By spatially and spectrally enhancing an image, pattern recognition can be performed with human eye. Spectral recognition can be more scientific using a computer system The pixels are sorted based on mathematical criteria. The classification process breaks down into two parts: training and classifying (using a decision rule).

Methodology: Digital Image Analysis for Landuse Classification

Step 1: Gathering/Acquiring Digital image

Images used in this project is Landsat 5 TM (1984 & 2004)

Step 2: a) Selection of Training sets for classification for specific cover types.

b) Establish spectral signature for each cover type.

Step 3: Signature statistics –

Mean Plots – The mean plots shows the separation of all the landcover signature in all channels.

Histograms – It shows the distribution of the signature in multiple bands.

Step 4: Classified Image.

Step 5: Accuracy Assessment.- Overall Accuracy 84%

Ancillary Data: Aerial Photography of Lucas County, Topozone Maps, and Teraserver.

Results: The supervised classification performed on both the Landsat 5 images is 84% accurate. The 2004 image shows sufficient urban growth in south of Lucas county and slight decrease in forested areas.