Building-change detection underpins many important applications, especially in the military and crisis-management domains. Recent methods used for change detection have shifted towards deep learning, which depends on the quality of its training data. The assembly of large-scale annotated satellite imagery datasets is therefore essential for global building-change surveillance. Existing datasets almost exclusively offer near-nadir viewing angles. This limits the range of changes that can be detected. By offering larger observation ranges, the scroll imaging mode of optical satellites presents an opportunity to overcome this restriction. This paper therefore introduces S2Looking, a building-change-detection dataset that contains large-scale side-looking satellite images captured at various off-nadir angles. The dataset consists of 5000 bitemporal image pairs of rural areas and more than 65,920 annotated instances of changes throughout the world. The dataset can be used to train deep-learning-based change-detection algorithms. It expands upon existing datasets by providing (1) larger viewing angles; (2) large illumination variances; and (3) the added complexity of rural images. To facilitate the use of the dataset, a benchmark task has been established, and preliminary tests suggest that deep-learning algorithms find the dataset significantly more challenging than the closest-competing near-nadir dataset, LEVIR-CD+. S2Looking may therefore promote important advances in existing building-change-detection algorithms.

Understanding the causes of poverty and identifying the transformation characteristics of poverty is the basis for achieving poverty eradication. In order to clarify the availability of construction land for poverty assessment, this paper explores the spatio-temporal synergy between urban built-up areas and poverty transformation in Tibet. The following conclusions are drawn: (1) the built-up areas in Tibetan counties have been growing from 2013 to 2019; (2) the proportion of counties with very low and low levels of relative poverty have decreased significantly, and the overall spatial characteristics of poverty are “high in the center and low in the surroundings”; (3) the overall coupling-coordination level between the built-up areas and the relative poverty level is gradually improving from the initial antagonism, and the relative-poverty index shows a significant negative correlation with coupling coordination (correlation coefficient of −0.63); and (4) the built-up area has a strong explanatory power for the spatial distribution of regional relative-poverty transfer compared to temperature, precipitation, elevation, and slope. The results of the study prove that the built-up area cannot be directly used as an indicator factor when constructing the multidimensional relative-poverty model and, instead, should use urban built-up areas by region to participate in poverty-estimation models based on regional economic development.

As resources are depleted, resource-based cities face unique challenges in the process of socio-economic development. We constructed a multidimensional socio-economic development level model by adopting Entropy Value Method, Analytical Hierarchy Process, time series weighting method, and Game Theory approach for the data of 10 indicators in 4 dimensions of 115 resource-based cities in China from 2004 to 2019 to explore the spatial and temporal divergence characteristics of multidimensional socio-economic development level and the driving mechanism of its pattern of evolution. The results show that: (1) the overall socio-economic development level of resource-based cities has improved from 2004 to 2019, but the overall level is low. Large differences exist in the spatial distribution of socio-economic development levels between cities with more significant regional spatial aggregation characteristics. (2) Secondary industry, tertiary industry, retail trade goods sales, urban construction land area, and total freight transport have a significant positive impact on socio-economic development; the correlation coefficient between the number of schools and the socio-economic development level index is negative. (3) Retail trade merchandise sales contribute the most to the Gini coefficient, where the percentage of secondary industry and urban construction land area have a higher cumulative contribution to growing cities (55.02%), the percentage of secondary industry has the lowest contribution to regenerating cities (10.94%), and the percentage of tertiary industry has an increasing contribution to declining cities year by year. Based on the above findings, some specific suggestions are provided to provide reference for resource-based city development planning.