Digital Globe Worldview 2/3 imagery and ARCSI

[Morgan Bond]

I have inherited some Worldview 2 and 3 imagery that I would like to calculate surface reflectance with. However, I get an error in ARCSI that I have a LV3D product and not a LV2A product. I messaged Pete, and it was his understanding that the LV3D imagery was already corrected to surface reflectance, and thus ARCSI adjustments would be unnecessary. However, I think from Digital Globe documentation that LV3D means that the imagery has been "3" orthorectified, and "D" a 1:12000 DEM was used. I think what they provide are DN's and not corrected values (each band has an integer value ranging from 0-2047), but I am not sure of this. In the peer reviewed literature, others with LV3D imagery seem to still need to perform atmospheric corrections, etc.  

From table 6.2 in this doc: https://dg-cms-uploads-production.s3.amazonaws.com/uploads/document/file/106/ISD_External.pdf

which I have attached.

Has anyone else worked with Worldview imagery and ARCSI and run into this problem? 

[Daniel Clewley]

Hi Morgan,

I’m not familiar enough with WorldView data to comment on if but if you believe that the same atmospheric correction applied to LV2A data in ARCSI should be applied to LV3D data you will need to edit ARCSI to accept this file type. For WV3 you need to change:

arcsllib/arcsisensorworldview3.py Line 180

if self.inProdLev != ‘LV2A':

to something like:

if self.inProdLev != ‘LV2A’ or self.inProdLev != ‘LV3D':

You might also need to make some updates to the metadata reading. Running arcsi.py with `--debug` should help you locate any problems easier.

Once you updated the code and tested it works please submit a pull request so the changes can be incorporated into the main ARCSI version.

If you can provide a sample of the file (this can be off list) either myself or Pete might be able to look at it but can’t promise how soon this will be. Making the changes yourself and submitting a pull request to be reviewed would be faster.

Thanks,

Dan

--
You received this message because you are subscribed to the Google Groups "RSGISLib Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rsgislib-suppo...@googlegroups.com.
To post to this group, send email to rsgislib...@googlegroups.com.
Visit this group at https://groups.google.com/group/rsgislib-support.
For more options, visit https://groups.google.com/d/optout.
<ISD_External.pdf>

[Daniel Clewley]

(link removed from chain and adding RSGISLib support back)

Not yet - have you had a look at editing the code yet?

Dan

On 25 Apr 2018, at 17:12, Morgan Bond - NOAA Affiliate <morga...@noaa.gov> wrote:

Hi Dan,

Just wondering if you had a chance to look at the worldview imagery I uploaded to my google drive? Thanks,

Morgan

On Fri, Apr 20, 2018 at 11:52 AM, Daniel Clewley <daniel....@gmail.com> wrote:

Hi Morgan,

Thanks for providing a link to an image subset. I’ll look when I get chance but don’t have much time to spend on it. If you need something working quickly I would recommend either trying to update ARCSI yourself or finding someone who can do it for you.

Apart from links to data which can’t be made publicly available it would be best to keep discussion on the list so others can pitch in if needed.

Thanks,

Dan

On 20 Apr 2018, at 19:43, Morgan Bond - NOAA Affiliate <morga...@noaa.gov> wrote:

Dan,

I am fairly new to remote sensing, so I am no expert on this. I wish there were better documentation on the Worldview "levels". Sadly, this imagery has been passed to me by a colleague, and I cannot re-order LV2A files. I have read a few published papers that use the LV3D imagery and they all do additional corrections for surface reflectance (if that is what they need). 

For example:

"The raw WV-2 satellite images were registered in UTM-50S and converted to geographical longitude and latitude WGS84 data. The satellite images were delivered as a level LV3D product to ensure that they were sensor corrected, radiometrically corrected, and orthorectified (DigitalGlobe Staff, 2013; Eckert, 2012; Liang, Li, and Wang, 2012). Two steps are involved in the conversion of WV-2 data to reflectance values. The first step involves converting the digital number (DN) in the range from 0 to 255 into radiance values, and the second step converts the radiance in watts per square meter per steradian per micron (W m 2 sr 1 lm 1) into reflectance values. This process requires relevant input information, such as the distance between the sun and the earth (in astronomical units), the day of the year (Julian date), the mean solar exoatmospheric irradiance, and the solar zenith angle."

And another, also using LV3D imagery:

"The satellite data were delivered in product level LV3D, meaning that the data had been sensor-corrected, radiometrically corrected, and orthorectified. Digital numbers were converted to top-of-atmosphere reflectance using the absolute radiometric calibration factors and effective bandwidths for each band, according to the formulas and directions provided by the satellite data provider DigitalGlobe [30]. The WorldView-2 data were then atmospherically corrected in order to reduce haze as well as other atmospheric and solar illumination influences. "

On Fri, Apr 20, 2018 at 8:41 AM, Morgan Bond - NOAA Affiliate <morga...@noaa.gov> wrote:

Dan,

Thanks for the insights. My images are large tiled 1200 km2 strips, but I have a small test image that I have been using that should have the same attributes (it is just much smaller). I have it in a google drive here if you can access that:

LINK REMOVED

I provided Pete with the same file, but I think he is too busy/remote with the class to download it and have a look. I would appreciate any thoughts you have. Although I do not think this file has been atmospherically corrected, it has been orthorectified (unlike the LV2A files). The DEM I was going to use is slightly higher resolution than the DEM that was used for the orthorectification. I don't know if or how this will pose a problem. Thanks,

Morgan  

Hi Morgan,

To unsubscribe from this group and stop receiving emails from it, send an email to rsgislib-support+unsubscribe@googlegroups.com.
To post to this group, send email to rsgislib-support@googlegroups.com.

Visit this group at https://groups.google.com/group/rsgislib-support.
For more options, visit https://groups.google.com/d/optout.
<ISD_External.pdf>

--

Morgan Bond, Fishery Biologist

NOAA-NWFSC

Contractor with Ocean Associates, Inc. 

Fish Ecology Division 

206-860-3428

--

Morgan Bond, Fishery Biologist

NOAA-NWFSC

Contractor with Ocean Associates, Inc. 

Fish Ecology Division 

206-860-3428

--

Morgan Bond, Fishery Biologist

NOAA-NWFSC

Contractor with Ocean Associates, Inc. 

Fish Ecology Division 

206-860-3428

[Morgan Bond]

Hi Dan,

I have not tried editing the code. To see if I could get things to run, I edited my xml file to indicate LV2A and immediately hit another error: "Expecting image to be projected as UTM WGS84". It looks like my Imagery is projected as "Geographic (Lat/Long)", with the datum "WGS84". I thought GDAL would be able to deal with this, as the imagery is brought in, but maybe I need to reproject it first? 

Dan

Hi Morgan,

To unsubscribe from this group and stop receiving emails from it, send an email to rsgislib-suppo...@googlegroups.com.
To post to this group, send email to rsgislib...@googlegroups.com.

Visit this group at https://groups.google.com/group/rsgislib-support.
For more options, visit https://groups.google.com/d/optout.
<ISD_External.pdf>

--

Morgan Bond, Fishery Biologist

NOAA-NWFSC

Contractor with Ocean Associates, Inc. 

Fish Ecology Division 

206-860-3428

--

Morgan Bond, Fishery Biologist

NOAA-NWFSC

Contractor with Ocean Associates, Inc. 

Fish Ecology Division 

206-860-3428