LiDAR resolution, not to be confused with accuracy, provides the framework for identifying and discerning objects.  Similar to aerial photography, higher resolution LiDAR information affords the opportunity to accurately identify smaller features.  In LiDAR vernacular, there are two closely coupled terms that relate to the resolution of LiDAR data:  Nominal Point Spacing and Density.   

Nominal point spacing refers to 1-dimensional measurement or a point-to-point distance. Recognizing that point distributions are not regularly or evenly spaced, it is quite uncommon to find a single point that has an equal distance to all of the points surrounding it.  Therefore, the nominal spacing would be a generalized value that attempts to quantify this characteristic.

Density is related to point spacing and it is therefore rational that the closer a group of points are to one another, the higher the point density and vice versa.  The point density is normally calculated from the actual data using the “box counting” method.   In “box counting” the area of a square is associated with the total number of LiDAR points inside the square – often presented as points per square meter (ppsm). Point density is a 2 dimensional parameter.

There are numerous variables that affect the resolution of LiDAR data, regardless of whether generated from aerial, terrestrial-mobile, or terrestrial-static systems.  Those with the largest influence include the sensor’s measurement frequency, distance to target, vehicle speed, scan angle and overlap.  Each of the variables can be compensated for by adjusting one or more of the others.  But modification may also have a negative impact on a collection.  For example, in aerial applications, the choice of a helicopter over a fixed wing aircraft provides more flexibility in slower speeds and lower flying heights to increase density.  However, the field of view is reduced when operating at a lower altitude.  For the most part, static scanning and aerial LiDAR operators have it easy compared to Mobile LiDAR.  There are fewer unknowns or influences for which to compensate.   

When performing static scanning, OEM programs, such as Leica’s Cyclone application, allow the user to have direct control over point densities.  Once the area of interest has been isolated, the operator simply keys- in the desired resolution (both horizontal and vertical) after providing or probing a distance to target.  The ability to isolate areas or objects provides flexibility in minimizing scan time, improving densities where required, and ensuring adequate coverage where needed.  The operator has complete control over the resolution of the data and can perform detailed review in real-time to validate results.  While static scanning seldom utilizes a single position to gather all information required, additional scan positions will increase the point density while adding a different perspective.

A portrait of our founder Michael Baker Jr. hangs in our conference room.  Here, a static scan with a horizontal and vertical resolution of 0.005’ at 30’ is shown.  The image represents approximately 1,375,000 points.

With few exceptions, Aerial LiDAR collections are performed to achieve a specified point density or nominal point spacing.  There are commercial-off-the-shelf (COTS) solutions that offer flight planning assistance to streamline the process and evaluate projected results from alternate flight parameters, as well as manufacturer-specific programs, such as Optech’s ALTM-NAV, to generate Aerial LiDAR mission planning parameters.  The variables that impact point densities, including flying height, aircraft speed, scan angle and overlap, are integrated to optimize a collection and achieve the desired resolution.  These densities remain relatively consistent over an Aerial LiDAR collection as aircraft flight patterns typically yield ample time to line-up the next flight line, while adjusting height and ground speed.

Over the next two articles, the following topics, with the requisite real-world examples, will be presented:

·         Collection variables impacting point densities:

o   Vehicle speed versus scanner rates;

o   Angle of incidence and the impact of flat surfaces;

o   Distance to target and effect of rotated sensors;

o   Collection paths and moving obstructions;

·         The minimum required resolution to identify objects; and

·         Tools that utilize points and photography to perform extractions.