Unmanned aerial vehicles (UAV) have become to play an important role in many tasks in agriculture. They can be used to take thermal, RGB, multispectral images among others. However, a useful information for farmers derived from this kind of images is required as a final product.
Yield estimations in apple orchard is crucial for farmers. This information helps them to make better decisions. Currently, these estimations are made by expert technician based on manual fruit counts. This methodology is laborious, expensive and time-consuming. Hence, a more accurate and faster approach would improve the current practices.
This thesis proposes the development of model base on pre-trained models (Yolo V3, Mask R-CNN, etc.) for fruit detection in real time on apple orchards. This thesis will be carried out within the follow steps:
- Annotate the images using an appropriate tool the;
- Algorithm design and training;
- Experiments in the already available dataset.
- Aggelopoulou, A. D., Bochtis, D., Fountas, S., Swain, K. C., Gemtos, T. A., & Nanos, G. D. (2011). Yield prediction in apple orchards based on image processing. Precision Agriculture, 12(3), 448–456.
- Bargoti, Suchet & P. Underwood, James. (2016). Image Segmentation for Fruit Detection and Yield Estimation in Apple Orchards. Journal of Field Robotics.
- Computer vision and UAV enthusiast
- Willing for mastering deep learning
- Excited to work in developing useful tools for farmers
Theme(s): Sensing & measuring