NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Back to Results
Machine learning for a Toolkit for Image MiningA prototype user environment is described that enables a user with very limited computer skills to collaborate with a computer algorithm to develop search tools (agents) that can be used for image analysis, creating metadata for tagging images, searching for images in an image database on the basis of image content, or as a component of computer vision algorithms. Agents are learned in an ongoing, two-way dialogue between the user and the algorithm. The user points to mistakes made in classification. The algorithm, in response, attempts to discover which image attributes are discriminating between objects of interest and clutter. It then builds a candidate agent and applies it to an input image, producing an 'interest' image highlighting features that are consistent with the set of objects and clutter indicated by the user. The dialogue repeats until the user is satisfied. The prototype environment, called the Toolkit for Image Mining (TIM) is currently capable of learning spectral and textural patterns. Learning exhibits rapid convergence to reasonable levels of performance and, when thoroughly trained, Fo appears to be competitive in discrimination accuracy with other classification techniques.
Document ID
19960008291
Acquisition Source
Legacy CDMS
Document Type
Conference Paper
Authors
Delanoy, Richard L.
(Massachusetts Inst. of Tech. Cambridge, MA, United States)
Date Acquired
September 6, 2013
Publication Date
October 1, 1995
Publication Information
Publication: NASA. Goddard Space Flight Center, The 1995 Science Information Management and Data Compression Workshop
Subject Category
Documentation And Information Science
Accession Number
96N15457
Distribution Limits
Public
Copyright
Work of the US Gov. Public Use Permitted.

Available Downloads

There are no available downloads for this record.
No Preview Available