The goal of the present study is to demonstrate that high-poverty counties and robust classification features can be identified by machine learning approaches using only DMSP/OLS night-time light imagery. To accomplish this goal, a total of 96 high-poverty and 96 non-poverty counties were classified using 15 statistical and spatial features extracted from night-time light imagery in China in 2010 formed a training set for identifying high-poverty counties. Seven machine learning approaches were adopted to classify high-poverty counties, and five feature importance measures were used to select robust features. The resulting metrics, including the user’s (>63%), producer’s (>66%) and overall (>82%) accuracies of the poor county identification (probability of poverty greater than 0.6), show that the seven machine learning approaches used in this paper exhibit good performance, although some differences exist among the approaches. The order of feature importance reveals that the relative importance of each feature differs among the models; however, the important features remain consistent. The nine most important features ranked in each approach are relatively robust for poverty identification at the county level. Both spatial feature and statistical features calculated in part from the central tendency, degree of dispersion, and the distribution of the night-time light data were identified as indispensable robust features in all the approaches, indicating that the complex social phenomenon of poverty requires analysis from different aspects. Previous studies that utilized primarily night-time light imagery applied single features related to the central tendency or the distribution features of the imagery; this study provides a new method and can act as a reference for feature selection and identification of high-poverty counties using night-time light imagery and has potential applications across several scientific domains.

Population spatialization reveals the distribution and quantity of the population in geographic space with gridded population maps. Fine-scale population spatialization is essential for urbanization and disaster prevention. Previous approaches have used remotely sensed imagery to disaggregate census data, but this approach has limitations. For example, large-scale population censuses cannot be conducted in underdeveloped countries or regions, and remote sensing data lack semantic information indicating the different human activities occurring in a precise geographic location. Geospatial big data and machine learning provide new fine-scale population distribution mapping methods. In this paper, 30 features are extracted using easily accessible multisource geographic data. Then, a building-scale population estimation model is trained by a random forest (RF) regression algorithm. The results show that 91% of the buildings in Lin’an District have absolute error values of less than six compared with the actual population data. In a comparison with a multiple linear (ML) regression model, the mean absolute errors of the RF and ML models are 2.52 and 3.21, respectively, the root mean squared errors are 8.2 and 9.8, and the R² values are 0.44 and 0.18. The RF model performs better at building-scale population estimation using easily accessible multisource geographic data. Future work will improve the model accuracy in densely populated areas.

Enriching Asian perspectives on the rapid identification of urban poverty and its implications for housing inequality, this paper contributes empirical evidence about the utility of image features derived from high-resolution satellite imagery and machine learning approaches for identifying urban poverty in China at the community level. For the case of the Jiangxia District and Huangpi District of Wuhan, image features, including perimeter, line segment detector (LSD), Hough transform, gray-level cooccurrence matrix (GLCM), histogram of oriented gradients (HoG), and local binary patterns (LBP), are calculated, and four machine learning approaches and 25 variables are applied to identify urban poverty and relatively important variables. The results show that image features and machine learning approaches can be used to identify urban poverty with the best model performance with a coefficient of determination, R², of 0.5341 and 0.5324 for Jiangxia and Huangpi, respectively, although some differences exist among the approaches and study areas. The importance of each variable differs for each approach and study area; however, the relatively important variables are similar. In particular, four variables achieved relatively satisfactory prediction results for all models and presented obvious differences in varying communities with different poverty levels. Housing inequality within low-income neighborhoods, which is a response to gaps in wealth, income, and housing affordability among social groups, is an important manifestation of urban poverty. Policy makers can implement these findings to rapidly identify urban poverty, and the findings have potential applications for addressing housing inequality and proving the rationality of urban planning for building a sustainable society.

Thanks to the recent development of laser scanner hardware and the technology of dense image matching (DIM), the acquisition of three-dimensional (3D) point cloud data has become increasingly convenient. However, how to effectively combine 3D point cloud data and images to realize accurate building change detection is still a hotspot in the field of photogrammetry and remote sensing. Therefore, with the bi-temporal aerial images and point cloud data obtained by airborne laser scanner (ALS) or DIM as the data source, a novel building change detection method combining co-segmentation and superpixel-based graph cuts is proposed in this paper. In this method, the bi-temporal point cloud data are firstly combined to achieve a co-segmentation to obtain bi-temporal superpixels with the simple linear iterative clustering (SLIC) algorithm. Secondly, for each period of aerial images, semantic segmentation based on a deep convolutional neural network is used to extract building areas, and this is the basis for subsequent superpixel feature extraction. Again, with the bi-temporal superpixel as the processing unit, a graph-cuts-based building change detection algorithm is proposed to extract the changed buildings. In this step, the building change detection problem is modeled as two binary classifications, and acquisition of each period’s changed buildings is a binary classification, in which the changed building is regarded as foreground and the other area as background. Then, the graph cuts algorithm is used to obtain the optimal solution. Next, by combining the bi-temporal changed buildings and digital surface models (DSMs), these changed buildings are further classified as “newly built,” “taller,” “demolished”, and “lower”. Finally, two typical datasets composed of bi-temporal aerial images and point cloud data obtained by ALS or DIM are used to validate the proposed method, and the experiments demonstrate the effectiveness and generality of the proposed algorithm.