Tree-TEC: Conteo de árboles con imágenes satélites en Costa Rica

Accurate tree counting is fundamental for forest management, biomass assessment, and understanding terrestrial ecosystems. However, traditional manual counting methods are laborious, costly, and error-prone, limiting the scale and frequency of forest inventories. This research project proposes developing an innovative tree counting system based on satellite imagery and artificial intelligence (AI), leveraging recent advances in remote sensing, deep learning, and computer vision.
The proposed system aims to provide an efficient and scalable alternative to traditional counting methods, enabling rapid and accurate quantification of trees across large land areas without extensive fieldwork. By utilizing high-resolution satellite images and advanced AI algorithms for tree detection and counting, the system promises to significantly reduce the time and resources required for tree counting while maintaining a high level of accuracy.
The research will focus on developing and optimizing deep learning models, such as convolutional neural networks (CNNs) and semantic segmentation architectures, specifically designed for automated tree detection and counting in diverse environments and forest types. These AI models will be trained on a diverse, manually labeled dataset to ensure their robustness and adaptability to different forest conditions

Proponer algoritmos para la detección y seguimiento de árboles en imágenes satelitales para conteo y monitoreo forestal.

Computing algorithms, AI and software design.