Conference Agenda

The pace of urbanization has been unprecedented. Rapid urbanization poses significant social and environmental challenges, including sprawling informal settlements, increased pollution and urban heat island effects, loss of biodiversity and ecosystem services, and increased vulnerability to disasters. Therefore, timely and accurate information on urban change patterns is crucial to support sustainable and resilient urban planning and monitoring of the UN 2030 Urban Sustainable Development Goal (SDG). The overall objective of this project is to develop innovative, robust and globally applicable methods, based on Earth observation (EO) big data and AI, for urban land cover mapping and urbanization monitoring.

Using ESA Sentinel-1 SAR, Sentinel-2 MSI and Chinese GaoFen-1 images, the EO-AI4Urban team has developed varous deep learning-based methods for urban mapping and change detection. For urban mapping, a novel Domain Adaptation (DA) approach using semi-supervised learning has been developed for urban extraction. The DA approach jointly exploits Sentinel-1 SAR and Sentinel-2 MSI data to improve across-region generalization for built-up area mapping [1]. For urban change detection, several novel methods have been developed including a dual-stream U-Net [2] and a Siamese Difference Dual-Task network with Multi-Modal Consistency Regularization [3]. Further, a high-resolution feature difference attention network (HDANet) is proposed to detect changes using the Siamese network structure [4]. Another novel procedure was designed to search for built-up changing patterns with the joint use of temporal and spatial properties, starting from high-frequency SAR time series. The methodology has been tested on the city of Wuhan and considering a SAR series from March 2018 to March 2021 [5] [6]. Additionally, a novel automatic deep learning-based binary scene-level change detection method that trains a Scene Change Detection Triplet Network (SCDTN) using the automatically selected scene-level training samples was proposed [8]. A machine learning method was also developed using Landsat time series, to map built-up areas and to analyze changes during 2000 to 2020 [9]. Finally, to identify similar urban areas quickly and to reduce the cost of manually labeled data, a multisource data reconstruction-based deep unsupervised hashing method was proposed for unisource remote sensing image retrieval, called MrHash, which consists of a label generation network and a deep hashing network [9].

Experiments conducted on a test set comprised of sixty representative sites across the world showed that the proposed DA approach achieves strong improvements upon fully supervised learning. The fusion DA offers great potential to be adapted to produce easily updateable human settlements maps at a global scale [1]. Using the OSCD dataset, the results showed that the dual-stream U-Net outperformed other U-Net-based approaches together with SAR or optical data and feature level fusion of SAR and optical data [2]. Using bi-temporal SAR and MSI image pairs as input, the Siamese Difference Dual-Task network with Multi-Modal Consistency Regularization have been tested in the 60 sites of the SpaceNet7 dataset. The method achieved higher F1 score than that of several supervised models when applied to the sites located outside of the source domain [3]. Using several public building change detection datasets, the experimental results showed that the HDANet can achieve a high building change detection accuracy, compared with the current mainstream methods, with public building change detection datasets [4]. Using Landsat time series, the results show that machine learning method could extract built-up areas effectively. To analyze urbanization in 13 cities in the Beijing–Tianjin–Hebei region, SDG indicator 11.3.1, the ratio of land consumption rate to population growth rate (LCRPGR) is calculated and the results show that the LCRPGR in Beijing–Tianjin–Hebei region fluctuated significantly. Apart from the megacities of Beijing and Tianjin, after 2010, the LCRPGR values were greater than 2 in all the cities in the region, indicating inefficient urban land use [7]. The results for the scene-level changes between the bi-temporal VHR images showed that the proposed SCDTN method achieved the highest F1 score of 81.85% [8]. Conducting experiments on Sentinel-2 and GF-1 satellite images, the results showed that MrHash yielded the best performance among all methods [9].

References:

[1] Hafner, S., Y. Ban and A. Nascetti, 2022a. Unsupervised Domain Adaptation for Global Urban Extraction Using Sentinel-1 and Sentinel-2 Data. Remote Sensing of Environment. Volume 280, 113192.

[2] Hafner, S., A. Nascetti, H. Azizpour and Y. Ban, 2022b. Sentinel-1 and Sentinel- 2 Data Fusion for Urban Change Detection Using a Dual Stream U-Net. IEEE Geoscience and Remote Sensing Letters, vol. 19, pp. 1-5.

[3] Hafner, S., Y. Ban and A. Nascetti, 2023. Multi-Modal Consistency Regular- ization Using Sentinel-1/2 Data for Urban Change Detection. International Journal of Applied Earth Observation and Geoinformation (under review).

[4] Wang, X., J. Du, K. Tan, J. Ding, Z. Liu, C. Pan, and B. Han, 2022. A high-resolution feature difference attention network for the application of building change detection, International Journal of Applied Earth Observation and Geoinformation, Volume 112, 102950.

[5] M. Che, A. Vizziello and P. Gamba, 2022. Spatio-temporal Urban Change Mapping with Time-Series SAR data, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[6] Che, M., A. Vizziello, P. Gamba. 2021. Spatio-temporal Change Mapping with Coherence Time-Series. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[7] Zhou, M., Lu, L., Guo, H., Weng, Q., Cao, S., Zhang, S., & Li, Q. (2021). Urban Sprawl and Changes in Land-Use Efficiency in the Beijing–Tianjin–Hebei Region, China from 2000 to 2020: A Spatiotemporal Analysis Using Earth Observation Data. Remote Sensing, 13(15).

[8] H. Fang, S. Guo, X. Wang, S. Liu, C. Lin and P. Du. 2023. Automatic Urban Scene-Level Binary Change Detection Based on a Novel Sample Selection Approach and Advanced Triplet Neural Network, IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1-18.

[9] Y. Sun, Y. Ye, J. Kang, R. Fernandez-Beltran, Y. Ban, X. Li, B. Zhang, and A. Plaza. 2022 Multisource Data Reconstruction-based Deep Unsupervised Hashing for Unisource Remote Sensing Image Retrieval. IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1-16.

Presenting Authors: Gao, Mingliang and Cartalis, Constantinos
-

The scope of the work is to demonstrate the potential of Earth Observation to support urban security. The work is deployed in a two-fold manner.

At a first stage, the evolution of groundwater flow field and the corresponding response of land subsidence along Yongding River (Beijing section) were analyzed by performing spatio-temporal analysis, time series decomposition, based on the data sets covering traditional hydrogeological data, groundwater observation data, and satellite-based images. Results showed that, at present, ecological water replenishment of Yondding River has no obvious impact on the formation deformation, but the rising groundwater level and differential land subsidence in some regions will pose a great risk to the safety of coastal areas in the future. In addition, the Beijing section of the Yongding River crosses multiple subway lines, and the affected area is close to the Beijing Daxing International Airport. Local groundwater level rising may cause underground facilities damage, and uneven land subsidence may cause surface & underground structure break, as well as the stability of electronic equipment, which affect the safe operation of airports and rail transit.

At a second stage, the dynamics of urban heat risk were analyzed by means of a tool that is based on the use of high-resolution Earth Observation (EO), climate, and socioeconomic data and exploits the potential of machine learning. The tool is developed in the cloud-based Google Earth Engine (GEE) platform that effectively addresses the challenges of big data analysis in studying urban heat risk. Urban heat risk maps are created for Beijing and Athens using clustering algorithms, which group areas with similar characteristics and assign them to different heat risk categories based on the spatiotemporal patterns of the above-mentioned indicators. The results effectively identify vulnerable regions that experience significantly higher heat risk and constitute intracity thermal heat spots. To this end, scientific evidence may be used in support of spatially differentiated resilience plans for climate extremes at the city scale.

Recommendations on the use of Earth Observation for urban security will be provided along with a discussion on other urban challenges that may be addressed accordingly.