Basics of Satellite Imagery Analysis, Part1
This article gives detailed information on satellite imagery as well as the various steps involved in analyzing the images.
What’s inside Satellite images?
Let’s do a simple comparison between normal images and satellite images. Normal images are images taken by a digital camera in our daily life. These photographs are made by the reflected rays of the spectrum in the visible region. In other words. Only the data regarding the visible spectrum is available in these types of photographs. Other spectrums like infrared, near-infrared, etc.. are not captured.
In the case of satellite images, cameras are capable of capturing a wide range of spectrums that a normal camera cannot achieve. For example, we need to find the temperature of an object present in the given image. The normal images fail this test. By employing the infrared region of the spectrum in satellite images, we can find the comparative temperature levels of objects in it.
Source for Satellite Images
If you have done any browsing regarding satellite imagery, I think that you will be seen some names like Landsat and Sentinel. Landsat satellites provide high-quality, multi-spectral imagery of the surface of the Earth. These moderate-resolution, remotely sensed images are not just pictures, but contain the visible and invisible light spectrum. Landsat satellites have the optimal ground resolution and spectral bands to efficiently track land use and to document land change due to climate change, urbanization, drought, wildfire, biomass changes (carbon assessments), and a host of other natural and human-caused changes.
Sentinel-2 multispectral sensor allows quality pictures of the terrain that can be implemented for land management, monitoring agriculture, and forestry.
Spectrum Bands
Sensors integrated into satellites are designed to acquire data in different ranges of frequencies along the electromagnetic spectrum. These ranges are called bands. These types of processes cannot be done using normal cameras. Many sensors on earth-observing satellites measure the amount of electromagnetic radiation (EMR) that is reflected or emitted from the Earth’s surface. These sensors, known as multispectral sensors, simultaneously measure data in multiple regions of the electromagnetic spectrum.
The bands above show will perform various applications like detecting vegetation, water content, roads, buildings, coastal areas, mineral and rock exploration, moisture sensing, etc.
Here I had taken Landsat for reference purposes only. The bands above mentioned can work on a combination level as well as an individual level based on the applications. The band combination of Landsat 8 is mentioned below
Natural Color (4, 3, 2)
The natural color composite uses a band combination of red (4), green (3), and blue (2). It replicates close to what our human eyes can see. While healthy vegetation is green, unhealthy flora is brown. Urban features appear white and grey and water is dark blue or black.
Color Infrared (5, 4, 3)
This band combination is also called the near-infrared (NIR) composite. It uses near-infrared (5), red (4), and green (3). Because chlorophyll reflects near-infrared light, this band composition is useful for analyzing vegetation. In particular, areas in red have better vegetation health. Dark areas are water and urban areas are white.
Short-Wave Infrared (7, 6 4)
The short-wave infrared band combination uses SWIR-2 (7), SWIR-1 (6), and red (4). This composite displays vegetation in shades of green. While darker shades of green indicate denser vegetation, sparse vegetation has lighter shades. Urban areas are blue and soils have various shades of brown.
Agriculture (6, 5, 2)
This band combination uses SWIR-1 (6), near-infrared (5), and blue (2). It’s commonly used for crop monitoring because of the use of short-wave and near-infrared. Healthy vegetation appears dark green. But bare earth has a magenta hue.
Geology (7, 6, 2)
The geology band combination uses SWIR-2 (7), SWIR-1 (6), and blue (2). This band combination is particularly useful for identifying geological formations, lithology features, and faults.
Bathymetric (4, 3, 1)
The bathymetric band combination (4,3,1) uses the red (4), green (3), and coastal bands to peak into water. The coastal band is useful in coastal, bathymetric, and aerosol studies because it reflects blues and violets. This band combination is good for estimating suspended sediment in the water.
I believe that you got an overview of the concepts of satellite imagery, image capturing, and composition. In the part 2 section, we will be focusing on the hands-on experience in the analyzing satellite image
