Abstract: Cultivated land is the foundation of agricultural development and a basic condition for ensuring human survival. With the rapid development of urbanization, which has increased non-agricultural employment opportunities and income, rural populations have migrated to urban areas, leading to large-scale abandonment of cultivated land. This phenomenon profoundly impacts national agricultural production and food security. As the Yellow River basin is a vital grain production base in China, accurately and efficiently obtaining the spatial distribution of abandoned cultivated land is crucial for preventing the further spread of this trend, holding significant theoretical and practical importance.This study focuses on the downstream regions of the Yellow River, specifically in Lankao County, Changyuan County, Fengqiu County, and Dongming County. It utilizes Landsat imagery from March to October, spanning the years 2003 to 2023, with the data undergoing preprocessing for this period. Considering the differences between cultivated land and other land types, spectral features, vegetation indices, texture features, topographic features, and Kauth-Thomas features were extracted to construct feature sets for each year from 2003 to 2023.A combination of stratified sampling and visual interpretation was employed to obtain the sample sets. This study employed a random forest probability model combined with the LandTrendr change detection method to extract abandoned and fallow lands, excluding urban areas.The accuracy of the abandoned cultivated land results was validated using the F1-Score, and the factors influencing abandoned cultivated land were analyzed. The results indicate that in the random forest probability model, EVI_p80, NDVI_p80EVI, and nir had high importance and discriminability in the output. Statistical analysis of the area of abandoned and fallow land within the study area revealed that from 2004 to 2022, the area of abandoned cultivated land followed a trend of initial increase followed by fluctuating decline, with an average area of abandoned cultivated land of 28.24 km². The maximum area of abandoned cultivated land was reached in 2007, amounting to 49.35 km². After 2020, the phenomenon of abandonment was effectively controlled. The area of fallow land was found to be relatively small and scattered. Utilizing the random forest probability model in conjunction with the LandTrendr change detection method, an F1-score of 0.87 was achieved for abandoned cultivated land, while a score of 0.89 was obtained for non-abandoned cultivated land, showing an improvement in accuracy compared to land cover change-based extraction. This demonstrates the effectiveness of the method for monitoring abandoned land in the Yellow River basin. To investigate the factors influencing the aabandoned ccultivated land, a multivariate regression analysis was conducted, incorporating total agricultural production value, total mechanical power, grain production, and the number of employed individuals in the primary industry, in relation to the area of abandoned cultivated land. The results indicate that an increase in the level of agricultural mechanization effectively reduces the abandonment of cultivated land, while an increase in grain production may raise the risk of abandonment. Although the number of employed individuals in the primary industry has decreased, its impact on land abandonment is limited, with policy regulation playing a significant role. The implementation of strict cultivated land protection policies has effectively reduced the area of abandoned cultivated land, which is crucial for ensuring national food security and promoting the sustainable development of agriculture. This study not only provides an effective technical approach for monitoring abandoned cultivated land in the Yellow River basin but also offers an in-depth analysis of the spatiotemporal changes and driving factors of cultivated land abandonment, providing valuable references for the formulation and implementation of agricultural policies in China.