Gully erosion is a widespread natural hazard. Gully mapping is critical to erosion monitoring and the control of degraded areas. The analysis of high-resolution remote sensing images (HRI) and terrain data mixed with developed object-based methods and field verification has been certified as a good solution for automatic gully mapping. Considering the availability of data, we used only open-source optical images (Google Earth images) to identify gully erosion through image feature modeling based on OBIA (Object-Based Image Analysis) in this paper. A two-end extrusion method using the optimal machine learning algorithm (Light Gradient Boosting Machine (LightGBM)) and eCognition software was applied for the automatic extraction of gullies at a regional scale in the black soil region of Northeast China. Due to the characteristics of optical images and the design of the method, unmanaged gullies and gullies harnessed in non-forest areas were the objects of extraction. Moderate success was achieved in the absence of terrain data. According to independent validation, the true overestimation ranged from 20% to 30% and was mainly caused by land use types with high erosion risks, such as bare land and farm lanes being falsely classified as gullies. An underestimation of less than 40% was adjacent to the correctly extracted gullied areas. The results of extraction in regions with geographical object categories of a low complexity were usually more satisfactory. The overall performance demonstrates that the present method is feasible for gully mapping at a regional scale, with high automation, low cost, and acceptable accuracy.

Remote sensing images with different spatial resolutions have different performance capabilities for gully extraction, so it is very important to study the suitability of different spatial resolutions for this purpose. In this study, part of the black soil area in Northeast China with serious gully erosion was taken as the study area, and Google Earth images with seven spatial resolutions ranging from 0.51 to 32.64 m, commonly used in gully erosion research, were selected as data sources. Combined with auxiliary data, gullies were extracted by visual interpretation. The interpretation results of images of different spatial resolutions were analyzed qualitatively and quantitatively, and the interpretation suitability of images of different spatial resolutions for different types of gullies under different classification systems was emphatically explored. The results indicate that the image with a spatial resolution of 1.02 m has the best performance when not considering the types of gullies. However, the image with a spatial resolution of 2.04 m is the most cost-effective and, therefore, the most suitable for general research. When it is necessary to distinguish the type of gully, the image with a spatial resolution of 0.51 m can be adapted for all situations. However, research on ephemeral gullies is of little practical significance. Therefore, the image with a spatial resolution of 1.02 m is the most universally useful image, being cheaper and easier to obtain. When the spatial resolution is 2.04 m or lower, it is necessary to select the spatial resolution according to the gully type required for practical application. When the spatial resolution is 8.16 or lower, the interpretation of gullies becomes very difficult or even impossible.