Drought stress is one of the major limiting factors in soybean growth and productivity. Leaf wilting is considered as a first visible symptom of soybeans under drought condition. In soybean breeding programs, genotypes with slow-wilting trait have been identified as drought tolerant cultivars. The traditional method uses visual ranking based on experience to discriminate slow-and fast-wilting genotypes, which is subjective and time-consuming. Recent developments in UAV- (Unmanned Aerial Vehicle) based high-throughput phenotyping have shown a great potential of measuring plant traits and detecting plant responses under stresses. The goal of this study was to investigate the potential of discriminating soybean genotypes with different wilting characteristics (fast- and slow-wilting) using UAV-based high-throughput phenotyping. Image-derived features, namely NDVI_std, gNDVI_max and hue_mean were found having significant differences between fast- and slow-wilting genotypes (Fig. 1).