High-throughput tools facilitate the rapid gathering of large amounts of plant phenotypic data that aid in improving the efficiency of breeding programs. However, under field conditions, there are several yield-contributing traits such as flower number or crop volume that cannot be measured or impractical to measure non-destructively can be estimated using high-throughput tools when combined appropriate sensors. The objective of this study to estimate the lentil flower number, crop volume, and canopy height using UAV (unmanned aerial vehicle) imagery. In the 2018 field season, three replications of 86 diverse lentil interspecific (Lens culinaris vs Lens orientalis) populations were planted in randomized complete block design within 1 X 1 m2 micro plots with 2 crop rows per plot. Ground-truth data of canopy height, crop biomass, and yield measured for all the populations and image acquisition was performed throughout the season. Segmented orthomosaics of individual micro plots were used to detect the lentil flowers (using convolutional neural networks, CNNs) and to estimate crop volume and plant height (using DSMs). The overall accuracy of CNNs for flower detection is 0.80 (trained vs tested plots). The accumulation of digital flower numbers over time was calculated using the “Area Under the Curve” (AUC) math function. The digital flower count based on AUC tends to have a good correlation (R=0.45) with the lentil seed yield. This study is the first attempt to detect lentil flowers using aerial imagery. Crop biomass and height also found a strong relationship with the digital plant volume (R=0.80) and canopy height (R=0.6) respectively. The results of this study evaluated the potential of aerial imagery in measuring some of the important lentil yield-contributing traits that will be of a great addition to the crop improvement programs.