MicaSense RedEdge processing to reflectance values without success

[Payne]

Dear all,

I tried to use the following tutorial to process MicaSense sensor data in Agisoft Metashape.

https://agisoft.freshdesk.com/support/solutions/articles/31000148780-micasense-rededge-mx-processing-workflow-including-reflectance-calibration-in-agisoft-metashape-pro

Regrettably the values in the orthomosaic are DN values and not reflectance values. I spend now hours to find a solution, without any success.
Also Appendix C (Controlling reflectance calculation) is fulfilled, the “Normalize band sensitivity” option is checked on.

Does anyone knows a solution for this problem?

Best wishes,
Norman Wilke

[Alexey Pasumansky]

Hello Norman,

Have you tried to follow the approach mentioned in the "Exporting results" section:

Metashape Professional is performing the reflectance calibration operation according to MicaSense recommendations. So the values in the output bands would sill be 16 bit integer values like the input values, but 100% reflectance for each band would correspond to the middle of the available range, i.e. to 32768 value. In case it is necessary to export the reflectance normalized to 0 - 1 range, then it is required to create Output bands in the Raster Calculator dialog and for each one of them input the formular that divides the source value by the normalization factor: B1/32768; B2/32768; B3/32768; B4/32768; B5/32768

[Patricia]

Dear all,
First,  I apologize for my low nevel of english.

I have doubt. I'm following this tutorial https://agisoft.freshdesk.com/support/solutions/articles/31000148780-micasense-rededge-mx-processing-workflow-including-reflectance-calibration-in-agisoft-metashape-pro  where it's said that "Metashape Professional is performing the reflectance calibration operation according to MicaSense recommendations. So the values in the output bands would sill be 16 bit integer values like the input values, but 100% reflectance for each band would correspond to the middle of the available range, i.e. to 32768 value. In case it is necessary to export the reflectance normalized to 0 - 1 range, then it is required to create Output bands in the Raster Calculator dialog and for each one of them input the formular that divides the source value by the normalization factor: B1/32768; B2/32768; B3/32768; B4/32768; B5/32768". However, when I do this, I get reflectance normalized values between 0 and 2. Therefore, would I have to divide by the higher value for each band which in this case is 65536 to get reflectance normalized values between 0 and 1?

Thanks
Patricia

[Alexey Pasumansky]

Hello Patricia,

Do you mean, that after using Tools Menu -> Calibrate Reflectance operation some values of the bands exceed 32768?

[ihb]

Hi Alexey,

I am having the same problem as Patricia. For the MicaSense RedEdge camera, I am trying to also normalize these values between 0 and 1, but I see that a lot of my vegetation values are above 32,768. This is especially true for the NIR band.

Since this camera takes 16-bit imagery photos, I am surprised that reflectance values are higher than 32,768 after reflectance calibration. What could we do to correct this?

Isaac

[Alexey Pasumansky]

Hello Isaac,

Is there any difference in behavior is you use only radiometric panel and/or DLS sensor data for the reflectance calibration?

If the issue is observed regardless of the calibration data used, would it be possible to share the subset of images which can be used to reproduce the problem and send it to support@agisoft.com?

[ihb]

Thanks for the reply Alexey. I actually only used the reflectance calibration panel, and did not use the DLS. I did this because this is the most common method for calibration used in the literature. I will have access to the photos this evening, and will send the email that is requested. Thanks!

[nprokofyev]

Since this camera takes 16-bit imagery photos, I am surprised that reflectance values are higher than 32,768 after reflectance calibration. What could we do to correct this?

Hi Isaac,

I've also seen in my projects that some values exceed reflectance>1 and here is an explaination. The reflectance calibration model assumes that the objects reflects light almost as a diffuse (Lambertian) reflector. Which is not always true, especially under complex lighting conditions such as a combination of direct and diffuse light. In such cases, plants may reflect light in more mirror-like way and in that case reflectance can (mathematically) exceed a value of 1 because it is not diffuse anymore. This is noticable in NIR particularly.

Other source of such effect is a situations where the illumination angle is high. The physics is same here - plants and other objects aren't behave  as a diffuse reflector.