Funding
Abstract
In this project, research will be focused towards introduction of new technologies and artificial intelligence in olive growing. First research direction is related to development of procedures and algorithms for olive trees analysis in order to detect olive fruits. This is especially challenging because intention is to develop a procedure that w ill enable detection of olive fruits in early stages of development w hen they are small. Second research direction is related to development of algorithms for estimation of fruit maturity and complete yield. Main hypothesis is that it is possible to develop a procedure that will enable quality information on fruits state and yield estimate based on olive groves images taken with an unmanned aerial vehicle. In this part, correlation between flowering intensity and yield will be investigated. During research, forming of an reference and publicly available labeled images database is expected (various time of day and seasons, various weather conditions). Procedures and algorithms for image preprocessing will be adopted for successful implementation on images collected with a drone and also to work on less constrains then approaches presented in the literature. Investigation of possible quality improvement of input data and, consequently, also the final results based on combination of information obtained RGB and multispectral sensors. Moreover, optimized architecture based on deep neural networks will be proposed in order to lower complexity and allow implementation on less demanding hardware.
Main research objectives
- Develop a new method for assessing the maturity and quantity of fruits
- New publicly available image sets for use by the wider research community.
Project details
Project start date: 16.12.2024.
Project end date: 15.12.2027.