Abstract

Our Image Change Detection Toolkit is a powerful group of algorithms useful for many applications in diverse disciplines, including remote sensing, surveillance, medical diagnosis and treatment, civil infrastructure, and underwater sensing. The main objective of the undergraduate research experience was to learn about change detection methods and the practicality of “Diverse problems, similar solutions”. The first application was an environmental one in which different change detection algorithms were applied to images of underground pollutants (DNAPL). These algorithms were used to detect the area of contamination in the soil by comparing two images of the same experiment acquired at different times. Using the results, R.O.C. curves were created to compare the accuracy of the algorithms. The second application was a biological one, which required an accurate quantification of the heartbeat region of a zebra fish embryo. This information was needed to determine the location, period, frequency and phase of each pixel intensity vector that corresponded to the changing region. Finally with the use of the phase value of each pixel intensity vector, the heart beat region was segmented into two groups which were out of phase by 180 degrees. In summary, we were able to use change detection algorithms to determine the heart beat region, make quantitative measurements, and highlight the two phases of the heartbeat.

Notes

Poster presented at the 2006 Thrust R2D Image Understanding and Sensor Fusion Methods Conference

Keywords

Image Change Detection, heartbeat, R.O.C., soil, DNAPL, civil infrastructure, surveillance

Subject Categories

Diagnostic imaging

Disciplines

Engineering

Publisher

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

Publication Date

2006

Rights Holder

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


View Full Text

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

Included in

Engineering Commons

Share

COinS