Image and Spatial Data Analysis Division
Kenton McHenry received a Ph.D. degree in computer science from the University of Illinois at Urbana-Champaign in 2008 after completing a B.S. degree from California State University of San Bernardino. He is currently a Research Scientist at the National Center for Supercomputing Applications (NCSA) and lead of the Image and Spatial Data Analysis (ISDA) group. His background is in computer vision with interests in the areas of image segmentation, object/material recognition, and 3D reconstruction. At NCSA he has applied this experience towards the task of cyber-infrastructure and digital curation, specifically, digital preservation and access.
In collaboration with National Archives Applied Research Kenton has worked on a series of tools focused around the need for file format conversions in digital archives. One such tool, NCSA Polyglot, is a file format conversion service that was designed as an extensible, scalable, and practical solution to accessing data among the many contemporary (and legacy) file formats available. The Polyglot service is built on top of another tool developed by the ISDA group, the Software Server, a service that runs on desktop machines to provide a uniform interface to capabilities within installed applications. These servers turn traditional desktop software into web based services and are what allow Polyglot to carry out a potentially large number of format conversions. More recently Kenton and the ISDA group are investigating means of providing low cost searchable access to digitized archives containing handwritten data. For example the 1930 Census data, consisting of over 3 million un-transcribed digitized forms the goal of this work is to explore methods of allowing users to find information within the image data without months of effort by thousands of human transcribers. Another project involving applied computer vision is GroupScope. With the goal of studying large group behavior, GroupScope brings together overlapping collections of video and audio and attempts to automate and/or aid in the identification of interactions between the people observed.
Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer