The development of algorithms can be key in the search for solutions in the productive sector
At the beginning of this semester, Dr. Yissedt Lara Diaz joined an interesting applied research project. It is called 'Deep-Hub: platform based on artificial intelligence for the analysis of aerial images of forest interest' (ID21I10354), financed by the Fund for the Promotion of Scientific and Technological Development, Fondef, of the National Agency for Research and Development, ANID, led by Dr. Guillermo Cabrera Vives -director of the Data Science Unit (UDS)- and co-directed by Dr. Simón Sandoval Rocha, academician from the Department of Forest Management and Environment of the Faculty of Forestry Sciences, both from the Universidad de Concepción, UdeC.
Dr. Lara obtained the highest academic degree in mid-2022 in the UdeC Program of Doctorate in Applied Sciences with a minor in Mathematical Engineering, and recently joined the UDS within the aforementioned project.
'Deep-Hub' integrates artificial intelligence algorithms with aerial images of forest masses, addressing problems of identification and segmentation at the individual tree level, classification of land use and detection of anomalies from temporal analysis of images.
In this context, the functions of Dr. Lara are related to the "proposition of mathematical models to address the problems that arise in the project, linking data that is already available or that has been obtained by using the technological platform that she has developed the research group for the processing of forest images”.
The scientist explains that this research initiative "arose from a need for companies and it was about finding a solution to count trees and from the images, they created an algorithm to solve questions such as how many trees are there in a certain number of hectares."
Currently, the challenge from a scientific point of view, details Yissedt, is "to expand the functions of the application, how to make better use of that information that is already available and, currently, I am working on that, proposing or raising a problem, clarifying it and then create algorithms with it.”
Based on the experience that the researcher already had in the development of models on the spread of diseases such as COVID, she has now begun to work on a line of research related to the spread of pests in the field of forest plantations. "They left me free to raise the model and access some information that is needed."
Yissedt explains that this Fondef project is close to completing its implementation period, which will also correspond to collaborating in the drafting of the final reports with a view to applying for the second stage of development of the initiative.
In this regard, Dr. Raimund Bürger, Director of the aforementioned postgraduate program and Deputy Director of the Center for Research in Mathematical Engineering, CI²MA, highlighted that "as former guiding professor of Dr. Lara, it has been great news to know that she has managed to successfully insert herself into in said project and has once again demonstrated his ability to carry out independent investigative work. The subject is, in any case, very interesting and I hope that through contacts like this the links between CI²MA and the Faculty of Forestry Sciences will be strengthened”.
Photo 1: Dr. Yissedt Lara Diaz
Photo 2: “The models I work on try to capture the dynamics of insect pests and trees in plantations. I essentially look for the infestation dynamics, and this depends on the life cycles of the pest and how they survive in the trees”, details the scientist