ABSTRACT
Urban planning applications (energy audits, investment, etc.) require an understanding of built infrastructure and its environment, i.e., both low-level, physical features (amount of vegetation, building area and geometry etc.), as well as higher-level concepts such as land use classes (which encode expert understanding of socio-economic end uses). This kind of data is expensive and labor-intensive to obtain, which limits its availability (particularly in developing countries). We analyze patterns in land use in urban neighborhoods using large-scale satellite imagery data (which is available worldwide from third-party providers) and state-of-the-art computer vision techniques based on deep convolutional neural networks. For supervision, given the limited availability of standard benchmarks for remote-sensing data, we obtain ground truth land use class labels carefully sampled from open-source surveys, in particular the Urban Atlas land classification dataset of $20$ land use classes across $~300$ European cities. We use this data to train and compare deep architectures which have recently shown good performance on standard computer vision tasks (image classification and segmentation), including on geospatial data. Furthermore, we show that the deep representations extracted from satellite imagery of urban environments can be used to compare neighborhoods across several cities. We make our dataset available for other machine learning researchers to use for remote-sensing applications.
Supplemental Material
- Saikat Basu, Sangram Ganguly, Supratik Mukhopadhyay, Robert DiBiano, Manohar Karki, and Ramakrishna R. Nemani 2015. DeepSat - A Learning framework for Satellite Imagery. CoRR Vol. abs/1509.03602 (2015).Google Scholar
- Marco Castelluccio, Giovanni Poggi, Carlo Sansone, and Luisa Verdoliva 2015. Land Use Classification in Remote Sensing Images by Convolutional Neural Networks. CoRR Vol. abs/1508.00092 (2015). http://arxiv.org/abs/1508.00092Google Scholar
- Dragos Costea and Marius Leordeanu 2016. Aerial image geolocalization from recognition and matching of roads and intersections. CoRR Vol. abs/1605.08323 (2016). http://arxiv.org/abs/1605.08323Google Scholar
- Marco De Nadai, Radu Laurentiu Vieriu, Gloria Zen, Stefan Dragicevic, Nikhil Naik, Michele Caraviello, Cesar Augusto Hidalgo, Nicu Sebe, and Bruno Lepri. 2016. Are Safer Looking Neighborhoods More Lively?: A Multimodal Investigation into Urban Life Proceedings of the 2016 ACM on Multimedia Conference (MM '16). ACM, New York, NY, USA, 1127--1135. http://dx.doi.org/10.1145/2964284.2964312 Google ScholarDigital Library
- Abhimanyu Dubey, Nikhil Naik, Devi Parikh, Ramesh Raskar, and César A. Hidalgo 2016. Deep Learning the City: Quantifying Urban Perception at a Global Scale. Springer International Publishing, Cham, 196--212. http://dx.doi.org/10.1007/978-3-319-46448-0_12 Google ScholarCross Ref
- Sebastian Grauwin, Stanislav Sobolevsky, Simon Moritz, István Gódor, and Carlo Ratti. 2015. Towards a Comparative Science of Cities: Using Mobile Traffic Records in New York, London, and Hong Kong. Springer International Publishing, Cham, 363--387. http://dx.doi.org/10.1007/978-3-319-11469-9_15 Google ScholarCross Ref
- Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2016. Deep Residual Learning for Image Recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Las Vegas, NV, USA, June 27-30, 2016. 770--778. http://dx.doi.org/10.1109/CVPR.2016.90 Google ScholarCross Ref
- Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2016natexlabb. Identity Mappings in Deep Residual Networks. In ECCV (4) (Lecture Notes in Computer Science), Vol. Vol. 9908. Springer, 630--645.Google Scholar
- Neal Jean, Marshall Burke, Michael Xie, W Matthew Davis, David B Lobell, and Stefano Ermon. 2016. Combining satellite imagery and machine learning to predict poverty. Science, Vol. 353, 6301 (2016), 790--794. Google ScholarCross Ref
- Maxime Lenormand, Miguel Picornell, Oliva G. Cantú-Ros, Thomas Louail, Ricardo Herranz, Marc Barthelemy, Enrique Frías-Martínez, Maxi San Miguel, and José J. Ramasco 2015. Comparing and modelling land use organization in cities. Royal Society Open Science Vol. 2, 12 (2015). http://dx.doi.org/10.1098/rsos.150449 Google ScholarCross Ref
Index Terms
- Using Convolutional Networks and Satellite Imagery to Identify Patterns in Urban Environments at a Large Scale
Recommendations
Critical Analysis of Urban Vegetation Mapping by Satellite Multispectral and Airborne Hyperspectral Imagery
GISTAM 2019: Proceedings of the 5th International Conference on Geographical Information Systems Theory, Applications and ManagementThe monitoring and management of urban vegetation is an important issue nowadays due to the multiple benefits of vegetation for people well-being and for maintaining the balance of ecosystem. In that context, the following study explore to what extent ...
Quantitative mapping of pasture biomass using satellite imagery
A knowledge of the amount of pasture biomass available in farm paddocks is crucial for improving utilization and productivity in the Australian grazing industry. A method to quantitatively map the biomass of annual pastures under grazing has been ...
Monitoring crop phenology with street-level imagery using computer vision
Highlights- Method to automatically collect side looking imagery for agriculture parcels.
- Crop types and phenology obtained from computer vision.
- Demonstration over monthly surveyed totalling 400,000 pictures for 17 crop types.
- Deep ...
AbstractStreet-level imagery holds a significant potential to scale-up in-situ data collection. This is enabled by combining the use of cheap high-quality cameras with recent advances in deep learning compute solutions to derive relevant thematic ...
Comments