Compute the principal components of multi-channel images.
principal_comp does a principal components analysis of multi-channel images. This is useful for images obtained, e.g., with the thematic mapper of the Landsat satellite. Because the spectral bands are highly correlated it is desirable to transform them to uncorrelated images. This can be used to save storage, since the bands containing little information can be discarded, and with respect to a later classification step.
The operator principal_comp takes a MultichannelImage with NumComp channels as input and transforms it to the output image PCAImage, which also contains NumComp channels, using the principal components analysis. The parameter InfoPerComp contains the relative information content of each output channel.
MultichannelImage (input_object) |
multichannel-image -> object : byte / int1 / int2 / int4 / real |
Multi-channel input image. |
PCAImage (output_object) |
multichannel-image -> object : real |
Multi-channel ouput image. |
NumComp (input_control) |
integer -> integer |
Number of channels. | |
Default value: 3 | |
Suggested values: 2, 3, 4, 5, 6, 7, 8, 9 | |
Range of values: 2 <= NumComp |
InfoPerComp (output_control) |
real-array -> real |
Information content of each output channel. |
The operator principal_comp returns the value TRUE if the parameters are correct. Otherwise an exception is raised.