As a multispectral camera , parrot sequoia has green/red/nir/rededge four bands, in order to calculate NDVI value from bands, reflectance of each band is needed. Now I can calculate irradiance images from following equation, how to calculate reflectance from irradiance images?
Normally, to deal with the problem described above:https://forum.developer.parrot.com/t/reflectance-estimation/5597
sun and equating
sequoia, then solving for
R we find:
R = (k' sequoia) / (k sunshine) or
R = K sequoia / sunshine.
One way of going around that is by using a calibration image with a target of known reflectance
Rref, often a spectralon panel. Which gives you
K = Rref sunshine / sequoia. This is known as reflectance target/tarp/panel calibration.
but pix4d could calculate reflectance map without panel calibration, how could you do this?
The radiometric calibration and correction are applied during Step 3. DSM, Orthomosaic and Index and can be set in the section Radiometric Processing and Calibration of the window Processing Options. The radiometric corrections allow the users to correct the image reflectance, taking the illumination and sensor influence into consideration.
The 3 types of radiometric corrections can be applied:
- Camera corrections: applied to the parameters that are written in the EXIF metadata and relate to the camera (vignetting, dark current, ISO, etc…), more info here.
- Sun irradiance corrections: take into account the information written in the XMP.Camera.Irradiance EXIF tag, for more information about camera XMP specification, please have a look at this article. This option is available when a sunshine sensor has been used during the image acquisition.
- Sun angle corrections: compensate for the directionality of the sun reflections. This correction can only be applied if we have a precise orientation of the sunshine sensor. For example, we can apply this correction with the eBee drone since the orientation between the camera and the sunshine sensor is fixed and known.
On top of that, you can use a radiometric calibration target to calibrate your images. The calibration target enables to have an absolute reference, which would allow you to get absolute reflectance values and make it possible to compare data coming from several cameras. For more information, please have a look at this article.
Thank you for the reference materials. I wanted to ask, when we using Pix4D to perform radiometric calibration with the radiometric calibration targets, will any empirical calibration equations be generated in the metadata for users to analyze? If not, is there a way for users to obtain the parameter values used to calibrate the imagery from a DN to reflectance? Thank you for your time.
Pix4D does not generate any empirical equation as output. The materials from @Yaoyao might be useful for you. You could also measure the reflectance value after calibration and irradiance value before calibration for some pixels and compare them for each band.