Bit depth in geospatial imagery refers to the number of bits used to represent each pixel’s color or intensity value in the image. Bit depth determines the range of possible values each pixel can have, influencing the image’s detail, quality, and analysis you can perform.
-
Unsigned integer
An integer data type that can represent only non-negative whole numbers.
-
Signed integer
An integer data type that can represent both positive and negative whole numbers.
-
Floating point
A numeric data type containing a decimal that can represent very small and very large numbers.
Each pixel in a given band can store up to 28 values. It uses unsigned integers, which range from 0 to 255.
The scale of colors aligns with how humans perceive them, so it’s used for optical imagery, such as RGB images and previews. It’s suited for visualization, mapping, and image detection algorithms.
Each pixel in a given band can store up to 216 values. It typically uses unsigned integers, which range from 0 to 65,535. It can also use signed integers, which range from -32,768 to 32,767.
Depending on the sensor, the image can have an effective dynamic range of 12-bit or 14-bit.
It’s used for multispectral and SAR imagery, as well as for some elevation models. It’s suited for visual classification and analysis.
Each pixel in a given band can store up to 232 values. It typically uses floating point values.
It’s used for SAR imagery and elevation models. It’s suited for advanced analysis which require high precision or complex numbers.
In a raster dataset, a NoData value marks pixels where data is missing or inapplicable. It helps identify and exclude these pixels during visual representation or analysis. It’s assigned to a pixel value that isn’t used as a valid value elsewhere in the dataset.