and Angels Rates-Borras, were recently published in the Institute of Electrical and Electronics Engineers’ (IEEE) monthly journal, Transactions on Pattern Analysis and Machine Intelligence (download here). The paper, “A Systematic Evaluation and Benchmark for Person Re-Identification: Features, Metrics, and Datasets,” provides an extensive review and performance evaluation of existing person re-identification algorithms.
Person identification, or re-id, matches observations of individuals across multiple camera views in a network of surveillance cameras, and represents a critical task in most surveillance and security applications. For example, a police officer may want to automatically follow a person of interest tagged at a check-in counter through the branching concourses of an airport. The research team’s recent review and evaluation of re-id algorithms helps characterize what re-id algorithms are currently capable of accomplishing, as well as what is missing and what is possible in the future. In this paper, the researchers discuss insights gained from their study, as well as put forth research directions and recommendations for re-id researchers that would help develop better algorithms in the future. Both Professor Camps and Professor Radke are involved with ALERT’s Research Thrust 4, which focuses on video surveillance and the analysis of video data with novel algorithms. This publication was the result of years of research and collaboration between their respective labs.
The fundamental re-id problem is to compare a person of interest as seen in a “probe” camera view to a “gallery” of candidates captured from a camera that does not overlap with the probe camera. If a true match to the probe exists in the gallery, it should have a high matching score, or rank, compared to incorrect candidates. Since the body of research in re-id is increasing, researchers can begin to draw conclusions about the best combinations of algorithmic subcomponents. In this paper, the researchers present a careful, fair, and systematic evaluation of feature extraction, metric learning, and multi-shot ranking algorithms proposed for re-id on a wide variety of benchmark datasets. Their general evaluation framework considers hundreds of combinations of (1) feature extraction and metric learning algorithms for single-shot datasets and (2) feature extraction, metric learning, and multi-shot ranking algorithms for multi-shot datasets.
The research team evaluated 276 algorithm combinations on 10 single-shot re-id datasets and 646 algorithm combinations on 7 multi-shot re-id datasets, making the proposed study the largest and most systematic re-id benchmark to date. Approaches were evaluated using 17 datasets that mimic real world settings, including the ALERT Airport Re-Identification Dataset. As part of the evaluation, the researchers built a public code library with an easy-to-use input/output code structure and uniform algorithm parameters that includes 11 contemporary feature extraction and 22 metric learning and ranking algorithms. Both the code library and the complete benchmark results are publicly available for community use.
Student Researchers Selected for ALERT Professional Development Award November 27, 2017
November 27, 2017
Three ALERT student researchers have been selected to receive the first ALERT Professional Development Award in November 2017. The winners are Qi Feng, a Ph.D. student working with Prof. Stan Sclaroff at Boston University; Ashraful Islam, a Ph.D. student working with Prof. Richard Radke at Rensselaer Polytechnic Institute; and Abubakar Siddique, a Ph.D. student working with Prof. Henry Medeiros at Marquette University.
The ALERT Professional Development Award is intended to encourage ALERT students to participate in professional development activities throughout the year and to facilitate their future participation in networking and career development opportunities. ALERT selects up to three students each year to win a $1,500 stipend that can be used towards attendance at a professional or academic conference and/or to visit and collaborate with a lab related to their ALERT research project. This year, applications were accepted August through October 2017. More information about next year’s award cycle is forthcoming.
Rappaport Delivers IEEE Distinguished Lectures in Australia and New Zealand November 27, 2017
November 27, 2017
ALERT Deputy Director and Northeastern University Electrical and Computer Engineering Distinguished Professor, Carey Rappaport, delivered three IEEE Distinguished Lectures in Australia and New Zealand earlier this month. The IEEE Antennas and Propagation Society (AP-S) selected Professor Rappaport as a Distinguished Lecturer for 2017-2019. The IEEE AP-S Distinguished Lecturer Program sends experts, the Distinguished Lecturers, to visit active IEEE AP-S Chapters around the world and give talks on topics of interest and importance to the Antennas and Propagation community. Professor Rappaport gave the following talks during his recent visit to Australia and New Zealand:
- “Electromagnetic Sensing and Treatment of Living Things: Using Microwaves to Detect and Treat Disease in Humans and Trees” (Melbourne, Australia)
- “A High Gain Toroidal Reflector Antenna for Multistatic 3D Whole Body Millimeter-Wave Imaging” (Queenstown, New Zealand)
- “Multifocal Bootlace Lens Design Concepts” (Sydney, Australia)
On the value of being an IEEE Distinguished Lecturer, Professor Rappaport states, “The IEEE Distinguished Lecturer Program supports one of the fundamental precepts of academic research: extending the dissemination of knowledge and cutting-edge discovery to as wide an audience as possible. In presenting my team’s engineering work around the world, I have been able to help engineers extend their scientific awareness beyond their traditional themes, and occasionally outside their comfort zones. Although the travel commitment is time-consuming, it has been a great opportunity to meet people and exchange technical ideas.” Professor Rappaport will be delivering another IEEE Distinguished Lecture in January 2018 to the IEEE AP-S chapter in Orlando, Florida.
Professor Rappaport has been a Northeastern University faculty member since 1987, becoming a full professor in Electrical and Computer Engineering in July 2000, and receiving an appointment as a College of Engineering Distinguished Professor in 2011. Professor Rappaport has written over 400 technical journal articles and conference papers on various topics, including electromagnetic wave propagation and scattering computation, microwave antenna design, and bioelectromagnetics. He has also received two reflector antenna patents, two biomedical device patents, and four subsurface sensing device patents.
Student Spotlight: Elizabeth Wig September 27, 2017
Congratulations to Elizabeth Wig, a Northeastern University (NU) Electrical Engineering undergraduate conducting ALERT research, for receiving the Society of Women Engineers GE Women’s Network Scholarship! Elizabeth will receive this award, which comes with a $5,000 stipend, in October 2017 at the SWE Annual Conference in Austin, Texas. Elizabeth has been working with ALERT R3 Thrust Leader, Professor Carey Rappaport since Summer 2016, conducting research on “Computational Models & Algorithms for Millimeter Wave Whole Body Scanning for AIT,” in collaboration with Smiths Detection (Project R3-A.2). When asked how Elizabeth got involved with the project so early on in her undergraduate career, she explained that she met Professor Rappaport at a NU-sponsored ski event and found out about his research while riding up the mountain on a chairlift. However, Elizabeth explained that her interest in this research began much earlier, “When my high school physics class did its electricity and magnetism unit, the symmetry was strikingly beautiful. I loved the way relatively few equations could describe so much of what makes up our world, from why sunrises are so beautiful to the way molecules hold together to Wi-Fi.” The aspect of her research that she is most passionate about is math, and learning about the different ways mathematics can be used to describe and explain our world. This fits in well with her role on the project, which involves developing the model used to detect and characterize potential explosives threats and eliminate false alarms using a millimeter-wave body scanner. She has been working to make and refine the model to improve the accuracy in characterization.
Beyond her recent award, Elizabeth has also published a paper on her work with Mahdiar Sadeghi, a Northeastern graduate student, and Professor Rappaport, and is currently working on her second paper. She and Mahdiar were also asked to present their work at the ADSA15 (Advanced Development for Security Applications) Workshop in November 2016.
Elizabeth has already gained valuable work experience through her Spring 2017 co-op position at Draper Laboratories in Cambridge, MA. There she worked on electrical engineering projects in their Sensors and Imaging Systems group. As for her future career, she hopes to continue her education and complete a Ph.D. program, and if possible, get the opportunity to travel more internationally and work with NASA!
ALERT Researcher Awards & Accomplishments September 27, 2017
September 26, 2017
Professor Bouman Nominated for ACM Gordon Bell Prize
Professor Charles Bouman of Purdue University has been nominated for an ACM (Association for Computing Machinery) Gordon Bell Prize. Professor Bouman co-leads research on the ALERT project, “Toward Advanced Baggage Screening: Reconstruction and Automatic Target Recognition (ATR)” (Project R4-B.1), with the overarching goal of finding the best mapping method from X-ray data to a decision on the relative safety of individual bags in security settings, such as airport checkpoints. Professor Bouman and his research team study the reconstruction problem with the end goal of detection, while also designing algorithms for image analysis that can best exploit the improved image quality in iterative methods. Their aim is to reduce the false alarm rate without sensitivity loss in detection. They hope to eventually reduce security costs to the transportation industry.
The Gordon Bell Prize is awarded each year to recognize outstanding achievement in high-performance computing. The purpose of the award is to track the progress over time of parallel computing, with particular emphasis on rewarding innovation in applying high-performance computing to applications in science, engineering, and large-scale data analytics. Prizes may be awarded for peak performance or special achievements in scalability and time-to-solution on important science and engineering problems.
Professor Jose Martinez-Lorenzo Awarded $546K DOE Grant
Professor Jose Martinez-Lorenzo of Northeastern University was awarded a $546K grant from the Department of Energy (DOE) for “Fusing Thermoacoustic, Electromagnetic and Acoustic/Seismic Wave Fields for Subsurface Characterization and Imaging of Flow Transport.” According to the DOE, “The overarching goal of this research program is to gain knowledge on the theory and experimental validation of a new unified sensing and imaging methodology for coupling Electromagnetic (EM), Acoustic/Seismic (AC/S), and novel Thermoacoustic (TA) physical fields, which will be applicable to multi-physics and multi-scale material characterization and underground imaging of fluid flow in porous media.” This research will help Professor Martinez-Lorenzo build upon his work with ALERT, specifically Project R3-B.1 and Project R3-B.2.
Professor Otto Gregory Awarded Patent for Gas Sensor System
Professor Otto Gregory of the University of Rhode Island was awarded a patent for “Systems and Methods for the Detection of Compounds” on September 12, 2017. Triacetone-Triperoxide (TATP) is an explosive commonly used in improvised explosive devices (IEDs) and is very difficult to detect using conventional explosives detection techniques, because most of these techniques were developed for nitrogen-based chemistries, not peroxide-based chemistries. In addition, TATP readily sublimes at room temperature, meaning that it can only be found in relatively high concentrations in the vapor phase compared to other commonly found explosives used in IEDs. Professor Gregory’s invention provides a gas sensor system for detection of a compound that decomposes upon exposure to a metal oxide catalyst, and incorporates the exposure of the compound to a microheater, which allows accurate detection to occur at smaller concentrations. For more information on Professor Gregory’s research with ALERT, see Project R2-B.1.
Patents Awarded to ALERT PI July 29, 2017
Dr. Rinaldi Awarded Two Patents for Resonators
Dr. Matteo Rinaldi, Associate Professor of Electrical and Computer Engineering at Northeastern University and principal investigator for Project R2-B.3, has been awarded two patents for his work on nano- and microelectromechanical resonators. Dr. Rinaldi’s research proposes to develop innovative Nano-Electro-Opto-Mechanical (NEOM) sensing technology platforms for the chemical analysis of trace explosives residues. This work contributed to the creation of the novel resonators patented earlier this month. Congratulations to Dr. Rinaldi and his research team for this exciting achievement!Read More
Kurt Jaisle Selected as Finalist in IEEE AP-S Student Paper Competition May 30, 2017
ALERT Student and Northeastern University Scholar, Kurt Jaisle has been selected as a finalist in the 2017 IEEE Antennas and Propagation Symposium’s (AP-S) Student Paper Competition for his paper, “Ray-Based Reconstruction Algorithm for Multi-Monostatic Radar in Imaging Systems.” Being selected as a finalist is quite an accomplishment, as each paper submitted to the IEEE AP-S Student Paper Competition undergoes three independent reviews from experts in the student’s field of study. Kurt’s submission was selected out of 159 papers, most of which were submitted by doctoral students. Kurt is a third year undergraduate student majoring in Electrical Engineering and conducts ALERT research with Professor Carey Rappaport on the R3 Research Thrust (Bulk Sensors & Sensor Systems).
The topic of Kurt’s paper is relevant to aviation security within the Homeland Security Enterprise. According to Kurt, “Today’s airport security scanners use very computationally demanding algorithms to process sensor data into an image of a passenger. As a result, these scanners require expensive, high-performance computers to complete the algorithms in a reasonable amount of time. Yet even with these powerful machines, it can still take several seconds for a scan to be processed.” In his paper, Kurt details a new algorithm that would result in significantly faster processing times (resulting in shorter lines at airport security checkpoints) and a reduction in the cost of the computer hardware used in scanners, potentially making the technology more accessible for broader security applications.
Under the guidance of Professor Rappaport, Kurt began coding the algorithm in the fall of 2015. Over the course of a year, Kurt brought the algorithm from a rudimentary 2D simulation to a functional 3D simulation worthy of publication. Reflecting on his experience conducting research with Professor Rappaport, Kurt states, “Aside from a great deal of technical knowledge, I think the most important thing I have learned from Professor Rappaport is to not leave an endeavor half-finished. Even when I was stuck on a technical challenge for weeks at a time, he would remind me that progress in research is non-linear and that it was worth seeing it through so that I could eventually share my work with the broader community.
Kurt’s interest in engineering was sparked during middle school, when he became involved in FIRST Robotics, a program that aims to develop young STEM leaders through robotics competitions. As time passed, Kurt became interested in the electrical side of engineering and decided to study Electrical Engineering at Northeastern University. After graduation, Kurt plans to pursue a master’s degree in the context of analog electronics, and is hoping that his upper level Electrical Engineering courses, co-op opportunities, and research experiences will help him choose a specific topic of study.
Kurt will present on his selected paper at the IEEE AP-S Symposium in San Diego, California in July. The Student Paper Competition Committee Chair will announce the first, second, and third place winners at the IEEE A-PS Symposium’s Annual Awards Ceremony.
ALERT Research Highlights March 31, 2017
March 31, 2017
ALERT Thrust R3 Project Investigators, Dr. Carey Rappaport and Dr. Jose Martinez-Lorenzo of Northeastern University were awarded a patent for “Signal Processing Methods and Systems for Explosives Detection and Identification Using Electromagnetic Radiation” (U.S. Patent 9,575,045) on February 21, 2017. This patent is for an algorithm designed to rule out non-explosive concealed foreign objects concealed under clothing and affixed to the skin, reducing the number of false alarms, and thus, the number of pat-downs needed, leading to greater accuracy in threat detection and shorter security lines. The improved reliability would benefit many: passengers, airlines, and the Transportation Security Administration; and possibly lead to the expansion of AIT Millimeter Wave Scanners into everyday use, such as railway stations, sporting venues, and other soft targets. Read more about their research here.
Dr. Rappaport was recently selected by the IEEE Antennas and Propagation Society (AP-S) as a Distinguished Lecturer for 2017-2019. The IEEE AP-S Distinguished Lecturer Program sends experts, the Distinguished Lecturers, to visit active AP-S Chapters around the world and give talks on topics of interest and importance to the Antennas and Propagation community. Read more here.
Dr. Martinez-Lorenzo was recently awarded a $500K National Science Foundation (NSF) CAREER Award for his work on developing a method for “4D mm-Wave Compressive Sensing and Imaging at One Thousand Volumetric Frames per Second.” Millimeter-wave sensing and imaging systems are generally used for a wide range of applications, such as security monitoring to detect potential threats at the airport and biological imaging for wound diagnosis and healing. Because this is the first four-dimensional millimeter-wave imaging system that can operate in quick-changing scenarios, it will benefit society greatly. Read more about how Dr. Martinez plans to use this award here.
ALERT Thrust R2 Project Investigator, Dr. Steve Beaudoin of Purdue University was recently awarded “Best Presentation” in his session for a paper he presented on at the Annual AlChE Meeting in November 2016. The paper was based on his research project that was recently selected as a new ALERT project. The new project, titled “A Novel Method for Evaluating the Adhesion of Explosives Residue,” aims to provide insight into the reasons why explosives residues stick to surfaces and what must be done to effectively detect those residues in air transportation security environments. Read more about Dr. Beaudoin’s research here.
Prof. Rappaport and Prof. Martinez-Lorenzo Present ALERT Research at EuCAP 2017 March 24, 2017
Dr. Carey Rappaport and Dr. Jose Martinez-Lorenzo of Northeastern University presented ALERT-related research at the 11th European Conference on Antennas and Propagation (EuCAP 2017) in Paris, France this week (March, 19-24, 2017). EuCAP 2017 is organized by the European Association on Antennas and Propagation (EurAAP), and since 2006, this major event has been bringing experts from academia, research centers, and industry together.
Dr. Rappaport presented a paper entitled, “Modeling the Response of Dielectric Slabs on Ground Planes Using CW Focused Millimeter Waves,” which he co-authored with researcher, Dr. Ann Morgenthaler, and ALERT students, Mahdiar Sadeghi and Elizabeth Wig. In this paper, the researchers present a novel non-iterative model based on ray analysis to characterize non-metallic, weak dielectric objects (like threat objects) on the surface of a highly conducting background (like the human body) using a focused continuous millimeter-wave sensor.
Dr. Martinez-Lorenzo presented a paper entitled, “High Capacity Imaging Using an Array of Compressive Reflector Antennas,” which he co-authored with ALERT students, Ali Molaei and Galia Ghazi, and researchers, Dr. Hipolito Gomez-Sousa and Dr. Juan Heredia-Juesas. In this paper, the authors propose to use an array of six compressive reflector antennas (CRAs), in order to be able to image an extended human-size region. A CRA is created by distorting the surface of a traditional parabolic reflector antenna. As a result of using CRAs, pseudo-random spatial codes are created at the imaging region. Because these spatial codes increase the information collected by each measurement, a smaller number of measurements are needed, which translates into less imaging time. Additionally, the electromagnetic cross-coupling between adjacent CRAs is used to enhance the sensing capacity of the system, as well as to extend the region that it can image. Current security checkpoints use a pause and pose sensing approach for passengers, and require divestment and recollection of passengers’ possessions. This results in a slow throughput in the overall system, and long lines at security check-points. In the presented work, the researchers have developed a fast, fully electronic system that will not require a pause and pose approach, resulting in quick and accurate screening of passengers and their belongings.
Professor Steve Beaudoin Awarded “Best Presentation” for ALERT Research at Annual AIChe Meeting March 6, 2017
ALERT researcher, Professor Steve Beaudoin of Purdue University was awarded “Best Presentation” in his session for a paper he presented on at the Annual AlChE Meeting in November 2016. The paper was based on his research project that was recently selected as a new ALERT project. The new project, titled “A Novel Method for Evaluating the Adhesion of Explosives Residue,” aims to provide insight into the reasons why explosives residues stick to surfaces and what must be done to effectively detect those residues in air transportation security environments.
The ultimate outcomes of this project include a library documenting the adhesion characteristics of explosives residues of interest against surfaces of interest, coupled with a spreadsheet that will allow members of the community to calculate adhesion forces between explosives and surfaces of interest. The end-users of the information developed by this project will include members of the Homeland Security community who are engaged with developing apparatuses, materials (swabs), and methods for contact sampling. Specifically, this research will help them to interpret the results of their developmental work and to guide the creation of next-generation materials and methods for detecting explosives residues.
Professor Beaudoin will be presenting the outcomes of this research thus far at the annual Trace Explosives Detection (TED) Workshop in April, 2017. This presentation will include all of the data collected, as well as a tutorial that illustrates how members of the Homeland Security community can use the data to calculate adhesion forces.