Minimum Class Confusion based Transfer for Land Cover Segmentation in Rural and Urban Regions
This work addresses land cover mapping for satellite image analysis, but it is incremental as it applies existing transfer learning methods to new datasets.
The study tackled land cover segmentation in satellite images by comparing transfer learning methods, finding that transfer learning improved segmentation performance by 3.4% MIoU in rural regions and 12.9% MIoU in urban regions.
Transfer Learning methods are widely used in satellite image segmentation problems and improve performance upon classical supervised learning methods. In this study, we present a semantic segmentation method that allows us to make land cover maps by using transfer learning methods. We compare models trained in low-resolution images with insufficient data for the targeted region or zoom level. In order to boost performance on target data we experiment with models trained with unsupervised, semi-supervised and supervised transfer learning approaches, including satellite images from public datasets and other unlabeled sources. According to experimental results, transfer learning improves segmentation performance 3.4% MIoU (Mean Intersection over Union) in rural regions and 12.9% MIoU in urban regions. We observed that transfer learning is more effective when two datasets share a comparable zoom level and are labeled with identical rules; otherwise, semi-supervised learning is more effective by using the data as unlabeled. In addition, experiments showed that HRNet outperformed building segmentation approaches in multi-class segmentation.