Russell Ives
New Member
Posts:4  
| 20 Apr 2018 11:48 AM |
|
|
I created a short API to take radiometrically calibrated, top of atmosphere reflectance data from Woldview 2 image and apply the DarkSubtractionCorrection Task to it. Where ever the pixel_state was set to 1, data values to ignore set to 0 in the input image, returns negative values in the resulting output image. How can this be corrected so the result doesn't apply the subtraction to the pixels to ignore?
|
|
|
|
MariM
Veteran Member
Posts:2396 
| 20 Apr 2018 12:27 PM |
|
|
Dark Subtraction is an older Classic tool that has never accepted masks and does not honor the data ignore value. You will need to mask those areas/values after running the process. If you choose to use the Band Minimum in the calculation, I would first calculate that minimum using a mask with the original data and using Calculate Statistics. Then provide this minimum as a user provided value in Dark Subtraction.
|
|
|
|
Russell Ives
New Member
Posts:4
| 20 Apr 2018 02:57 PM |
|
|
Am I to understand then that the new ENVItask for dark subtraction was not updated when it transitioned? Can this be dded to the punch list for the next release?
|
|
|
|
MariM
Veteran Member
Posts:2396
| 23 Apr 2018 08:50 AM |
|
|
Many of the tasks are wrappers about the classic code and in this case, the original tool never supported masks. There is a feature request to add masking/DIV support for Dark Subtraction and I have added your name as a user who would like to see this request implemented.
|
|
|
|