Three Common Manual Utility LiDAR Classification Errors


LiDAR classification is traditionally a manual process which can be time-consuming and expensive. Manually classifying LiDAR across a 10k square mile network area can take 80k hours and cost nearly $30m – not only is it expensive, but even the most painstaking manual effort is subject to common avoidable errors.

Classification errors are problematic for two reasons – they create expensive, repeat labor that constrains geospatial teams’ capacity and leave utilities with limited visibility into critical operations.

First, geospatial teams have to make re-classification attempts and sometimes need to repeat the capture exercise. Most capture exercises require scheduling expensive resources before the team even starts the data processing phase over again. Secondly, errors leave utilities with deficient data, which impairs their ability to prioritize operations and keep their communities safe.

Raw LiDAR scan to Automatically classified LiDAR

Aside from mounting costs, this dynamic can also delay projects by an average of a month. Geospatial teams end up pouring more resources into existing projects and have less capacity for new projects, while utilities operate for prolonged periods without high-quality data.

Geospatial teams are increasingly embracing automatic LiDAR classification because they can significantly reduce manual error frequency. Automation has long been a popular means of saving time, but until recently, was not sufficiently accurate.

Today’s Al/ML technology can not only help geospatial companies and utility divisions speed classification, but it can also significantly improve accuracy and consistency. Automatic LiDAR classification leverages neural networks that allow rapid pattern- matching across vast databases to accurately identify objects that the naked human eye cannot.

The result: Clean, accurately classified LiDAR scans that geospatial teams can stand behind and utilities trust.

Download the PDF below to learn more about three of the most common manual LiDAR classification errors and how automation can help.

Reach out. We’re very friendly.

Contact us to learn more, schedule a demo, inquire about a pilot project, or discuss other needs.