Processing LiDAR point cloud data, particularly for complex terrain and urban environments, often requires significant manual effort for classification. This manual approach leads to extended project timelines and inconsistent results, hindering efficient data analysis and application.

IRIS is an automated point cloud processing system that leverages machine learning to classify LiDAR data. It offers enhanced classification accuracy for challenging features, reducing the need for manual intervention and thereby decreasing man-hours by 50-70%. The system provides flexible deployment options, allowing for on-premises training and secure data handling. Furthermore, IRIS enables unlimited model fine-tuning, ensuring users can optimize data classification to meet specific project requirements and achieve high-quality outputs efficiently.

Machine learning-driven automated point cloud classification engine.

Achieves up to 98.6% accuracy on bathymetric classes (Seabed, Bathy Water Surface) and 99.5% on topographic classes (High Vegetation, Ground).

Processes up to 30 million points per minute, significantly accelerating data throughput.

Reduces manual classification effort by 50-70%, leading to quicker project turnarounds.

Supports on-premises model training for enhanced data security and control.

Offers unlimited model fine-tuning capabilities for customized classification performance.

Capable of classifying complex urban features such as roads, highways, and dense building structures.

Handles large datasets, processing over 60 billion points in urban mapping projects.

The primary users are geospatial data processing firms, surveying companies, and urban planning agencies that require efficient and accurate classification of LiDAR point cloud data.

Revenue is generated through a flat pricing model with options for on-premises deployment and unlimited data optimization.