Thesis subject

MSc thesis topic: High-resolution UAV imagery for fruit counting in orchards utilizing deep learning

Advances in sensor miniaturization are increasing the global popularity of unmanned aerial vehicle (UAV) in agriculture. In the context of smart agriculture, site-specific management in fruit orchards is required throughout the whole growing season – from flowering, fruitlet development, ripening, harvest, till tree dormancy. Yet, the application of UAV in orchard management is in its infancy. Are you the brave and ambitious explorer who can not wait to redefinite the tiny orchard world with the sharp sword of UAV?

Detailed spatial apple yield information is key for growers to facilitate efficient utilization of resources and to optimize and streamline the harvest process. So far, however, there has been little discussion about the feasibility of utilizing UAV in apple counting, though it has great potential to provide efficient and more reliable estimation results than manual work. Further studies regarding the individual tree identification and apple counting, based on multiple UAV products, such as high resolution RGB images, orthophoto and 3D point clouds, would be worthwhile.

Objectives

  • Literature review: yield estimation in orchards
  • Algorithm design, deep learning architecture building
  • Experiments in the already available datasets

Literature

Theme(s): Sensing & measuring