I am completing a remote sensing paper and am doing most of the work in PCI Geomatica, where I have 2 separate files. One from running a texture analysis algorithm containing 6 band layers from a satellite image. The other set from running an unsupervised classification on the same image with the original 6 band layers, as well as the texture analysis bands amalgamated to aid the classification, classification results which are two bands (input, output channels) giving me a total of 16 band layers. The classification itself is aggregated into 3 classes. My desire is to compare the correlation between the results of the classification (3 classes) against the results of the texture analysis quantitatively.
I am aware the GIS does not like pix files. I am considering converting both to TIFF files. My query is do I need to take any additional steps to run zonal statistics on these two image files correctly?