Corn planting quality assessment in very high-resolution RGB UAV imagery using Yolov5 and Python

Abstract. Uniform plant spacing along crop rows is a primary concern in maximising yield precision agriculture, and research has shown that variation this uniformity detrimental effect on productive potential. This irregularity needs to be evaluated as early efficiently possible facilitate effective decision-making. Traditionally, seedling sampled manually site, however recent technological developments have made it refine, scale automate process. Using machine-learning (ML) object detection techniques, plants can detected very high-resolution RGB (redgreen-blue) imagery acquired by an unmanned aerial vehicle (UAV), after processing geometric analysis of the results measurement variability intra-row distances obtained. proposed technique superior traditional methods since sampling over more area less time, are representative objective. The main benefits speed, accuracy cost reduction. work aims demonstrate feasibility automatically assessing sowing quality any number images, using ML Shapely Python library for geometrical analysis. prototype model detect 99.35% corn test data from same field, but also detects 1.89% false positives. Our algorithm been robust finding planting orientation interplant lines cases. result coefficient (CV) calculated per sample image, which visualised geographically support