Sloping cropland can serve as one of the most fundamental land resources for agricultural production in mountainous areas. It is often required to reveal the spatial distribution and the driving factors of sloping cultivated land, particularly for the prevention and control of soil and water loss. However, existing slope extraction can rely typically on raster calculations without considering the boundaries of cultivated land patches. The technical challenge has also restricted to accurately identify the overall slope distribution of cultivated land at different scales. In this study, a batch processing algorithm was proposed to calculate the overall slope of cultivated land patches using digital elevation model (DEM) data. The accuracy of slope calculation was then enhanced to integrate the patch boundaries with topographic data. The sloping cultivated land was extracted within the Southwest Alpine Canyon Area of China. Furthermore, kernel density estimation, spatial autocorrelation analysis, and the geographical detector were integrated to systematically reveal the stock characteristics, multi-scale distribution patterns, driving factors, and their interactions on sloping cultivated land. The results indicate that: 1) The slope calculation was better agreement with manual measurements, with a coefficient of determination of 0.9 and a Mean Absolute Error (MAE) of 2.5°. This approach was effectively resolved the conventional algorithms, specifically the slope overestimation in large patches on gentle slopes caused by pixel noise, and the underestimation in small patches on steep slopes after averaging. 2) Southwest Alpine Canyon Region contained a total of 918 000 sloping patches of cultivated land, covering a total area of 12 500 km², which accounted for 72.6% of the total cultivated land in the region. The overall pattern of distribution was presented by “aggregation in the south and sparsity in the north.” High concentrations were observed in the southern counties accounting for 7.4% of the total area. The watersheds were accounted for 13.2% of the total area, indicating the higher population densities and lower elevations. 3) Sloping cultivated land was predominantly distributed in the subtropical climate zone. The primary soil types included the red soil (25.9%), brown soil (16.9%), yellow-brown soil (15.7%), and yellow soil (10.2%). Notably, 43.3% of the land exceeded the prohibited cultivation slope limit, and approximately one-third exceeded 30 degrees, indicating severe potential for soil erosion. These lands were concentrated mainly in the middle mountain zone at elevations of 2 000–3 500 m (56.7%), while the low mountain and river valley zone below 2 000 m (33.2%). 4) At the county scale, the differentiation of sloping cultivated land was primarily controlled by population density, followed by topographic factors. At the watershed scale, the distribution was influenced by the synergistic effects of multiple factors. The influencing factors on the distribution of slop cultivated land were dominated by interactive enhancement. Particularly, the synergistic explanatory power of population density was combined with soil, topography, and hydrothermal conditions. This finding can provide a robust reference to remediate the sloping cultivated land use in mountainous regions, such as the Southwest Alpine Canyon Area of China.