Learning Aerial Image Segmentation From Online Maps
- 21 July 2017
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 55 (11) , 6054-6068
Abstract
This paper deals with semantic segmentation of high-resolution (aerial) images where a semantic class label is assigned to each pixel via supervised classification as a basis for automatic map generation. Recently, deep convolutional neural networks (CNNs) have shown impressive performance and have quickly become the de-facto standard for semantic segmentation, with the added benefit that task-specific feature design is no longer necessary. However, a major downside of deep learning methods is that they are extremely data hungry, thus aggravating the perennial bottleneck of supervised classification, to obtain enough annotated training data. On the other hand, it has been observed that they are rather robust against noise in the training labels. This opens up the intriguing possibility to avoid annotating huge amounts of training data, and instead train the classifier from existing legacy data or crowd-sourced maps that can exhibit high levels of noise. The question addressed in this paper is: can training with large-scale publicly available labels replace a substantial part of the manual labeling effort and still achieve sufficient performance? Such data will inevitably contain a significant portion of errors, but in return virtually unlimited quantities of it are available in larger parts of the world. We adapt a state-of-the-art CNN architecture for semantic segmentation of buildings and roads in aerial images, and compare its performance when using different training data sets, ranging from manually labeled pixel-accurate ground truth of the same city to automatic training data derived from OpenStreetMap data from distant locations. We report our results that indicate that satisfying performance can be obtained with significantly less manual annotation effort, by exploiting noisy large-scale training data.Keywords
All Related Versions
This publication has 42 references indexed in Scilit:
- Road networks as collections of minimum cost pathsISPRS Journal of Photogrammetry and Remote Sensing, 2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Endmember Extraction Using a Combination of Orthogonal Projection and Genetic AlgorithmIEEE Geoscience and Remote Sensing Letters, 2011
- Morphological Attribute Profiles for the Analysis of Very High Resolution ImagesIEEE Transactions on Geoscience and Remote Sensing, 2010
- Feature Sets and Dimensionality Reduction for Visual Object Detection2010
- Human detection using partial least squares analysis2009
- Feature selection by genetic algorithms in object-based classification of IKONOS imagery for forest mapping in Flanders, BelgiumRemote Sensing of Environment, 2007
- Spatial Metrics and Image Texture for Mapping Urban Land UsePhotogrammetric Engineering & Remote Sensing, 2003
- Backpropagation Applied to Handwritten Zip Code RecognitionNeural Computation, 1989
- Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in positionBiological Cybernetics, 1980