Dicamba is an herbicide used for postemegence control of several broadleaf weeds in corn, grain sorghum, small grains, and non-cropland. Currently, dicamba-tolerant (DT) soybean and cotton are under development, which provide new options to combat weeds resistant to glyphosate, the most widely used herbicide.  With the use of DT-trait cotton and soybean, off-target dicamba drift onto susceptible crops will become a concern. To relate soybean injury to different rates of dicamba applications, field experiments were conducted in 2012, 2013, and 2014 at the research farm of the USDA-ARS Crop Production Systems Research Unit in Stoneville, Mississippi, USA. For the experiments, thirty-two soybean plots were established on a 4.5 ha field of the research farm and, at the four-trifoliolate-leaf stage, treated with dicamba at six rates following a randomized complete block design with four replications. In 2014, five weeks after dicamba treatment, RGB color images were acquired with a GoPro camera on a small octocopter flying over the soybean field. Also, at one and half weeks and 10 weeks after dicamba treatment, RGB color images and near-infrared (NIR) images were acquired by a Canon digital camera and a customized Canon digital camera with NIR pass filter on a fixed-wing remotely controlled plane over the field. RGB color images and NIR images were stacked to generate color infrared (CIR) images of the entire soybean field.  Vegetation indices, such as normalized difference photosynthetic vigor ratio extracted from the RGB images and normalized difference vegetation index from CIR images over the dicamba-treated soybean field, were highly correlated with soybean yield. A monotonic trend of the vegetation indices was also observed with increasing rates of dicamba. This study demonstrated that the RGB and CIR images acquired on small UAVs have a great potential in assessing crop injury from dicamba spray in the field.