A Drone-Powered Deep Learning Methodology for High Precision Remote Sensing in California’s Coastal Shrubs

Wildland conservation efforts require accurate maps of plant species distribution across large spatial scales. High-resolution mapping is difficult in diverse, dense communities, where extensive ground-based surveys are labor-intensive and risk damaging sensitive flora. satellite imagery available at scales needed for community areas, but can be cost prohibitive lack resolution to identify species. Deep learning analysis drone-based aid classification these communities regions. This study assessed whether deep modeling approaches could used map complex chaparral, coastal sage scrub, oak woodland communities. We tested the effectiveness random forest, support vector machine, convolutional neural network (CNN) coupled with object-based image (OBIA) diverse shrublands. Our CNN + OBIA approach outperformed forest machine methods accurately tree shrub species, vegetation gaps, even distinguishing two congeneric similar morphological characteristics. Similar accuracies were attained when applied neighboring sites. work key identification scale research monitoring chaparral other wildland Uncertainty model application associated less common intermixed canopies.