Remote sensing

Inferring properties of a distant object from measured signals, often electromagnetic radiation. Spatial sampling, spectral information and observation cadence must jointly resolve the physical process of interest. Received radiance alone need not uniquely separate temperature, emissivity and atmospheric attenuation.

Range measurement using transmitted laser light and its returned signal. For a pulse traveling a round trip in air, the elementary range is $r\simeq c\Delta t/2$, where $c$ is the propagation speed and $\Delta t$ the measured travel time. Sensor position, pointing and time synchronization place returns in a three-dimensional point cloud. Scanning lidar can record multiple returns from vegetation and the ground; a return is an observation of a reflecting surface, not automatically bare-earth elevation.

Lidar acquired from an aircraft with surveyed trajectory and orientation. Ground-return classification removes canopy and structure returns before constructing a bare-ground digital elevation model. Penetration through forest is sampling through canopy gaps, not unrestricted transmission through leaves. Areas with few or no ground returns require uncertainty assessment rather than treating interpolated heights as observations.^[ref]

Recovery of object positions and shape from photographs with overlapping views. Known or estimated camera geometry and matched image features constrain three-dimensional point positions. A metric reconstruction requires scale and a coordinate reference; image pixel size is not automatically the positional accuracy of the reconstructed surface. Stereo photogrammetry and structure-from-motion photogrammetry use related geometry with different acquisition and camera-estimation strategies.

An identifiable image feature or target with independently surveyed coordinates, used to georeference and constrain a photogrammetry model. A check point is withheld from the fit to assess reconstruction accuracy. Distribution across the site, accurate survey and consistent height datums matter as well as the number of points.

A photogrammetry workflow jointly estimating camera geometry and scene structure from many overlapping images, commonly followed by dense surface reconstruction. Ground control points or accurately surveyed camera positions establish scale and georeferencing. Independent check points test accuracy rather than participating in the fit. A fine image ground-sampling interval does not alone establish equal vertical accuracy.

Three-dimensional reconstruction using two or more overlapping photographs from different viewing positions. Corresponding image points define viewing directions whose intersection estimates the surface position. Surveyed camera positions and orientations, camera calibration and ground control points establish the coordinate geometry. Ordinary optical stereo models the visible surface, which can be forest canopy rather than bare ground.

Imaging in many narrow, usually adjacent wavelength bands. Relative to broad-band multispectral imaging, it can resolve narrower mineral absorption features in spectral reflectance, helping distinguish surface lithologies, at the cost of data volume and often reduced signal per band.

Removal or modeling of atmospheric absorption and scattering contributions to a measured signal to estimate surface or water-leaving optical quantities.

The Geostationary Operational Environmental Satellite series, used for frequent weather observations from geostationary orbits.

A European meteorological satellite series including geostationary orbit missions for repeated regional weather observations.

A European radar mission using Sun-synchronous orbits and supporting interferometric synthetic-aperture radar.

A European optical multispectral imaging mission using Sun-synchronous orbits for repeated observations of land and coastal areas.

A series of Earth-observation missions using repeated optical and multispectral imaging from near-polar Sun-synchronous orbits.

Imaging in several wavelength bands. Differences in spectral reflectance can distinguish surface materials, including vegetation and water.

An image formed by dividing corresponding calibrated pixels in two spectral bands. If their signals have a common multiplicative factor, $L_1=k{\rho }_1$ and $L_2=k{\rho }_2$, their ratio is ${\rho }_1/{\rho }_2$ and does not depend on $k$. This explains suppression of some common illumination effects and enhancement of spectral contrasts. Additive atmospheric radiance and small denominators prevent exact cancellation and can amplify errors. Normalized difference vegetation index is a related normalized spectral contrast.

The normalized contrast of red and near-infrared reflectances. Large positive values indicate a vegetation-like spectrum, but soil, atmosphere, shadows and canopy saturation affect its interpretation.

A displayed image assigning measured bands to colour channels that differ from ordinary visual perception. Assigning near-infrared to red is a common way to highlight vegetation.

The steep increase of healthy green vegetation reflectance between red absorption and the high near-infrared scattering plateau, typically around 0.68–0.75 micrometres.

Comparison of coherent radar phases from repeated observations to infer relative line-of-sight displacement. A complete deformation fringe represents half a radar wavelength after other phase contributions have been removed.

An image of the relative radar phase between co-registered coherent observations. A displacement changes the two-way path length, giving phase change proportional to $4\pi d_{\mathrm{LOS}}/{\lambda }_r$, with sign set by the adopted range convention. A full wrapped phase cycle therefore corresponds to half a radar wavelength of line-of-sight displacement. Orbital geometry, topography, atmospheric delay and scattering changes must be separated from deformation; a fringe is not uniquely an elevation contour or an earthquake displacement.^[ref]

A normalized measure of consistency of complex radar returns over a neighborhood. Low coherence makes interferometric phase unreliable and can arise from vegetation, surface changes, viewing differences or extreme displacement gradients.

The component of a displacement vector along a specified sensor–target viewing direction. Its sign must state whether motion toward or away from the sensor is positive.

The phase difference of complex radar returns. It includes deformation, topography, orbital geometry and propagation contributions; the displayed wrapped phase is defined modulo $2\pi$.