Whatever Happened to Data Fusion (Part 1 of 2)

A few years ago the big buzzword in the industry was data fusion, and most times people were referring to the merging of image and LIDAR technologies to provide a more comprehensive dataset. This buzz led to many vendors purchasing systems that contain not only a LIDAR system, but also a camera of some sort. Typically the camera is strapped to the LIDAR unit such that there is a known relationship between the two systems. This fusion of data would then lead to gains in the processing and extraction techniques being used. However, time and time again we hear how many data providers are acquiring LIDAR data and imagery simultaneously, but not doing anything with the imagery data in the case of mobile systems, or only generating orthophotos in the case of airborne. In the meantime, software has evolved over the years to incorporate tools for fusing these datasets. In this first article of the two part series we will look at the tools more applicable to the airborne side of the industry.
A number of years ago Terrasolid implemented a fairly simple, yet robust methodology for associating the RGB values from the co-acquired imagery to the corresponding laser point. The result was the not only a nice visualization of the acquired dataset (see Figure 1), but when coupled with the Classify By Color routine, provide an additional means of distinguishing the classification assignments for individual LIDAR points. Note that the assignment of RGB values to the laser points does not necessitate generating an orthophoto. The big plus to this is that one may fuse the two datasets together very early on in the process to assist in not only automated classification, but probably more importantly, in manual classification review work that can be difficult and very laborious without an orthophoto.

Whatever Happened to Data Fusion (Part 1 of 2)


Darrick Wagg

Darrick Wagg has written 3 articles

Leave a Reply