Most commonly we refer to remote sensing as the imagery information obtained from sensors housed on satellite platforms .
Satellite sensors record the various waves emitted/reflected from/to the Earth's surface.
A sensor is a device that detects a signal and it is part of a data acquisition system. A signal is a form of energy and we can classify signals according to the energy they detect.
A sensor is a type of transducer ( it converts a signal in one form of energy to another form of energy). The output signal of the sensor may be described in terms of amplitude, frequency and phase.
So a sensor detects electromagnetic energy that travels through space at the speed of light.
The signal is in fact proportional to the electromagnetic energy that reaches the sensor. The sensors capture the reflected or emitted energy from the surface and convert it to an electrical signal. The radiation is reflected, absorbed and emitted in varying proportions depending on the characteristics of the surface and the wavelength. The atmosphere, as the energy travels through it, absorbs, distributes, and modifies both the incident radiation and the one directed towards the sensor.
We use the signal received by the sensor to manufacture images. Sensors are divided into:
- Imaging sensors (Scanning Radiometer, SLAR and SAR)
- Non-imaging sensors (radar altimeter, weather, atmospheric sounder)
We can get information measuring the emitted or reflected electromagnetic radiation of distant objects in certain bands (infrared, visible light, microwaves) of the electromagnetic spectrum.
The electromagnetic spectrum is the range of all possible frequencies of electromagnetic waves and is divided into seven regions (or bands) depending on the wavelength.
Digital sensors acquire these regions with different spectral, geometric, radiometric and temporal resolutions.
- Spectral resolution - the number of bands detected and their width: the greater the number of bands, the lower their width, the greater will be the ability to identify objects according to the proportion of incident light that a given surface is able to reflect .
- Geometric resolution: the minimum area on the ground that an instrument can view from a given height at a given time and is represented by the size of the surface element recognizable in a recorded image.
- Temporal resolution: the time period between two successive shots of the same area.
- Radiometric resolution: the minimum signal difference that the sensor is able to distinguish (i.e. How sensitive is the sensor in recording small differences in reflected or emitted energy).
There are passive and active remote sensing sensors. A passive instrument is one that detects electromagnetic radiation reflected or emitted from natural sources (the Sun is an example). Signal logging occurs only when the Sun illuminates the Earth.
Active sensors emit radiation directed towards the object (LIDAR, RADAR). The sensor record and measure the radiation reflected from the object.
We gather information from sensors mounted on orbiting satellites. The digital data acquired by satellites are transmitted to ground stations. These digital data, once processed, will give us images.
We can classify the satellite imaging systems according to the spectral regions used in data acquisition systems:
- Ultraviolet and Visible
- Visible and Reflective Infrared
- Thermal Infrared
The sensor, as we have seen, measure the electromagnetic radiation reflected or emitted in certain bands of the electromagnetic spectrum. On an image these bands are displayed using the primary colors (red, green, blue) so that people can look at them.
The digital images resulting from the sensing process are records of the amount of light that stroke the sensor. Data transmitted to the ground are affected by errors of various types.
A digital image is made up of pixels (picture elements). In our context a pixel represents the smallest discernible area element on the Earth's surface and it can be addressed by its physical coordinates. The location of each pixel must be identified exactly. Geometric errors (i.e. errors due to the rotation of the Earth, to track movement simultaneously on the ground during a cycle of satellite observation, etc...) do not allow the exact identification of that position. These errors are compensated by suitable computerized correction programs.
A pixel has also an intensity value, the average measured quantity (reflected, emitted and scattered electromagnetic radiation). This value is stored as a digital number.
Coded numbers about a generic pixel are not always properly relevant energy levels at various wavelengths, due to the radiometric errors. A cause of error is the spread of radiant energy in the atmosphere, which decreases the contrast of images. This error is correct trough a procedure known as removal of haze.
The second most common cause of error is due to the radiometric sensors. The "answer" of the instrument may be different in spite of equal reflectance levels. By reflectance we mean the incident light that a surface is able to reflect. The data that they provide must therefore be calibrated at each detection cycle.