Unmanned Aerial Vehicles (UAVs) have become popular for use in plant phenotyping of field based crops, such as maize and sorghum, due to their ability to acquire high resolution data over field trials. Field experiments, which may comprise thousands of plants, are planted according to experimental designs to evaluate varieties or management practices. For many types of phenotyping analysis, we examine smaller groups of plants known as "plots." In this paper, we propose a new plot extraction method that will segment a UAV image into plots. We will demonstrate that our method achieves higher plot extraction accuracy than existing approaches.