Feature extraction from multiple data sources using genetic programming.

Conference · Mon Dec 31 23:00:00 EST 2001

DOI:https://doi.org/10.1117/12.478765· OSTI ID:976121

Szymanski, J J ^[1]; Brumby, Steven P; Pope, P A ^[2]; Eads, D R ^[3]; Galassi, M C ^[4]; Harvey, N R ^[5]; Perkins, S J ^[6]; Porter, R B ^[7]; Theiler, J P ^[8]; Young, A C ^[9]; Bloch, J J ^[10]; David, N A ^[11]; Esch-Mosher, D M ^[12]

---

1. John J.
2. Paul A.
3. Damian R.
4. Mark C.
5. Neal R.
6. Simon J.
7. Reid B.
8. James P.
9. Aaron Cody
10. Jeffrey J.
11. Nancy A.
12. Diana M.

Feature extration from imagery is an important and long-standing problem in remote sensing. In this paper, we report on work using genetic programming to perform feature extraction simultaneously from multispectral and digital elevation model (DEM) data. The tool used is the GENetic Imagery Exploitation (GENIE) software, which produces image-processing software that inherently combines spatial and spectral processing. GENIE is particularly useful in exploratory studies of imagery, such as one often does in combining data from multiple sources. The user trains the software by painting the feature of interest with a simple graphical user interface. GENIE then uses genetic programming techniques to produce an image-processing pipeline. Here, we demonstrate evolution of image processing algorithms that extract a range of land-cover features including towns, grasslands, wild fire burn scars, and several types of forest. We use imagery from the DOE/NNSA Multispectral Thermal Imager (MTI) spacecraft, fused with USGS 1:24000 scale DEM data.

• Research Organization: Los Alamos National Laboratory
• Sponsoring Organization: DOE
• OSTI ID: 976121
• Report Number(s): LA-UR-02-1551
• Country of Publication: United States
• Language: English

Related Subjects

59 BASIC BIOLOGICAL SCIENCES
99 GENERAL AND MISCELLANEOUS
ALGORITHMS
CULTURAL OBJECTS
GENETICS
IMAGE PROCESSING
PROCESSING
PROGRAMMING
RANGELANDS
REMOTE SENSING
SIMULATION