Our dedicated support team is ready to assist you with any queries or issues you might have. Check out our FAQs or contact us directly at [email protected]. Your satisfaction is our priority.
Flash Point Classifier also has extensive in-app documentation to help guide you through your lidar classification workflow. You can read the user manual when you download and install the software.
Currently, points can be categorized as low/high noise, ground, building (roof and facade), power lines, poles, low/medium/high vegetation, paved surfaces (roads and parking lots), and water.
Yes, running FPC does require a valid license for the software, which you can purchase from here. Licenses are perpetual, and a valid license grants you the ability to use FPC for as long as you operate your computer. In addition to the license, the purchase of a license also comes with a 365-day maintenance & support term.
Yes, licenses for Flash Point Classifier are perpetual. Once you purchase a license, you can use the software indefinitely, as long as you comply with the terms of the license agreement. You can also transfer your license to another computer if needed, so long as your Maintenance and Support (M&S) term is valid. M&S terms should be renewed annually to maintain access to software updates and support and the ability to transfer licenses.
A valid M&S term grants the user the ability to run any updates to the software during the term. It also allows the user to contact WG Inc. ([email protected]) with queries about installation and usage of the software. A valid M&S term is also required to transfer your license to another computer. That is, your M&S term grants you the ability to use the floating license model; you may use FPC on any computer but cannot run it on more than one computer simultaneously. When your M&S term expires, you can purchase a new extended M&S term for your license from here.
Yes, anyone can download the FPC software from our site and start using it for evaluation purposes. Without a valid license, the software will automatically add a random error to the output classified point clouds; the magnitude of error increases each day until you register a license for the software. A message will be issued to the user warning them that they are operating on a trial license and that random error is being added to the output.
Yes, part of the FPC workflow allows users to selectively remove specified point classes from data sets.
Yes, Flash Point Classifier can process multiple input files in parallel, leveraging multi-core processors to improve performance. Give all those CPU cores some exercise!
No, at the moment, is not possible to change the order of individual classifiers in a classification workflow; you can only turn individual classifiers on/off. This is because certain classifiers require that other classifiers have been applied earlier. For example, each of the building, powerline, vegetation, water, and road classifiers require that ground points have already been labelled.
FPC can read LAS and LAZ formats. User may also input e57 point clouds and the software will automatically convert these files into a LAS file, which is then used for classification.
FPC will write LAS, LAZ and PLY files. For LAS/LAZ file output, the user may specify the version of the LAS specification (*View Application Settings* -> *Default output LAS format*). Output files can optionally contain color data (RGB values) associated with each class category (see *Output Files* -> *Colorize points by class* in the *Classification Workflows* panel). The user may also optionally choose to calculate and associate point normal vectors when the output is a PLY file (see *Output Files* settings in the *Classification Workflows* panel).
Currently, FPC is available on Microsoft Windows and MacOS. We do not support Linux based operating systems at present.
Registering your license requires Internet access, but once registered, you can use Flash Point Classifier offline.
No, all processing is done locally on your machine. You do not need to upload your data to a remote server.
FPC has been designed with airborne lidar, including UAV-based lidar systems and platforms, in mind. While it is possible to apply it to point clouds collected with ground-based lidar systems, it may not work as well on these data sets.
No, the FPC classifiers do not rely on machine-learning based or pre-trained deep-learning based classifiers. These approaches offer certain advantages in lidar point classification applications, most notably, they have been shown to frequently provide accurate classifications with minimal user intervention required in the way of adjusting parameters. However, this can be viewed as an disadvantage of ML and AI based approach as well. As with all classification applications, there are times when advanced ML and AI based methods will perform poorly on certain data sets. Given the black-box nature of these models, it is impossible when this happens to improve the performance of the classification model because the user does not have the ability to adjust the internal parameters. With FPC, you can always experiment with adjusting each classifier's parameter values until you find an optimally performing parameter set. Remember however, all of these classification models, including deep-learning AI models, are imperfect and some manual classification is sometimes necessary, depending on the application and the accuracy requirements of the project.
