To increase student’s knowledge and to improve hands-on skills and to prepare them for professional work environment as well as for graduate studies, teaching activities will be fostered with research activities for undergraduate students. Students will be supported to have summer research experience and internships as discussed in Section 2.3. Moreover, each new course will have a research or project component. Topics will be developed in coordination with the DHS Center for Border Security and Immigration at UTEP. Some examples as outlined below.
a.) UAV assembly and operations for wide area surveillance (Dr. Selahattin Ozcelik, ME)
Motivation: UAVs have proven remarkably useful in the skies since they have been used for various military operations and are being used by CBP’s office of Air and Marine for border surveillance and reconnaissance. Increasingly, DHS has explored the use of UAVs in support of U.S. Border Patrol’s mission to augment its agents’ ability to patrol the border. Apparently, UAVs are becoming an integral part of DHS in border security as well as for purposes of surveillance and reconnaissance during natural disasters. Therefore, this project aims giving hands-on experience to students in the fundamentals of flying mini aircrafts equipped with vision system.
Project Description: The project will involve an operation of a small-scale remotely controlled aircraft for surveillance and image capturing tasks. The project will involve the assembly of some equipment, design and development of necessary interfacing for communication between the airplane and ground station, collect video recordings and images and analyze them for the given tasks of the project.
Undergraduate Research Opportunities: Students in teams will have the opportunity to gain hands-on experience on simple UAVs. Multidisciplinary teams formed by students from different engineering and science majors will learn major components of UAVs, investigate how to control a UAV and how to design communication and vision systems. They will also investigate on image capturing techniques.
b.) Image Enhancement Techniques with Spatial Domain Filters/ Edge Detection for Homeland Security Applications (Dr. Nuri Yilmazer, EE)
Motivation: The images taken from the UAV cameras contain some undesired effects such as noise, EM interference and degradations effects. These unwanted effects can be minimized by applying some pre-processing and filtering techniques as well as restoration techniques. The images usually contain Gaussian, impulse noise as well as periodic noise usually arises during the image acquisition from electrical or electromechanical interference which is spatially dependent noise. There are signal processing techniques that can be used for restoring the degradations in the images. The objective of this work is to reduce noise in UAV images.
Project Description: In this project, spatial filtering techniques will be investigated to enhance the images. The students will investigate some spatial filtering techniques such as mean filters, order-statistic filter, and adaptive filters for noise reduction in images. Sharpening spatial filters known as Laplacian and Gradient filters will be employed as well as some intensity transformations such as log, power-law transformations, histogram equalization, and matching techniques to enhance the images.
Undergraduate Research Opportunities: Students will participate in developing image processing techniques to enhance the images. They will conduct the simulations by using MATLAB programming. The students will learn the spatial filtering techniques and edge and line detection techniques which are widely used in image processing field. They will gain the great in-depth of different types of filters and compare the performances of each one of them.
Motivation: Operations research (OR) has had a long and distinguished history of work in homeland security applications. OR models and algorithms have been developed and used in airline security, transportation of hazardous materials, emergency preparedness and response, and threat analysis. However, papers on border security are just beginning to appear recently.
Project Description: Most of current border security research is using fingerprint images to detect the entry of criminals, suspected terrorists, and illegal immigrants. In this proposal, we propose to user UAVs to catch images, and then use image data mining to identify the patterns of suspected illegal behaviors and individuals. After that, optimization and decision making models will be used to schedule the border’s control inspection activities and locations. A multi objective optimization and decision making support model will be developed to minimize the illegal entry to the states and threat risk using the available resources.
Undergraduate Research Opportunities: Students will participate in developing the decision making support system. They will conduct the simulations by using SOLVER and MATLAB programming. The students will learn and use the different approaches and tools to solve multi objective optimization problems.
d.) Data Processing and Mining (Dr. Mais Nijim, CS)
Motivation: Data mining is becoming one of the fastest growing technologies as the need is becoming more to finding the important information and features from the captured data. As it is well known, not all data captured is important and related to the problem. This is the case especially, in video and image capturing for purposes of surveillance and monitoring, in which data is collected on a regular basis and analyzed for later use. There is a need to identify and develop algorithms to classify and find the related data. There are some image processing techniques and data mining algorithms that could be used for the purpose of classifying images into clusters for analysis and usage. The major objective of this work is to come out with a best image mining approach that effectively classifies images and model them for real time usage.
Project Description: The process of capturing images for monitoring and remote sensing is becoming very important. There are several mining algorithms that could be used and developed to extract patterns and behaviors from these images. In this project, we will focus on the several feature extraction techniques along with image processing and analysis algorithms and several prediction schemes. There are several advantages to using data and image mining to extract patterns and features from the captured images. Several preprocessing, processing and post processing image mining algorithms will be developed. These algorithms will find the patterns and cluster the features together while storing them, so that it is easier to extract features and help semi automating the whole process from capturing to storing.
Undergraduate Research Opportunities: Students will participate in developing image and data mining techniques that will help in the image mining algorithms. Students will be able to learn the current related theoretical algorithms and conduct simulations. They will learn feature extractions and predictions that could be used in other computer science or engineering related fields. This will broaden the knowledge of the students in real life applications and help them develop their skills for practical problems that face them as they use and collect information for their personal and professional use as they graduate.
e.) Information Analysis and Modeling (Dr. Hua Li, IE)
Motivation: As stated in the DHS report , improved analysis and decision-making tools are part of the representative technology needs. The effort is needed to research ways to fully integrate multiple domains, including technology, managerial, policy, organizational, political, and contextual, to enhance decision making. The man-made events are stochastic and hard to control because almost every single event is unique and represents its own specific situation. Meanwhile, it is not feasible to evaluate the efficiency and effectiveness of different response policies using actual, real homeland security events which have great impacts on the national security. The major objective of this work is to improve the efficiency and accuracy of information analysis and modeling related to homeland security .Basic.net to conduct information analysis and modeling. They will also collect necessary information and data from various national database, such as global terrorism database. By participating in this project, students can improve their simulation and modeling skills and increase their awareness of homeland security related issues.
Project Description: Simulation-based optimization (so-called simulation optimization) has become an active research area in homeland security . Many simulation optimization techniques have been developed, e.g., scatter search  and surrogate search , but few place emphasis on the evolutionary perspective of decision making for dynamic systems. Furthermore, a substantial number of simulation evaluations are normally required to obtain a satisfactory solution. This has not been feasible in real-time decision making cases where large-scale, expensive simulations are involved . This proposed research project will develop efficient simulation optimization algorithms to assist in making high-quality, timely decisions and managing online systems. The algorithms will utilize analytical formulations and offline experimental results to guide the online search and provide near-optimal solutions quickly. The methodology will have many potential applications for the limited homeland security resource decision making problems [19-22].
Undergraduate Research Opportunities: Students will participate in developing and optimizing the simulation algorithms. They will learn and use Arena and Visual Basic.net to conduct information analysis and modeling. They will also collect necessary information and data from various national database, such as global terrorism database. By participating in this project, students can improve their simulation and modeling skills and increase their awareness of homeland security related issues.
This page was last updated on: July 30, 2013