The effectiveness of unmanned aerial systems (UASs) in monitoring crops and capturing detailed aerial images which are essential for estimating crop areas has been demonstrated. This study evaluated the utility of spectral, textural, and morphological transformations obtained from UAS-based sensor images in distinguishing between maize and soybean in a mixed-cropping environment. The researchers conducted a study on a plot of land measuring 0.324 ha, dividing it into 25 plots with varying mixtures. During the tasseling stage of maize, we used a Da-Jiang Innovations (DJI) Matrice 300 drone to capture digital red, green and blue (RGB) images 48 days after planting. To determine the best features for crop identification, we extracted different types of features (spectral, textural, and morphological) from each plot. For supervised classification using the random forest algorithm, we used 26 variables derived from high-spatial-resolution images. The findings of this study revealed that morphological features achieved the highest accuracy of 0.93 and an F1-score of 92, followed by a combination of textural and morphological features, with spectral features being the least effective. These results emphasize the importance of morphological analysis for classifying crops in mixed-cropping systems.