In this paper, a data augmentation method Conditional Residual Deep Convolutional Generative Adversarial Network (CRDCGAN) based on Deep Convolutional Generative Adversarial Network (DCGAN) is proposed to address the problem that the accuracy of existing image classification techniques is too low when classifying small-scale rock images. Firstly, Wasserstein distance is introduced to change the loss function, which makes the training of the network more stable; secondly, conditional information is added, and the network has the ability to generate and discriminate image data with label information; finally, the residual module is added to improve the quality of generated images. The results demonstrate that by applying CRDCGAN to the augmented rock image dataset, the accuracy of the classification model trained on this dataset is as high as 96.38%, which is 13.39% higher than that of the classification model trained on the non-augmented dataset, and 8.56% and 6.27% higher than that of the traditional dataset augmented method and the DCGAN dataset augmentation method, respectively. CRDCGAN expands the rock image dataset, which makes the rock classification model accuracy effectively improved. The data augmentation method was found to be able to change the accuracy of the classification model to a greater extent.

Spatio-temporal changes to the eco-environmental quality index (EQI) and determination of their spatial differentiation characteristics are important bases for land management and ecological environment protection. This study evaluates the changes in EQI and its spatial distribution characteristics with reference to the three dominant functions of land use, namely “production-ecology-living” (PEL), based on the interpretation of land use remote sensing data in 2000, 2010 and 2018. The spatial diversity of ecological environment quality and its driving factors were quantitatively analyzed by gravity center transfer, cold and hot spot analysis, and the GeoDetector model. The results showed that: (1) The transformation of land in Ningxia from 2000 to 2018 mainly manifested by the increase in industrial and mining production land (IMPL), urban living land (ULL) and rural living land (RLL), and the decrease of grassland ecological land (GEL), especially in the north of Ningxia. (2) The ecological environment quality decreased slightly during the research period, but there was an improvement trend in the north. High environment quality values were concentrated in the Liupan Mountain area in the south of Ningxia, while the low values were mainly in the desert areas of Shapotou County and Zhongning County in the west. (3) The interaction between land use intensity and topographic factors led to spatial change in EQI in the research area. Effects of land use intensity are the dominant factor, reflecting the degree of impact of human activities on natural ecosystems. Our results suggest that topographic factors and human disturbances should be fully taken into account in future land and spatial development decisions to minimize human-ecological conflicts.