As a further step in our digitalisation approach, IDVIA is carrying out the HEL OR project, which aims to develop a system for detecting damage to woody crops (especially citrus crops) after frosts in areas at risk.
To achieve this, our engineers are developing an innovative methodology based on the processing of multispectral images obtained at low altitude using drones, through which the detection of trees is carried out using SLIC unsupervised machine learning algorithms.
Once the independent trees are identified, an analysis of their various vegetation indexes can be carried out. From these indexes, and using a neural network previously trained with pre-existing data of trees with frost damage, we are able to quickly obtain a prediction of the state of the trees analysed.
The results obtained after such processing are displayed on a digital platform, being the objective to be able to have several tabs or views with different information: trees geospatially located on a map with the information of their condition for that specific flight, average vegetation indexes, evolution of such indexes, etc.
|The HEL OR project will be completed in mid-2023 thanks to funding from the Spanish Ministry of Science and Innovation and the State Research Agency in the framework of the call for proposals RETOS COLABORACIÓN 2019 (RTC2019-006863-2)|