samrae Posted June 22, 2012 Report Share Posted June 22, 2012 Hi, I've used 2011 SPOT-5 data to create an NDVI using ERDAS IMAGINE (the NDVI function available on the Raster toolbar). Ive got the output but am slightly concerned with the values. Its a rural area that consists of pine plantation(stretching from the south-west) , tea plantation (south-east corner) and natural forest (amongst plantation), but the highest NDVI value is 0.206. The forest, pine plantation and water seem to read this value or lower (to 0.128). 1. Is water meant to measure similar values to trees? 2. Is it not possible to see the difference between pine plantation and forest? Looking at the NIR composite the two classes seem more distinguishable 3. Are my values strange? Am I missing any vital pre-processing steps? Thanks Sam p.s. I was going to upload an image but dont have a place to upload it to. Quote Link to comment Share on other sites More sharing options...
nomlas Posted June 23, 2012 Report Share Posted June 23, 2012 Hi You have to check the band sequence of your spot image, if you imported using dimap format then the band sequence is 3214, and it should be 1234 ,if you band sequence is correct than the vegetation classes should display a higher NDVI value Quote Link to comment Share on other sites More sharing options...
rahmansunbeam Posted June 24, 2012 Report Share Posted June 24, 2012 Hi You have to check the band sequence of your spot image, if you imported using dimap format then the band sequence is 3214, and it should be 1234 ,if you band sequence is correct than the vegetation classes should display a higher NDVI value Just in time! I thought it was only me, but now it seems global! The layer staking utility of Erdas seems buggy in some sense. It cannot identify which layer to place where in the img, whereas ENVI places them quite accordingly. Couples of months ago this 'bug' ate my head after doing a layer staking job on GeoEye 1. The layers staked alphabetically (BGIR) and did 'wonderful job' on ndvi and so on. What a crap! Integraph should work on it. @ samrae, the ndvi should be higher in dense veg and slope down toward non-veg, IF the pre-processing is done well. Have a doubt? try the ndvi option of arcgis 10 to quick check the outputs. Still struggling? try the Model Maker of any of two to build a custom model. Try these on the raw layer TIFFs from scratch, you can't miss it! :wink: 1 Quote Link to comment Share on other sites More sharing options...
samrae Posted June 24, 2012 Author Report Share Posted June 24, 2012 @nomlas and @rahmansunbeam - thank you, you were right! The bands were 3214. I've changed it to 1234 - the NDVI output looks much better now, and the water values are low as they should be! Quote Link to comment Share on other sites More sharing options...
kalymairo Posted July 1, 2012 Report Share Posted July 1, 2012 Any one has an idea witch Indices could I use to estimate SOC of soil? Quote Link to comment Share on other sites More sharing options...
hariasa Posted July 2, 2012 Report Share Posted July 2, 2012 I really don't like the ERDAS vegetation index calculator thingy. So what I did instead is program a python script to calculate all the indices i need automatically from multiband optical imagery. Quote Link to comment Share on other sites More sharing options...
3dbu Posted July 2, 2012 Report Share Posted July 2, 2012 of my experience man is good to say that no all time when you made a NDVI you have all data like you desire , is necessary make and edition and analyst and check if all class or NDVI value are right according reality...you should make analyst of class of soil in the site study ...regards Quote Link to comment Share on other sites More sharing options...
gisguy Posted July 24, 2012 Report Share Posted July 24, 2012 (edited) good info Edited July 24, 2012 by gisguy Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.