Learning to Detect Fortified Areas

TL;DR

This paper explores machine learning methods, including neural networks and gradient boosting, to classify fortified areas using LiDAR and orthophotos, addressing sensor variability with a novel embedding architecture, achieving over 96% accuracy.

Contribution

It introduces a neural network embedding architecture that normalizes data from different LiDAR sensors, enhancing generalization in fortified area detection.

Findings

Accuracy exceeds 96% on real-world data

AUC score surpasses 0.99

Effective sensor data normalization achieved

Abstract

High resolution data models like grid terrain models made from LiDAR data are a prerequisite for modern day Geographic Information Systems applications. Besides providing the foundation for the very accurate digital terrain models, LiDAR data is also extensively used to classify which parts of the considered surface comprise relevant elements like water, buildings and vegetation. In this paper we consider the problem of classifying which areas of a given surface are fortified by for instance, roads, sidewalks, parking spaces, paved driveways and terraces. We consider using LiDAR data and orthophotos, combined and alone, to show how well the modern machine learning algorithms Gradient Boosted Trees and Convolutional Neural Networks are able to detect fortified areas on large real world data. The LiDAR data features, in particular the intensity feature that measures the signal strength of the return, that we consider in this project are heavily dependent on the actual LiDAR sensor that made the measurement. This is highly problematic, in particular for the generalisation capability of pattern matching algorithms, as this means that data features for test data may be very different from the data the model is trained on. We propose an algorithmic solution to this problem by designing a neural net embedding architecture that transforms data from all the different sensor systems into a new common representation that works as well as if the training data and test data originated from the same sensor. The final algorithm result has an accuracy above 96 percent, and an AUC score above 0.99.