A camera’s bit depth defines the number of distinct shades available for each pixel. There are many imaging applications that do not need a bit depth of more than 8 bits; however, when color accuracy is required, high bit depth is vital for detecting even the slightest deviation from specific color tones.
For example, as explained in this Lumenera case study on paint inspection applications, when inspecting paint in automotive assembly lines, high bit depth enables the industrial camera to perform the color analysis with greater accuracy.
What is Bit Depth?
The term bit depth is derived from the binary nature of data, as it is stored electronically. Each bit can contain one of two values – a zero or a one, and each extra bit unlocks twice the data, as it is an exponential relationship. The number of shades available for each pixel can be measured by taking two to the power of the bit depth.
Taking an image at a bit depth of one is the simplest example of bit depth, and an ideal example of this is black text on a white background. This data is clearly not enough to represent more complex images with different color shades.
For this reason, most images and computer monitors consist of images with a bit depth of 8 bits for each color channel, or 24 bits in total, meaning that the image is comprised of 16,777,216 or 2 (8x3) colors. Although this amount of color is acceptable in routine images, systems with higher bit depths can be used when accurate color analysis is needed.
When cameras are able to capture images at 12 and 14 bits for each channel, they can analyze much finer differences between color shades, as they are able to distinguish between more than 68 billion (68,719,476,736) colors, and 4 trillion (4,398,046,511,104) colors for 12-bit and 14-bit cameras, respectively.
This can be accomplished by sampling each pixel with finer granularity. Although the pixel does not capture more light, it has a far more extensive range of values to relate to it. A most common analogy used to explain a pixel is a bucket of water. You equate capturing light to fill the bucket. In terms of bit depth, the same analogy can continue by way of measuring the water.
A simple measuring cup would work in almost all situations, or a graduated cylinder can be used to attain much higher accuracy when measuring the quantity of water. During the analysis, the quantity of water stays the same, but you have a much better idea of how much water you have.
Most displays are not designed to display images that are more than 8 bits. Instead, the images would have to be converted from a higher bit depth down to 8 bits. In this case, the original finer granularity data will be grouped together into coarser color tones; losing the advantages of the high bit depth of the camera. Computer monitors that display images at higher bit depths are available and can be used for high-performance imaging applications.
Factors that Impact Bit Depth
Compression algorithms and image format can also influence the bit depth and other aspects of image data in an unrecoverable way.
Compression algorithms can be used to minimize the size of an image to either increase the transmission rates or decrease the space needed to store them. They usually target aspects of images that are not easily visible to the human eye. This can affect the operation of machine vision systems as important data, including bit depth, can be lost. Therefore, it is important that transmission and storage of raw data are used to ensure that no data is lost.
A USB 3.0 camera and cable will ensure sufficient bandwidth to transmit raw images at a high bit depth without affecting the frame rate.
If users are unsure about the bit depth needed for their system, they can contact Lumenera’s imaging experts, who will help them choose a camera that meets their specific application.
This information has been sourced, reviewed and adapted from materials provided by Lumenera Corporation.
For more information on this source, please visit Lumenera Corporation.