The development of high yielding wheat varieties is crucial for wheat growing regions in South Asia. Plant height is an important agronomic trait in wheat due to its relationship with biomass and grain yield. The plant height trait is conventionally scored on a plot basis with a measuring stick, which is a subjective, low-throughput and labor intensive, and therefore prone to errors. The unmanned aerial systems (UAS)-based high-throughput phenotyping can be an alternative to the low-throughput visual/manual height estimation. Using a semi-automated analysis pipeline, we built a three-dimensional digital elevation model from multispectral images using MicaSense RedEdge camera onboard a quadcoptor UAS during the grain filling stage of wheat at Ludhiana, India in 2017-18. The plot-level digital height from the experiments consisting 300 plots were extracted and compared with visual/manual height scores. We found strong correlations (0.45-0.64, p-value <0.001) between visual and digital height scores across the experiments. Correlation between digital height and yield ranged from 0.03 to 0.38 and between visual height and yield -0.08-0.21. These results suggest the potential of UAS-based high-throughput height estimates in field-breeding experiments. The UAS system based model will improve the accuracy of selection and consequently the development of highyielding cultivars of wheat suitable South Asia.