Deep Learning classifies land cover/land use
Identifying the physical aspect of the earth’s surface (Land cover) as well as how we exploit the land (Land use) is a challenging problem in environment monitoring and many other subdomains. This can be done through field surveys or analyzing satellite images (Remote Sensing). While carrying out field surveys is more comprehensive and authoritative, it is an expensive project and mostly takes a long time to update.
With recent developments in the Space industry and the increased availability of satellite images (both free and commercial), deep learning and Convolutional Neural Networks has shown a promising result in land use classification.
A project conducted by Abdishakur used the freely available Sentinel-2 satellite images to classify 9 land use classes and 24000 labeled images. The original dataset contains 10 classes and 27000 labeled images and is available here.
As a result, Abdishakur train a model with 0.94 accuracy. While this is less than the accuracy reported in the original paper with the dataset (0.98), it is relatively high for its objectives.
Knowledge about land use/land cover has become important to overcome the problem of biogeochemical cycles, loss of productive ecosystems, biodiversity, deterioration of environmental quality, loss of agricultural lands, destruction of wetlands, and loss of fish and wildlife habitat. Currently, it is very convenient and accurate to use deep learning model to monitor the LC/LU changes, providing us with a powerful tool to better manage the valuable land resources.
