System for inspection of large high-resolution radiography datasets

High-resolution image collections pose unique challenges to analysts tasked with managing the associated data assets and deriving new information from them. While significant progress has been made towards rapid automated filtering, alignment, segmentation, characterization, and feature identification from image collections, the extraction of new insights still strongly depends on human intervention. In real-time capture and immediate-mode analysis environments where image data has to be continuously and interactively processed, a broad set of challenges in the image-driven verification and analysis cycle have to be addressed. A framework for interactive and intuitive inspection of large, high-resolution image data sets is presented, leveraging the strength of the human visual system for large-scale image processing. A case study is provided for an X-ray radiography system, covering the scanner-to-screen data management and representation pipeline, resulting in a visual analytics environment enabling analytical reasoning by means of interactive and intuitive visualization.