Post

2 followers Follow
0
Avatar

mosaic map shrunk, doesnt fit Google Maps basemap

I recently flew two copter missions, something routine for me.    The same copter was used on both missions - a Typhoon-H, but the first mission used a CGO3+ with standard RGB bands, the second, used a CGO3+ outfitted with a ZB2 lens to pick up NIR bands.    The resolution of both cameras, and the lenses for both cameras, were identical.   

 

When i processed the images in MME, using standard workflow, the NIR image stitched very nicely, and was very accurately registered against the GoogleMaps basemap, with a GSD of about 2.5 in/pixel.     

However, the RGB image, while it stitched together very nicely, is clearly about 80% the size it should be, and doesnt register against the GoogleMaps basemap at all...   and still has a GSD of about 2.5in/pixel.    

Why?   both missions flown at same altitude, both with same copter, and only about 30 mins apart.   The coverage/overlap report for both is similar, with 99% of all points Dark Blue w over 10 images/point (each map was composed of about 500 images).    

What would cause a shift in location to be introduced?

This has happened several times in the past couple of months, but this was my first test where I had one where a second map/camera was available and produced a perfect image to compare against.

The correct map is

https://www.mapsmadeeasy.com/maps/detail/b3a1ceedad884c20b95fb315da89bc99/

 

https://www.mapsmadeeasy.com/maps/detail/07f920d67ba14466b513d408cb3f4749/

 

 

 

Doc Vogt

Official comment

Avatar

It looks like you downscaled the resolution quite a bit but didn't likely adjust all of the tags to reflect the data properly. This is likely a large source of the error. 

Also, your overlap is definitely in the dark blue which is usually good but your data collection is all over the place. Lots of images were taken while flying side ways and backwards. This is a problem that is called out in the Data Collection guidelines. 

The biggest issue though is likely the fact that the majority of almost all of the pictures is a large reflect water surface that changes with every image.

All of these things combined basically broke the system's ability to properly create an accurate calibration for the camera and the results is what you are seeing. 

Please use an app to properly manage your data collection efforts and follow the Data Collection guidelines carefully. As long as our guidelines are followed the results will be better. 

Zane
Comment actions Permalink

Please sign in to leave a comment.

1 comment