Technical Report Number
Segmentation of 3D and time-varying volumetric (4D) image data is considered a time and resource intensive bottleneck in scientific endeavors. Automatic methods are becoming more reliable, but many data sets still require manual intervention. This can mainly be attributed to the characteristics of the image data not being amenable to automated methods, the existence of variations in or poor image quality, or the need for an expert to review and edit results from an automatic technique. Manually segmenting volumetric data is a challenge even for those more experienced. Understanding the 3D nature of the data and navigating through the 3D environment poses some of the main difficulties of the task. Understanding what it means to segment data, and what a contour and a set of contours portrays is key to producing meaningful and usable results. Our goal is to construct a system that will allow even novices to segment 3D data and produce results that experts and scientists can review and use. Through the use of fe tures such as tutorials, contouring protocols, navigation aids, and other interface enhancements, we believe that we can create a workflow that will allow any level of user to work with and understand 3D data. We propose a guided segmentation system that (1) lets experts create data specific segmentation aids, (2) helps users create a meaningful segmentation using these aids, and (3) allows the experts to use the results to provide helpful information to scientists.
Vaughan, Michelle; Grimm, Cindy; Sowell, Ruth; Pless, Robert; and Kobourov, Stephen, "Specializing Interfaces for Citizen Science Segmentation of Volumetric Data" Report Number: WUCSE-2012-42 (2012). All Computer Science and Engineering Research.