One of the biggest challenges, which affects people's livelihoods and the environment, is land degradation, with effects on climate change, but also on the loss of productivity and biodiversity, being the result of the discrepancy between land quality and cultivation intensity. Remote sensing data can provide information on crops by calculating different vegetation and / or soil indices. To be able to create maps with different types of crops, for many years plays an important role, from environmental to economic. The advantage of using satellite images in the study of crops is that the information can be extracted in a relatively short time, only benefiting from an adequate infrastructure, without interfering with the lands / crops. The main objective of this study was to evaluate the accuracy of the information obtained after the Random Forest classification over a period of 2 years, crops that were affected by land-use changes and soil properties. The study area was in Valea Oii catchment, Romania (47°21ʹ0.86ʺ N to 47°13ʹ23.32ʺ and 26°49ʹ37.07ʺ to 27°10ʹ35.68ʺ E), an elevation between 64 and 425 m (110 m to Baltati), a continental climate with an average annual temperature range of 8-10.4oC and an annual precipitation of 500-700 mm. 62% of the area is represented by agricultural uses, respectively 3700 ha other land uses, with chernozems (60 %) and phaeozems (10 %) followed by anthrosols (12 %) and aluviosols (6 %). More precisely, the study was conducted in the Baltati area (4518 ha), part of the Valea Oii catchment. Land use changes, without a correlation between soil and crop type, lead to affecting their quality and quantity. In this analysis, soil and vegetation indices were used, such as NDVI and NDWI, BI, SBL, and GOSAVI, which aimed to highlight the vegetation, being able to differentiate the types of crops. Also, the Random Forest algorithm was used for the classification of crop types, by introducing various indices in the model, and the resulting information was validated with field data. The results show a high accuracy of the information, making this research a particularly appropriate approach when land management measures are required, as in soil complexes involving degraded land. The results show a high level of information accuracy, making this research a particularly appropriate approach, especially when land management measures are required.