Purpose of the Analysis : Timely and accurate change detection of Earth’s surface features
by Cloud Ararat Team | News | 0 Comment
Timely and accurate change detection of Earth’s surface features is extremely important for understanding relationships and interactions between human and natural phenomena in order to promote better decision making. Remote sensing data from different sensors such as Earth observation satellites, manned aircrafts and UAV imagery are primary data sources extensively used for change detection over the last decade. This is an introductory, simplified imagery analysis to demonstrate spatial, spectral, radiometric and temporal changes in Earth Observation capabilities.
In this post, we will keep the context within the spatial and spectral resolution of aerial imagery.
This is a simplified and limited scope post. We are able to provide further stats, explanation and custom solutions. Please contact us to discuss your business case further. Remote sensing imagery data can be characterized in several ways:
- Spatial Resolution
- Spectral Resolution
- Radiometric Resolution
- Temporal Resolution
*** For custom applications for your business, please contact Cloud Ararat team.
There are various change detection techniques which have been developed over the past decade. Pixel-based change detection has been, and remains, an important research topic in remote sensing. When observing a common scene, a high-spatial-resolution image provides more details than a lower-spatial-resolution image. However, this increased spatial resolution generates high spectral variability within geographic objects, which typically reduces the change detection and classification accuracy when using pixel-based algorithms.
Image classification is another important process for extracting information classes, such as land cover categories or human-made objects, from multi-band remote sensing imagery or aerial imagery, and still one of the most challenging problems in understanding high-resolution remote sensing images Deep learning techniques, especially the convolutional neural network (CNN), have improved the performance of remote sensing image scene classification due to the powerful perspective of feature learning and reasoning, but traditional methods such as pixel-based and object-based classification are appropriate for most of the cases and still used in many studies.
In the above imagery, the revisit time is 2 years. Please review the circled, human made buildings and vegetation change.
Parameters to decide the suitable data type for your business depend on your ultimate goal or mission. While there are so many aerospace and computer science technologies, we observe increased misinformation in the market, leading businesses to miss the value of data driven decision making. Some of the application areas include but not limited to object detection, construction progress analysis, earth surface changes, forestation, inspections, insurance claims review…etc. Please feel free to reach out to us to discuss your business case.
• Spatial Resolution : Most people are familiar with the spatial resolution, defining the length of each side of one pixel. When these pixels are referred to the area on the Earth’s surface, it’s called Ground Sampling Distance (GSD).
• Radiometric Resolution : provides information about number of bits In a given image band spectrum.
• Temporal Resolution : Time difference of aerial imagery of a scene in two subsequent data acquisition. In aerospace, this is also known as Revisit time.
• Spectral Resolution : refers to the number of color bands that an image contains. As an example, black and white photos has 1 band while RGB has 3 bands.
* Pros : Global data coverage. Frequent revisit time as low as 30 minutes to few hours. Low cost raw imagery. Access to historic imagery. Wide spectrum of sensors (EO, IR, SAR, Multi-Spectral…)
* Cons : Lower resolution (30cm to 30 meters) compared to low altitude aircraft imagery.
Manned Aircraft Imagery
* Pros: Ability to carry very high resolution and heavy sensors. Ability to fly over cities with easier restrictions.
* Cons: Expensive compared to satellites. Low availability. Requires crew or professionals
* Pros: Ability to fly low altitude. Carry sufficient payload for missions up to 1-2 hours. Commercially available easily. Low CapEx and OpEx
* Cons: Flight regulations and restrictions on populated areas and cities. Limited options and flight time for heavy sensors such as Lidar.