<title>Image content engine (ICE): a system for fast image database searches</title>

Abstract

The Image Content Engine (ICE) is being developed to provide cueing assistance to human image analysts faced with increasingly large and intractable amounts of image data. The ICE architecture includes user configurable feature extraction pipelines which produce intermediate feature vector and match surface files which can then be accessed by interactive relational queries. Application of the feature extraction algorithms to large collections of images may be extremely time consuming and is launched as a batch job on a Linux cluster. The query interface accesses only the intermediate files and returns candidate hits nearly instantaneously. Queries may be posed for individual objects or collections. The query interface prompts the user for feedback, and applies relevance feedback algorithms to revise the feature vector weighting and focus on relevant search results. Examples of feature extraction and both model-based and search-by-example queries are presented.

FY2006 Engineering Tech Base Report: Defect Detection in Large CT Image Sets

“…One can extract a given row (elevation) from each radiograph in this sequence and stack them, creating a sinogram in which small features form a distinct sinusoid pattern as the object rotates. We looked for these sinusoid patterns using the Gradient Direction Matching (GDM) [1] algorithm developed under the Image Content Engine (ICE) LDRD-SI [2], a powerful model-based matching tool (Figure 2). This approach was attractive due to the success of GDM in similar problems, the usefulness of sinogram data to human analysts, and the fact that it does not require image reconstruction.…”

Using gradients, alignment and proximity to extract curves and connect roads in overhead images

“…Systems such as the Image Content Engine (ICE) cue human analysts for content, such as specific types of buildings or moving vehicles, in broad area search images [1]. Once ICE has matched the necessary patterns or extracted the appropriate features from images to be searched, it provides prioritized cues to human analysts for manual confirmation.…”