top of page
Search

Understanding Sub-Meter Satellite Imaging and GSD

Satellite images are a primary source of geospatial data used in mapmaking, urban planning, environmental observations, infrastructure management, defense, agriculture, disaster relief operations, and many other fields. With advances in satellite imaging technology, the user community benefits from ever-improving satellite imagery with higher spatial resolution.


Sub-meter satellite imagery represents one of the most important definitions of high-resolution Earth observation. It is closely associated with the definition of Ground Sample Distance (GSD).


Sub-Meter Satellite Imaging and GSD
Sub-Meter Satellite Imaging and GSD

What Is Sub-Meter Satellite Imaging?


Sub-meter satellite imagery means satellite images with a resolution better than 1 meter per pixel. This implies that each individual pixel on the image has a ground size of less than 1 square meter.


For example:


  • 80 cm imagery: Each pixel has a ground size of 80 × 80 cm

  • 50 cm imagery: Each pixel has a ground size of 50 × 50 cm

  • 30 cm imagery: Each pixel has a ground size of 30 × 30 cm


The smaller the size of the pixels, the better the detail level of the satellite image.


Sub-meter imagery is often used when ordinary medium-resolution satellite images do not have sufficient detail for mapping purposes.


What Is Ground Sample Distance (GSD)?


The Ground Sample Distance (GSD) is the distance on the ground represented by a single pixel in the image.


In simple terms, GSD refers to an image with a GSD of 30 cm, where each pixel represents about 30 centimeters on the Earth's surface.


In other words, GSD can be understood as:


Low GSD = low ground area covered by a single pixel = high spatial resolution.


Why Does GSD Matter in Satellite Imagery?


The value of the GSD is in its impact on the features that can possibly be identified in the image.


The bigger the GSD (5 m), the more complicated it becomes to detect features such as buildings, thin roads, or small infrastructure. When using sub-meter images, smaller features become visible.


Some examples include sub-meter imagery that allows the detection of:


  • Building outlines

  • Roads and transportation networks

  • Large vehicles

  • Construction sites

  • Utility infrastructure

  • Solar power plants

  • Wind turbines

  • Changes in surface

  • Urban expansion


However, not all the features will be identifiable because the process of identification is also affected by other parameters such as contrast, shapes, shadows, image resolution, spectral characteristics, viewing angle, and processing.


Common Sub-Meter Satellite GSD Levels


Sub-meter satellite data can usually be acquired at several resolutions:


80 cm Resolution


An 80 cm GSD can offer much greater detail compared to a 1 meter GSD. This GSD can be appropriate for general infrastructure mapping and other regional uses.


50 cm Resolution


A 50 cm GSD is commonly used for detailed mapping and visual interpretation. This GSD can be beneficial for buildings, roads, infrastructure, and cities.


30 cm Resolution


A 30 cm GSD offers a great deal of spatial detail and can be useful where detailed visual interpretation is necessary.


Applications include:


  • Urban mapping

  • Infrastructure mapping

  • Building mapping

  • Transport planning

  • Construction monitoring

  • Detailed change detection


The appropriate resolution is always a function of the needs of the specific project and not merely the minimum available GSD.


Sub-Meter Satellite Imagery for AI and Machine Learning


High-resolution satellite imagery is essential in GeoAI and geospatial machine learning.


AI models can use sub-meter imagery to detect and categorize the following features:


  • Buildings

  • Roads

  • Vehicles

  • Construction

  • Solar panels

  • Wind turbines

  • Land cover change

  • Infrastructural assets


The lower the GSD, the more spatial details the AI model can get. Yet higher resolution does not necessarily ensure better performance.


Factors influencing machine learning results include:


  • Training data quality

  • Accuracy of annotations

  • Consistency of images

  • Sensor type

  • Seasonality

  • Atmospheric effects

  • AI model design

  • Feature size


When performing satellite image analysis on a massive scale, it is necessary to consider the resolution, processing cost, data storage needs, and other factors.


Sub-Meter Satellite Imagery Applications


Urban Planning


High-resolution satellite images can assist in urban growth analysis, transportation network evaluation, building density measurement, and land use assessment.


Infrastructure Monitoring


Sub-meter images can aid in the monitoring of road networks, railroads, oil/gas pipelines, power networks, construction sites, and any other major infrastructure.


Construction and Development


Repeating satellite images can be helpful in monitoring the progress of construction and any other changes taking place on construction sites.


Environmental Monitoring


High-resolution images can aid in vegetation monitoring, coastal zone monitoring, land cover assessment, habitat analysis, and environmental change detection.


Disaster Response


After any natural disaster such as floods, earthquakes, storms, and others, high-resolution satellite images can assist in damage and infrastructure identification.



Even though in most agricultural applications high resolution is less valuable than multispectral resolution and revisit frequency, high-resolution satellite images can still assist in field boundary delineation and land use assessment.


The Future of Sub-Meter Satellite Imagery


Innovation in the fields of sensors, onboard computation, cloud computing, artificial intelligence (AI), and automated image processing will continue to add value to high-resolution Earth observation.


The future workflow will consist increasingly of:


  • High-resolution optical images

  • Synthetic aperture radar (SAR)

  • Multispectral images

  • Hyperspectral images

  • LiDAR

  • Drone images

  • AI-driven feature extraction

  • Cloud-native geospatial processing


Not using only one type of data but rather different types of sensors together is becoming a common practice for building more comprehensive geospatial intelligence.


Sub-meter satellite imagery will continue to be a part of this system, especially in cases where high-resolution, repeatable observations of extensive territories are required.


Sub-meter satellite imaging is characterized by spatial resolution better than one meter and can be used to obtain more details on buildings, infrastructure, transportation networks, land use, and other features of the Earth's surface.


Ground Sample Distance (GSD) is an important parameter that allows us to estimate the level of detail provided by this kind of satellite image. The lower the GSD, the smaller the area of the Earth's surface corresponds to a single pixel.


However, GSD cannot be considered in isolation from other parameters. The image quality, its positional accuracy, spectral properties, revisitation period, processing time, and project requirements are no less important.


The best way to select satellite imagery is to find the correlation between the size of objects that need to be analyzed and GSD.


For contemporary cartography, GeoAI, infrastructure and urban analysis, and earth observation applications, the correlation between sub-meter satellite imaging and GSD should be known.


For more information or any questions regarding Sub-Meter Satellite Imaging, please don't hesitate to contact us at


USA (HQ): (720) 702–4849

India: 98260-76466 - Pradeep Shrivastava

Canada: (519) 590 9999

Mexico: 55 5941 3755

UK & Spain: +44 12358 56710


Comments

bottom of page