Photogrammetry and Remote Sensing

Photogrammetry and remote sensing encompass the acquisition, processing and interpretation of image-based data to measure and map terrestrial and aquatic environments. Photogrammetry derives three-dimensional geometry from overlapping two-dimensional photographs or imagery, employing principles of central projection and stereopsis to reconstruct terrain, infrastructure and vegetation structure. Remote sensing extends beyond passive photography to include active systems—radar, lidar and hyperspectral scanners—that record reflected or emitted electromagnetic energy across multiple wavelengths. Together these tools deliver spatially extensive, repeatable observations spanning centimetre- to kilometre-scale resolutions, enabling applications from topographic mapping and urban planning to forest biomass estimation, coastal monitoring and precision agriculture. Advances in sensor miniaturisation, unmanned aerial vehicles and data fusion have driven unprecedented gains in accuracy, temporal coverage and thematic insight, underpinning scientific research, resource management and disaster response worldwide.

Research from Nature Portfolio

A human–machine collaborative framework has been introduced to infer fine-scale economic development from publicly available satellite imagery. By combining lightweight subjective ranking annotations with deep convolutional networks, the system predicts grid-level socioeconomic indicators in data-poor regions without relying on traditional ground surveys. Applied to several low-resource nations, the method yields high-resolution maps of development trends, demonstrating the power of remote sensing to guide sustainable planning where conventional data are scarce.

A fully integrated photonic-electronic coherent lidar engine has been realised on a single wafer-scale platform. Combining a tunable Vernier laser with a monolithic erbium-doped waveguide amplifier and high-voltage driver in CMOS-compatible technology, the system achieves 10 m range measurements at 10 cm precision with 50 kHz acquisition rates. This turnkey coherent ranging solution illustrates a path towards compact, mass-manufacturable lidars for environmental mapping and terrain modelling in remote sensing applications.

Research from all publishers

Benchmarking of NASA’s ICESat-2 ATL08 photon-counting lidar product over North American boreal forests has validated new 30 m × 11 m data subsets against high-density airborne lidar references. Across both standard (100 m × 11 m) and enhanced sampling grids, terrain and canopy height biases remain below half a metre with root-mean-square errors under 4 m, providing critical guidance for ecosystem modelling and carbon cycling studies.

Innovations in lidar photon classification have addressed low-signal returns over complex topography. An adaptive rotating-ellipse local outlier factor (LOFR) algorithm aligns search areas to terrain slope, effectively segregating ground, canopy and noise photons without prior digital elevation models. Validation against airborne lidar yields mean absolute height errors below 2 m, enabling robust large-scale processing of spaceborne photon-counting data.

At finer scales, the G-LiHT airborne imager integrates lidar, hyperspectral and thermal sensors to deliver centimetre-level mapping of canopy architecture, foliar spectral traits and surface temperature. Synergistic fusion algorithms produce detailed point clouds and spectral mosaics, underpinning new approaches to biodiversity assessment, biomass estimation and calibration of satellite-borne vegetation products across diverse ecoregions.

Photogrammetry and Remote Sensing publication trend

The graph below shows the total number of articles in photogrammetry and remote sensing across all publications each year (not limited to Nature Index journals).

Technical terms

Photogrammetry: The science of measuring three-dimensional coordinates of points on an object or terrain by analysing two-dimensional images from multiple viewpoints.

Remote sensing: The acquisition of information about the Earth’s surface or atmosphere from a distance, typically using aerial or satellite sensors that detect reflected or emitted electromagnetic radiation.

LiDAR: Light Detection and Ranging, an active remote-sensing method that emits laser pulses and measures the travel time of returned photons to generate high-resolution three-dimensional point clouds.

Hyperspectral imaging: The collection of image data in tens to hundreds of contiguous narrow spectral bands, enabling detailed discrimination of surface materials based on their unique spectral signatures.

Digital elevation model (DEM): A gridded representation of ground-surface elevation, derived from photogrammetric stereo reconstruction, lidar point clouds or radar interferometry.

Photon counting: A lidar detection mode in which individual returning photons are recorded to maximise vertical precision in low-energy regimes.

References

1. A human-machine collaborative approach measures economic development using satellite imagery. Nature Communications (2023).
2. Photonic-electronic integrated circuit-based coherent LiDAR engine. Nature Communications (2024).
3. A systematic evaluation of multi-resolution ICESat-2 ATL08 terrain and canopy heights in boreal forests. Remote Sensing of Environment (2023).
4. NASA Goddard’s LiDAR, Hyperspectral and Thermal (G-LiHT) Airborne Imager. Remote Sensing (2013).