Intensity Revisited

  1. I wanted to highlight an important comment on last week’s “intensity” blog post.
  2. Jarlath O’Neil-Dunne from the UVM Spatial Analysis Lab is finding intensity to be a valuable attribute.
  3. In high contrast regions LiDAR intensity can provide important clues for land use mapping.

I received 2 important comments about last week’s post on intensity. I would like to make everyone aware of the information provided by one of In the Scan’s early supporters, Jarlath O’Neil-Dunne. Jarlath is a geospatial analyst at the University of Vermont’s Spatial Analysis Laboratory. He has been providing me with the real world experience and insight that I lack, so I want to thank Jarlath for his support.

As he pointed out, intensity has proven to be a valuable LiDAR attribute, particularly for impervious surface mapping, when used in a high contrast situation such as in the figure below. This is from a post to the lab’s blog where Jarlath explains their procedure in more detail.

They have been using a data fusion approach combining LiDAR data with CIR imagery and Definiens/eCognition software suite. I discussed the power of this software in one of my first blogs. He also notes the use of QT Modeler, another software package that seems to be showing up more frequently.

Thanks for the comments, and please keep them coming.

This entry was posted in Education, Research, Software and tagged , . Bookmark the permalink.

3 Responses to Intensity Revisited

  1. Thank you Gene for the excellent descriptions you provide. A good quote that was passed on to me regarding the use of LiDAR intensity: “one man’s trash is another man’s treasure.”

  2. Martin Flood says:

    Intensity – or more correctly return pulse amplitude – is invaluable in airborne data analysis. Used as a relative filter (since as was pointed out it is not radiometricly calibrated in the instrument) it can significantly improve the results when classifying ground returns. It can also be used in other feature extraction algorithms to improve object identification (roads, buildings and water bodies being some examples). At a minimum it allows for a visual reference via a pseudo grayscale image when no supporting imagery is available. For those people using lidar stereo (lidargrammetry), grayscale stereo models generated from the lidar surface and the intensity data can provide a very effective way to extract 3d vector information from the point data set. I do suspect it is not used as much in mobile or static applications since these instruments seem to have an integrated digital camera much more often than airborne instruments. Perhaps intensity will be depreciated in the ground-based markets, but I don’t know too many airborne operators who would give-up their intensity data without a fight!

Leave a Reply

Your email address will not be published. Required fields are marked *

*