Recent approaches to crater detection have been inspired by face detection's use of gray-scale texture features. Using gray-scale texture features for supervised machine learning crater detection algorithms provides better classfication of craters in planetary images than previous methods. When using Haar features it is typical to generate thousands of numerical values from each candidate crater image. This magnitude of image features to extract and consider can spell disaster when the application is an entire planetary surface. One solution is to reduce the number of features extracted and considered in order to increase accuracy as well as speed. Feature subset selection provides the operational classifiers with a concise and denoised set of features by reducing irrelevant and redundant features. Feature subset selection is known to be NP-hard. To provide an effcient suboptimal solution, four genetic algorithms are proposed to use greedy selection, weighted random selection, and simulated annealing to distinguish discriminate features from indiscriminate features. Inspired by analysis regarding the relationship between subset size and accuracy, a squeezing algorithm is presented to shrink the genetic algorithm's chromosome cardinality during the genetic iterations. A significant increase in the classification performance of a Bayesian classifier in crater detection using image texture features is observed.
Collaborative Data Mining Research Center (DMRC) (University of Massachusetts President's Science and Technology ST Fund 2011) The DMRC builds on existing data mining research at UMass Boston and designs and implements efficient and effective data mining algorithms to solve challenging research problems in environmental, social, economic and health-related issues, bringing together faculty from several colleges and disciplines at the University of Massachusetts Boston.
Reliably Predicting the Location, Time, and Likelihood of Future Criminal Activity in Boston (Department of Justice)
Working with the Boston Police Department, the work intents to explore a methodology for reliably predicting the location, time, and/or likelihood of future criminal activity. The work will allow BPD to deploy police resources in the areas where they can most effectively reduce the incidence. The project will focus on property crime because it comprises the bulk of Part I crime in the city of Boston.
Automatic Detection of Sub-Kilometer Craters in High Resolution Planetary Images. (NASA Proposal No. 08-AISR08-0022) Impact craters, the structures formed by collisions of meteoroids with planetary surfaces, are among the most studied geomorphic features in the solar system because they yield information about the past and present geological processes and provide the only tool for measuring relative ages of observed geologic formations. Thus, surveying impact craters is an important task in the planetary research. Planetary probes deliver ever increasing volume of high resolution images; however, the scientific utilization of these images is hampered by the lack of tools for their effective automated analysis. This project seeks to develop a processing pipeline for fast and accurate surveys of small craters from high resolution images. Such system will make possible assembling global, “million crater” catalogs of sub-kilometer craters on Mars, Mercury, and the Moon.
Discovery of Geospatial Discriminating Patterns from Remote Sensing Datasets. (UMass Boston Proposal Development Grant) Remote sensing data, pertaining to geosciences, consists of satellite observations of climate, vegetation cover, terrain topography, lithology, soil properties, etc. A large number of such datasets is available in the public domain within the framework of Geographic Information Systems (GIS). Remote sensing datasets offer an unprecedented opportunity for studying various aspects of Earth Science and to predict and address environmental and other ecological problems. For example, understanding the spatial variation of drainage density—the density of land surface dissection by river networks—is related to the problem of assessing the risk of damage and degradation of the landscapes. Studying the spatial distribution of carbon flux, controlled by land precipitation, land and ocean temperature, and terrestrial biomass loss, leads to a better understanding of global warming. Sufficient data exist to address such problems, what is lacking is a methodology that can efficiently distill vary large amount of data into a usable knowledge. Data mining techniques are well suited to provide such methodologies. In this project, we study geospatial discriminate patterns and similarity measures to capture and summarize complex interactions among geospatial variables.
Past Research Projects
A Gaze-Controlled Interface to Virtual Reality Applications for Motor- and Speech-Impaired Users. (UMass Joseph P. Healey Grant Program) A virtual reality is a computer-simulated online environment, and its users create avatars to inhabit it and interact with each other. Because the multifaceted nature of virtual worlds offers vivid and rich perceptual stimuli, the experience can feel astonishingly real. As our society progressively invents and integrates more advanced technologies including virtual worlds, overcoming technology access barriers for different user groups and promoting equalization across the whole society become even more critical. Research shows that virtual worlds can be especially beneficial to physically challenged users, not only to enhance their independence and mental health, but also to increase career and education opportunities for them. The aim of the proposed project is to overcome the access barriers of virtual worlds for motor- and speech-impaired users by building a gaze-controlled interface for virtual worlds that will enable them to interact with the virtual world by just moving their eyes. The proposed interface will specifically address the requirements of virtual worlds with regard to real-time interaction among multiple users through multimodal communication channels.
Crater Seeker Video Game
We have established a web-based Mars Crater Seeker video game for K-12 students, teachers, and the public using real data from mars missions. This video game is designed to showcase the challenges and fun of driving rovers on Mars. This ongoing outreach project uniquely combines the data, systems, and resources of four existing NASA-funded research projects.