Forum

Author Topic: Minimum image coverage options  (Read 2015 times)

BrianZimmer

  • Newbie
  • *
  • Posts: 19
    • View Profile
Minimum image coverage options
« on: September 19, 2017, 10:29:12 PM »
Hello All,
I am currently trying to reconstruct imagery of Athabasca glacier from a set of historical air photos. Unfortunately there are only seven images and their overlap is certainly sub-optimal. While I know the resolution of the render will be limited, I am still working to make the best model possible. I have gotten the images to align by adjusting the keypoints to 60,000 and the tie points down to 2,000. The problem is that I have two spots that appear to not have enough coverage (See attached image). Interestingly, the two wedges that are missing in the dense cloud are in areas that have three projections and are some of the better imaged parts of the data set. The geometry of the missing wedges suggests to me there is something systemic to my process that is removing data from those places as opposed to there not being coverage.


Any ideas on how to get those wedges of missing data to render?

Thanks!
Brian

Dave Martin

  • Full Member
  • ***
  • Posts: 170
    • View Profile
Re: Minimum image coverage options
« Reply #1 on: September 19, 2017, 10:52:17 PM »
Brian,

I've never experienced anything like this myself, but you mention that the missing areas have three-fold coverage whereas most of the remainder has perhaps only two-fold coverage. I wonder if issues with the original images mean that, where there are three sources, the results are so divergent that PhotoScan is rejecting the tie points.  In absence of any better suggestions, I wonder if you re-processed and eliminated one of the images there, what the result might be?

Dave

BrianZimmer

  • Newbie
  • *
  • Posts: 19
    • View Profile
Re: Minimum image coverage options
« Reply #2 on: September 20, 2017, 03:05:53 PM »
Hi Dave,
Thanks for responding. I tried disabling each of the three overlapping images to see if it would make a difference, but it did not. I've also tried adjusting the quality of the alignment (highest to low) with no further success. Are there any additional manual alterations I can make during alignment that would prevent Agisoft from dropping those wedge points even if their error is anomalously high for one reason or another?

Thanks!