What is remote sensing?
Remote sensing means getting information about objects or areas on the Earth's surface from a distance, without touching them. We record the electromagnetic energy they reflect or give out.
Your eyes are a simple remote sensor: they record reflected light. Cameras and satellite sensors do the same, and can also "see" light our eyes cannot, such as infrared.
Aerial photos taken from balloons and planes were the first remote sensing. Satellites (from the 1970s) now cover the whole Earth again and again.
Stages in remote sensing
- Source of energy: usually the Sun (passive sensing). Radar sends its own energy (active sensing).
- Propagation through the atmosphere: gases, water vapour and dust absorb or scatter some energy. Sensors use "windows" where the air lets energy pass.
- Interaction with the surface: objects absorb, transmit or reflect the energy.
- Reflected or emitted energy travels back through the atmosphere to the sensor.
- Detection by the sensor: the sensor records the energy.
- Transmission to a ground station: data is sent down, e.g. to the NRSC station at Shadnagar near Hyderabad.
- Conversion into data products: photos or digital images.
- Extraction of information and use: experts interpret the images and make maps for users.
Spectral signature
Each object reflects different amounts of energy at different wavelengths. This pattern is its spectral signature, like a fingerprint.
- Healthy vegetation: reflects some green, very little red (chlorophyll absorbs it), and a lot of near-infrared (NIR).
- Clear water: reflects little in visible light and almost nothing in NIR, so it looks dark.
- Dry soil: reflection rises steadily with wavelength.
By comparing bands, we can separate crops from forest, wet land from dry, clean water from muddy water.
Platforms
A platform is what carries the sensor.
- Ground-based: towers, cranes or hand-held instruments.
- Airborne: balloons, aircraft and drones; give very detailed images of small areas.
- Spaceborne: satellites; cover large areas repeatedly.
Two main kinds of satellites
- Sun-synchronous satellites: orbit about 700–900 km high, passing near the poles. They cross the same latitude at the same local time each day, so the lighting is similar in every image. Used for mapping land, crops, forests and water. Examples: IRS/Resourcesat, Landsat.
- Geostationary satellites: orbit about 36,000 km above the Equator, moving with the Earth's rotation, so they stay over one spot. They watch a whole hemisphere all the time; used for weather and communication. Example: INSAT series.
Sensors
A sensor is the device that records energy.
- Photographic (camera) sensors: record the whole scene at one moment on film or a chip. Used in aerial photography.
- Non-photographic (scanning) sensors: build the image bit by bit.
- Whiskbroom (across-track) scanner: a rotating mirror sweeps side to side across the path, recording one point at a time.
- Pushbroom (along-track) scanner: a long row of detectors records a whole line at once; the satellite's forward motion adds line after line.
- Multispectral sensors record several bands (e.g. blue, green, red, NIR) at the same time.
Resolution
- Spatial: the ground size of one pixel (e.g. 5 m, 23.5 m). Smaller = sharper.
- Spectral: how many and how narrow the bands are.
- Radiometric: how many brightness levels (e.g. 0–255).
- Temporal: how often the same place is imaged (revisit time).
Data products
Photographic products
Aerial photographs and film prints. Tones are continuous (smooth). They can be viewed in 3D with a stereoscope using overlapping pairs.
Digital products
A digital image is a grid of tiny squares called pixels. Each pixel stores a number, the digital number (DN), for its brightness, often from 0 (black) to 255 (white). Computers can enhance, classify and measure them.
Colour composites
Three bands are shown through blue, green and red. In a standard false colour composite (FCC), NIR is shown as red, red as green, and green as blue. So healthy vegetation looks bright red, clear water dark blue or black, and towns bluish-grey.
Interpreting images
Interpreters look for clues: tone/colour, size, shape, texture (smooth or rough), pattern (e.g. rows of an orchard), shadow and association (a factory near a railway). Uses include crop estimates, forest and flood mapping, city growth, drought watch and disaster relief.
Try it
Open a phone photo and zoom in until you see little squares: those are pixels. Now use the slider in the last 3D step: at what pixel size can you no longer spot the red house?
Key formulas and definitions
- Stages: source → atmosphere → surface → return → sensor → station → product → user
- Digital image = grid of pixels, each with a DN (e.g. 0–255)
- Standard FCC: NIR → red, red → green, green → blue
- Resolution: spatial, spectral, radiometric, temporal
Worked examples
1. In a standard FCC a large patch is bright red. What is it likely to be?
Healthy vegetation (crops or forest), because plants reflect strongly in NIR, which is shown in red.
2. Which satellite type would you use to watch a cyclone all day?
A geostationary satellite such as INSAT, because it stays above the same place and images it again every few minutes.
3. A 10 km × 10 km area is imaged with 5 m pixels. How many pixels along one side?
10,000 m ÷ 5 m = 2,000 pixels per side (2,000 × 2,000 = 4 million pixels).
4. Why is water dark in NIR images?
Water absorbs almost all near-infrared energy, so almost nothing is reflected back to the sensor.
Common mistakes
- Thinking remote sensing needs satellites only. Aerial photos and ground sensors are remote sensing too.
- Mixing up satellites: sun-synchronous are low (~800 km) and polar; geostationary are high (36,000 km) over the Equator.
- Thinking red in a false colour image means red objects. In a standard FCC, red means strong NIR, usually healthy plants.
- Believing a bigger pixel gives a sharper image. Smaller pixels mean higher spatial resolution.