0

NDVI values raster computed by using the band 3 and band 4 reflectance values b4-b3/b4+b3. The output raster is of 32bit type formate although the possible ranges that shows in statistics parameter are from -0.35 to +0.65. Are this is due to abnormal values that have been accounted for no data values. If yes then how can i remove those values from the output raster or make them out during the calculation of NDVI in raster calculator.

@Whuber. commented in a post normalize raster, but cannot understand how to do into my case of ration images. there are 5 images that i took for comparison but how do i normalize them to compare easily

Improve this question
0

2 Answers 2

Reset to default
1

You are asking a lot of questions, not just one.

Are this is due to abnormal values that have been accounted for no data values.

NoData values will be excluded from calculations. If the value of 0 is being used to notate NoData, you are going to have bad results. Check the attribute table to see what your data uses (other popular choices are -999 and the like).

the possible ranges that shows in statistics parameter are from -0.35 to +0.65.

The valid range of NDVI values are {-1, 1}. Your value ranges are within the expected range of NDVI values.

NDVI raster 0-255

This is not a good idea, but it is possible using a bit of algebra.

Not a good idea because it is already normalized for comparison between images (that's what the N stands for in NDVI). The range of NDVI values are {-1, 1}. Usually < 0 means not vegetation and > 0 means vegetation. It is likely that you need to calibrate your NDVI to your specific images by comparing various pixel thresholds with manually classified data (vegetation/no vegetation) to somewhere around .1-.2.

Improve this answer
2
  • 3
    +1 Good answers. However, it is often useful to stretch NDVI from 0 - 255, such as with NAIP imagery, in order to store data in unsigned 8-bit integer format rather than float. This can save a lot of disk space and increase processing efficiency.
    – Aaron ♦
    Jun 25, 2015 at 21:51
  • floor((x+1)*127.9999) will give you the 0-255 conversion
    – aaryno
    Jun 25, 2015 at 22:51
0

NDVI raster 0-255

You cannot represent NDVI values to some wide grey scale 0-255 because that doesn't make any sense.

NDVI only shows areas with vegetation cover and those that are not. Hence, geting some ranges to represent some kind of shrubs, built-up areas or water are not actually correct. Many though do try to do so by thresholding but believe I believe thats not scientific but ended up having fun playing around with data.

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.