PP-LinkNet: Improving Semantic Segmentation of High Resolution Satellite Imagery with Multi-stage Training
This work addresses the expensive and labor-intensive task of mapping road networks and building footprints for applications like city planning and disaster response, representing an incremental improvement in semantic segmentation for satellite imagery.
The paper tackles the problem of extracting road networks and building footprints from high-resolution satellite imagery by proposing a two-stage transfer learning technique and an improved deep neural network called PP-LinkNet, achieving results such as 78.19% meanIoU on the SpaceNet building footprint dataset and up to 77.11% on road topology metrics.
Road network and building footprint extraction is essential for many applications such as updating maps, traffic regulations, city planning, ride-hailing, disaster response \textit{etc}. Mapping road networks is currently both expensive and labor-intensive. Recently, improvements in image segmentation through the application of deep neural networks has shown promising results in extracting road segments from large scale, high resolution satellite imagery. However, significant challenges remain due to lack of enough labeled training data needed to build models for industry grade applications. In this paper, we propose a two-stage transfer learning technique to improve robustness of semantic segmentation for satellite images that leverages noisy pseudo ground truth masks obtained automatically (without human labor) from crowd-sourced OpenStreetMap (OSM) data. We further propose Pyramid Pooling-LinkNet (PP-LinkNet), an improved deep neural network for segmentation that uses focal loss, poly learning rate, and context module. We demonstrate the strengths of our approach through evaluations done on three popular datasets over two tasks, namely, road extraction and building foot-print detection. Specifically, we obtain 78.19\% meanIoU on SpaceNet building footprint dataset, 67.03\% and 77.11\% on the road topology metric on SpaceNet and DeepGlobe road extraction dataset, respectively.