Practicing GIS requires a careful understanding of the important distinction between precision and accuracy. Photogrammetry for dynamic geospatial data, precision and accuracy are two important concepts. Failure to understand the contradiction between precision and accuracy can have a profound negative impact on the way data is processed and the final results of geospatial analysis.
This article aims to eliminate confusion, explore the difference between precision and accuracy and the correct use of each term, and use drone photogrammetry as an example.
What is accuracy in GIS?
Accuracy: Accuracy refers to the degree to which a measurement result conforms to a correct value or standard. In GIS, accuracy is often used to describe the correctness of feature locations or attribute values in geographic data or maps.
What is precision in GIS?
Precision is defined as the "quality of accuracy" and refers to the closeness between two or more measurements, whether or not those measurements are accurate. Precision measurements can also be inaccurate. In GIS, precision is often used to describe the repeatability and reliability of a measurement tool or method.
What is the difference between accuracy and precision?
Accuracy and precision both reflect how close a measurement is to the actual value, but they are not the same. Accuracy reflects how close a measurement is to a known or accepted value, while precision reflects how repeatable a measurement is, even if they are far from an accepted value. Measurements that have both accuracy and precision are repeatable and very close to the true value.
In geospatial photogrammetry, high-accuracy and high-precision data information helps decision makers make further analysis and decisions. In GIS, accuracy is usually related to the resolution of the data. Compared with low-resolution data, high-resolution data (such as aerial images collected by drones) can make more accurate measurements and capture finer details.
Accurate and accurate, or accurate and accurate?
High-accuracy and high-precision data collection is the most critical foundation for drone photogrammetry. In actual photogrammetry of a specific object, high accuracy does not guarantee high accuracy. That is, when actually measuring and recording the position, the measurement result always deviates from the true position, but the measurement value is almost the same each time, then it is accurate, but not accurate.
Which is more important in drone photography, accuracy or precision?
Accuracy and precision are not mutually exclusive. Ideally, we strive to obtain data that is both accurate and precise. Drone photography is the process of processing images taken by drones to generate 3D models.
GPS drones help us find the exact location of objects to be measured, and drones with RTK modules help us improve the accuracy of photo shooting. When we start drone photogrammetry, drones can take thousands of pictures at a time according to a set route mission, and then stitch them together in the drone mapping software to form a 3D model.
How do drones collect accurate data?
When choosing drone photogrammetry, please give priority to Autel RTK drones, which will improve the accuracy and precision of geospatial data collection.
The Autel EVO II Pro RTK V3 drone integrates a high-resolution 1-inch CMOS camera sensor and a centimeter-level precision RTK module, which can realize RTK, PPK, and network RTK, and can achieve accurate drone mapping without ground control points. The compatibility of this drone with RTK and PPK GNSS correction systems further enhances positioning accuracy in various mapping applications.
Does drone image resolution affect data accuracy?
Drone image resolution is the precision of the screen image, which refers to the number of dots that the display can display. The spatial resolution (GSD) of drones/remote sensing satellites refers to the spatial distance represented by one pixel of aerial photos/remote sensing images.
For example, suppose there is a digital elevation model (DEM) dataset with a resolution of 10 meters, which means that each pixel in the dataset represents a 10x10 meter area on the ground. If a terrain feature (such as a hill or depression) falls within the boundaries of a single pixel, the elevation value assigned to the pixel may not accurately represent the true elevation of the feature.
However, if the dataset has a higher resolution, such as 1 meter, the terrain feature will be represented by multiple pixels, allowing a more accurate portrayal of its elevation.
In general, higher data resolution can capture finer details and nuances in spatial data, thereby improving data accuracy. However, it is important to note that data resolution alone does not guarantee data accuracy. Other factors, such as data collection methods, processing techniques, and error sources, can also affect the accuracy of geospatial data.
Related More:
Can Drone Photogrammetry Replace Traditional Ground Surveying Techniques?
