Simms, Daniel M.2020-10-222020-10-222020-10-20Simms DM. (2020) Fully convolutional neural nets in-the-wild. Remote Sensing Letters, Volume 11, Issue 12, 2020, pp.1080-10892150-704Xhttps://doi.org/10.1080/2150704X.2020.1821120http://dspace.lib.cranfield.ac.uk/handle/1826/15908The ground breaking performance of fully convolutional neural nets (FCNs) for semantic segmentation tasks has yet to be achieved for landcover classification, partly because a lack of suitable training data. Here the FCN8 model is trained and evaluated in real-world conditions, so called in-the-wild, for the classification of opium poppy and cereal crops at very high resolution (1 m). Densely labelled image samples from 74 Ikonos scenes were taken from 3 years of opium cultivation surveys for Helmand Province, Afghanistan. Models were trained using 1 km2 samples, subsampled patches and transfer learning. Overall accuracy was 88% for a FCN8 model transfer-trained on all three years of data and complex features were successfully grouped into distinct field parcels from the training data alone. FCNs can be trained end-to-end using variable sized input images for pixel-level classification that combines the spatial and spectral properties of target objects in a single operation. Transfer learning improves classifier performance and can be used to share information between FCNs, demonstrating their potential to significantly improve land cover classification more generally.enAttribution-NonCommercial 4.0 Internationalhttp://creativecommons.org/licenses/by-nc/4.0/FCN8CNNsOpium PoppyLandcover ClassificationConvnetsSemantic SegmentationArticle28145491