Key Definitions
Spatial Resolution: This refers to the smallest level of detail that can be captured in a satellite image. Higher resolution often correlates with finer detail, which is crucial for precise applications. Resolution in this context refers to the ground area size represented by a single pixel in a satellite image. (Note that other types of resolution are associated with satellite imagery, such as temporal or spectral.)
Relative Accuracy: The accuracy of measurements taken between points in a satellite image in relation to one another. It helps in understanding discrepancies over varying distances and different geographic areas.
Absolute Accuracy: This measure indicates how closely a measured point in a satellite image aligns with its actual geographic location on the Earth's surface.
Root Mean Square Error (RMSE): A statistical measure used to evaluate the accuracy of a model or a set of measurements. RMSE quantifies the difference between the actual (true) values of elevations or heights on the Earth's surface and the values measured by the satellite. In other words, a lower RMSE indicates higher accuracy, meaning that the satellite's measurements closely match the actual elevations. This metric is crucial for satellite topographic mapping, where precise height data is essential.
Linear error (LE): Refers to the discrepancy or deviation between an object's actual position on the Earth's surface and the position reported by satellite measurements. This discrepancy can happen due to various factors, including satellite positioning errors, atmospheric conditions, and inaccuracies in the signal processing.
PhotoSat lists many of these metrics in our surveys and reports, enabling engineers to understand the data presented comprehensively. Knowing the margin of error allows engineers to make well-informed decisions with greater precision.
How Stereo Topography Works
Understanding the mechanics of stereo photos is essential to grasping the concept of accuracy in this application. Satellites orbit the Earth, capturing images of the same geographical area from different angles. With two or more stereoscopic pairs of images, it is possible to produce ground elevation measurements.
Stereo satellite photos are particularly valuable for topographic mapping, as they allow for the reconstruction of the Earth's surface in three dimensions. By comparing two images taken from slightly different positions, it is possible to discern variations in terrain elevation, such as mountains, valleys, and plains. This depth information enables engineers to create highly accurate digital elevation models (DEMs), which are essential for various applications in operating mines.
PhotoSat uses proprietary deep learning technology, based on geophysical data processing systems, to produce highly accurate elevation surveys. Final corrections can be applied to the survey and orthophoto data using surveyed ground control surveys.
Orthorectification
The accuracy and true resolution of a satellite photo are affected by several geometrical factors, such as its look angle, or the angle between the satellite’s viewing direction and the point directly below it. While flat terrains may present minimal discrepancies in elevation, areas of different topography may experience significant shifts, often leading to inaccuracies of up to hundreds of meters if left uncorrected.
The process of orthorectification becomes critical here. Orthorectification is the process of correcting geometric distortions in satellite imagery using accurate topographic data from other verified sources.
Accuracy in Satellite Monitoring
When discussing accuracy, it is essential to distinguish between absolute and relative accuracy. Absolute accuracy pertains to the precision of a point on a map compared to its true geographic location on the ground. For instance, if a satellite image indicates the position of a mine site, absolute accuracy determines how closely that indicated position aligns with the actual GPS coordinates of the site.
In contrast, relative accuracy focuses on how coordinates within the same survey area compare to one another. This concept can be particularly relevant in expansive mining sites where distances can span kilometres. As engineers evaluate how these points correlate, understanding absolute and relative accuracy becomes crucial in ensuring the integrity and reliability of the data collected.
By understanding these critical parameters in satellite monitoring, mining professionals can enhance operational efficiency, prioritize safety, and improve the overall sustainability of their projects.
The Importance of Ground Control
When capturing satellite images or data, the raw output often contains inherent positioning inaccuracies. Note that even without ground control, the internal geometry within the satellite imagery can support high relative accuracy. However, to ensure the most accurate results, it is best practice to establish high-precision ground control points, which serve as reference markers for calibrating or registering data. This serves as a critical standard in the industry, providing a framework for achieving reliable and consistent results.
We look for specific reference points, denoted by “X” or “plus” marks on a photo. If the location of an “X” in the captured image does not align perfectly with our intended ground control point, we make necessary adjustments to realign it accurately. This process of horizontal matching sets the stage for the subsequent calibration of vertical components.
Once the correct (horizontal) position of our control points is established, we work closely with surveyors to obtain elevation data. This data enables us to compare our calculated survey points with the established benchmarks, identifying any vertical discrepancies that require correction.
PhotoSat has conducted extensive accuracy studies, yielding reliable data that support an average 15-centimetre RMSE in vertical accuracy.
Putting it all together
Grasping the intricacies of accuracy, resolution, and orthorectification allows mining engineers to unlock a powerful advantage in managing their sites, including tailings facilities. These crucial data points empower them to make more informed decisions, leading to improved outcomes in diverse applications, from optimizing the design and construction of new tailings facilities to ensuring the stability of existing ones and enhancing overall operational efficiency and environmental safety.
PhotoSat provides satellite surveying for the entire mine site lifecycle, from exploration to closure. Our expertise is rooted in over 70 years of combined engineering and geological knowledge, as well as proprietary deep-learning processes. Contact us to discover how we can assist with the management of your tailings storage facility.
