In a previous Blog-post, we implemented an application to identify water surfaces using Synthetic Aperture Radar (SAR) images produced by radar-based sensors. The main advantage of using this type of sensor is that it can overcome the limitation of optical spectral sensors in the case of clouds and weather conditions that block spectral reflectance. An example is the Sentinel 1 satellite, which uses microwaves that can penetrate haze, light rain and clouds to illuminate the earth and measure the backscatter and travel time of the transmitted waves reflected by objects on the ground. It has the clear advantage of being far less dependent on the weather and can acquire images at night to obtain near-real-time information and monitor the earth over a large area. Furthermore, SAR images provide a lower backscatter reflectance for water surfaces as compared to non-water surfaces. This significant difference can be used to distinguish water surfaces and other land areas.
Deep Learning algorithms are being used more often to process remote sensing data. Data captured from satellite platforms requires reliable and generalized methods to periodically extract and analyse useful information. Environmental planners are becoming increasingly interested in applications, such as land cover mapping ranging from urban development to crop detection, to monitor climate change and wildfires.