Abstract

The use of satellite-based remote sensing is a cost-effective approach to documenting changes over large geographic regions. However, the large volumes of multi- or hyper-spectral data make systematic exploitation of earth observation data challenging. There is a long history of developing automated change detection (CD) systems to aid users in interpreting vast arrays of data. Change detection is a technique that employs two or more time-separated images taken of the same geographic location and labels each spatial location according to whether there is a significant change in the images over time. Traditional CD methodologies employ a single change hypothesis, and do not interpret the causes behind the observed change. On the other hand, automatic interpretation of detected changes enables automatic labeling of images with meaningful semantic descriptions which can be used to aid in database querying, early warning systems, ecological system exploration, and defense applications. In this work, we construct in automatic, qualitative explanation of object-level changes in multi-spectral remote sensing imagery. Our proposed object-level change interpretation system employs change models that directly reflect the nuisance effects of the data acquisition process, unimportant causes of observed object variation such as illumination variation or spectral composition variation, and important object changes like object shape or location change. This work develops an object-level change understanding framework along with change models for multi-spectral remote sensing imagery, adding value to the R2C and R2D CenSSIS research areas.

Notes

Poster presented at the 2007 Thrust R2D Image Understanding and Sensor Fusion Methods Conference

Keywords

Multispectral Remote Sensing Imagery change models, CD methodologies, Object-Level Change Interpretation, nuisance

Subject Categories

Remote sensing

Disciplines

Engineering

Publisher

Bernard M. Gordon Center for Subsurface Sensing and Imaging Systems (Gordon-CenSSIS)

Publication Date

2007

Rights Holder

Bernard M. Gordon Center for Subsurface Sensing and Imaging Systems (Gordon-CenSSIS)



Click button above to open, or right-click to save.

Included in

Engineering Commons

Share

COinS