by Richard Horrocks
With more and more people having access to LiDAR files, a key question is how to maximize their use while extracting all the relevant information. One option is to use Asset Collection, where an element that was scanned and is present on the LiDAR as a Point Cloud can be converted into a Feature with its own properties and attributes.
This type of functionality can be found in Autodesk InfraWorks. In this case, you can use the LiDAR files to create a new model, and define a Point Cloud Terrain to run the Point Cloud Modelling. Point Cloud Modelling will then classify the Point elements from the Point Cloud and convert them into Features.
We went through this process and identified some tips to get a better result and make better use of your time.
Preparing Your LiDAR Files
Before starting this process, you have to use Autodesk ReCap to create a project where you can load the LiDAR files and clean them up. At the exact moment you create the ReCap project, you must assign a coordinate system – if you don’t do it in this step, you won’t be able to do it later and you won’t be able to properly geo-reference your model in InfraWorks.
Keep in mind also that when you load your files into a ReCap Project (in rcp/rcs format) it will generate larger files with the 3.5 ratio.
Using Point Cloud Terrain and Point Cloud Modelling in InfraWorks
In InfraWorks, after creating your model and adding your ReCap Project from Data Sources > Point Cloud, run the command Point Cloud Terrain, to create the Terrain and a raster. This could take more than an hour, so if your ReCap model is too big, you might instead consider using several small ReCap models, and processing them in blocks.
Next, you’ll run Point Cloud Modelling to classify the Points (this also can take more than an hour). Keep in mind that if your Points in the LiDAR have good density, you will get better results.
Before starting Point Cloud Modelling, ensure that your library of Point Features is complete with all the Features that you need; for instance, if you have just one traffic light or street sign, all the Features will be the same. You can change them later, but it’s better if you do it in this stage when you are looking in the Point Cloud.
When the Point Cloud Modelling process ends it will offer you a dialog box from where you can browse each of the detected Points to review and place the right feature. When a feature is not detected, a yellow box is place on the model, at which point you can decide which element should be there. Sometimes many of these yellow boxes are placed because of offset Points; in this case, it’s easier and faster to go directly to them and delete the ones you know are not representing any asset. Then you can continue with the remaining Points from the dialog box.
Because you might consider exporting these Points at the end of this process, you have to ensure that when a Feature is placed, you don’t move the anchor Point – if you do, you will move the X and Y that you want to preserve. This is worth mentioning because you will find that some of the Features have the anchor Point in the center of the selection block, and not where the Feature visually touches the ground, which won’t look good in the model even though it is placed correctly.
When you finish this process, you should be able to export the Features with the attributes and load them into another application.
Need More Guidance?
These few tips could make your first experience working with Point Cloud Modelling a little smoother, but if you would like some more guidance through this process, contact us today. If you’re a current AIMS user thinking about switching to InfraWorks, check out our recent article.