1

I am calculating various gridmetrics using the Lidar processing tools in FUSION. The dataset I am working with contains 154 flight lines of high density data, at a nominal pulse spacing (NPS) of 1.5 meters (full metadata available here). Specifically, I am using the grid metrics command in FUSION to calculate all returns above 2m (column 53). The attached figure shows the problem I have been encountering. You can see there are detectable grid values at the scan line overlap areas. There are even detectable values in the non-overlap scan line areas over lakes (observed as white blob in screenshot).

How should I be interpreting these apparently faulty results--is it a result of the Lidar dataset, or is it some type of processing issue with FUSION?

enter image description here

Improve this question
8
  • 1
    It looks like levelling hasn't been done, or at least not done well. Can you elaborate on returns above 2m, is that 2m from surface (lowest) or 2m above sea level?
    – Michael Stimson
    Jul 22, 2014 at 21:38
  • Returns above 2m refers to 2m above surface.
    – Aaron ♦
    Jul 22, 2014 at 21:40
  • 1
    Is there something wrong with your surface at the ephemera of each strip? As the laser hits the extents ground points become less frequent in vegetated areas and more interpolation needs to be employed. How does it look with 1m above surface?
    – Michael Stimson
    Jul 22, 2014 at 21:43
  • Many of the FUSION grid metrics based raster data look great (e.g. cubic mean elevation, 1st return cover above mean), although many look like the screenshot (e.g. all count above mode, all count, 1st count above mode...).
    – Aaron ♦
    Jul 22, 2014 at 21:50
  • 1
    Rotating mirrors tend to give better results at the edges as there is some aberration to do with oscillating mirrors accelerating and decelerating. It is a kindness to class the points captured with an oscillating mirror overlapping the adjacent strip as overlap points (class 12 ASPRS) and remove those from further processing, also don't forget High/Low noise classes - assuming of course your data is properly classified.
    – Michael Stimson
    Jul 22, 2014 at 22:17

1 Answer 1

Reset to default
1

The good folks at the Remote Sensing Applications Center (RSAC) noticed from the metadata that the Lidar data specs are insufficient to calculate many of these grid metrics. In particular:

  • NPS is 1.0 – 1.5 pulses/sq m
  • Side Lap (Minimum): 25%
  • Field of View (full): 40 degrees

These three parameters, especially when combined, will likely result in data gaps on the edge of flight lines. RSAC recommend the following specs when using FUSION grid metrics:

  • NPS (4-8 pulses/sq m),
  • A minimum 50% side lap,
  • A narrower FOV (30 degrees or less).

However, the data gaps may not be a problem depending upon how the data are used.

Improve this answer
1
  • @Andre Good points all around--I appreciate your perspective. The AOI is primarily forested and the soil type is very rocky with a heavy clay component.
    – Aaron ♦
    Jul 25, 2014 at 22:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.