The capability of a machine vision camera to capture the details of a scene is defined by several parameters with dynamic range at the top of the list. High contrast images require a high dynamic range. One problem is there can be different ways to calculate dynamic range, which makes it difficult to compare cameras and sensors on paper. Also, dynamic range and the signal to noise ratio (SNR) are sometimes considered interchangeable for CCD and CMOS image sensors and cameras providing further confusion.
Dynamic range is the ratio between the maximum output signal level and the noise floor at minimum signal amplification (noise floor which is the RMS (root mean square) noise level in a black image). The noise floor of the camera contains sensor readout noise, camera processing noise and the dark current shot noise. Dynamic range represents the camera’s ability to display/reproduce the brightest and darkest portions of the image and how many variations in between. This is technically intra-scene dynamic range. Within one image there may be a portion that is in complete black and a portion that is completely saturated.
It is expressed in dBs (decibels). The largest possible signal is directly proportional to the full well capacity of the pixel where the full well capacity is the maximum number of electrons per pixel. Therefore, dynamic range is the ratio of the full well capacity and the noise floor.
This should be clarified as the “useful full well capacity.” Camera designers cannot always use the full well capacity of the image sensor if they want to maintain linearity also not all pixels saturate at the same level. If the full well capacity of the sensor is used instead of the usable full well value, this can provide an artificially high dynamic range specification and is something to be aware of.
Dynamic range is not equal to digitization level such that a camera with a 12-bit A/D converter does not necessarily have 12 bits of dynamic range because this does not consider the noise. The causality is reversed; if a camera has 12-bit of dynamic range that the A/D converters need to be at least 12 bits as well and preferably higher.
Dynamic range is also not the same as signal to noise ratio even though they are both expressed in dBs. The signal to noise ratio is simply the ratio of the signal level to the noise level. Since the absolute noise level depends on the average charge and the PRNU (photo response non-uniformity) on the signal level so does the SNR. Therefore, SNR is the ratio of the average signal level to the rms noise level.
The SNR for higher signal levels is dominated by shot noise. The maximum SNR is obtained at full-scale output. Note that this SNR cannot be measured in practice since the noise becomes clipped near full-scale output.
Dynamic range can be measured and calculated using the photon transfer curve if desired. For more information on the Photon Transfer Curve, click here.
Dynamic range provides a much more useful indication (compared to SNR) regarding the ability of the camera to provide the desired image details. When comparing dynamic range values from different cameras, be sure to verify they were measured under the same conditions.
For more details on how to compare cameras: