A deep learning technique was used to recognize, count, and categorize date palm trees in the Kingdom of Bahrain. The MRCNN model of convolutional neural networks has been presented and assessed. The test area was comprised of the entire land region of Bahrain’s Northern Governorate, which encompassed around 175 square kilometers. The raw material for extracting palm tree distributions and specifying the exact inventory areas that can contribute to Bahrain’s overall food security has been very high-resolution satellite images. This paper has demonstrated the value of incorporating deep learning techniques into the GIS environment to aid in the successful generation of agricultural data from satellite imagery.
Paper
Full text
Developing Date Palm Tree Inventory from Satellite Remote Sensed Imagery using Deep Learning
Semantic Scholar · Environmental Science · 2021
Abstract
A deep learning technique was used to recognize, count, and categorize date palm trees in the Kingdom of Bahrain. The MRCNN model of convolutional neural networks has been presented and assessed. The test area was comprised of the entire land region of Bahrain’s Northern Governorate, which encompassed around 175 square kilometers. The raw material for extracting palm tree distributions and specifying the exact inventory areas that can contribute to Bahrain’s overall food security has been very high-resolution satellite images. This paper has demonstrated the value of incorporating deep learning techniques into the GIS environment to aid in the successful generation of agricultural data from satellite imagery.