Using imaging and computer vision to precisely identify and quantify fruits at different stages of plant development is important not only to optimize labour-intensive manual measurements of phenotypic data but also to make significant progress towards task automation.  The estimation of fruit yield plays a pivotal role in Precision Agriculture, aiding growers in more precisely forecasting market planning, workforce scheduling, procurement of suitable equipment, and other related considerations. There is also a high demand for automated methods to estimate fruit yield accurately and reliably in orchards. The advancements in Deep Learning models have been a great boon in this regard. In this work, YOLOv5 model is employed to detect and count mangoes on trees. The model has achieved a mean Average Precision@0.5 (mAP@0.5) as 99.5% and the MAP across the range of 0.5 to 0.95 as 81.9%, with a Mean Absolute Error (MAE) of 1.5 during testing.