Signal-to-noise ratio (SNR) and dynamic range are related sensor-performance measures, but they are not interchangeable. Dynamic range is not replaced by SNR; SNR is one of the quantities used to characterise the usable signal range. The term “Dynamic Range 2.0” is not an established industry standard. This guide treats it only as an editorial label for a practical, system-aware discussion of sensor measurement.
For a practical guide to measuring dynamic range across camera systems, see dynamic range in digital cameras.
For comparable measurements, use a recognised characterisation method and state the operating conditions. The EMVA 1288 Standard for Characterization of Image Sensors and Cameras defines measurement procedures and quantities including temporal dark noise, sensitivity threshold, saturation capacity, SNR and dynamic range.
Signal-to-noise ratio
SNR expresses the ratio between a measured signal and the noise associated with that measurement. The exact definition depends on the signal, noise estimate, measurement domain and test method. For an image sensor, relevant noise sources include photon shot noise, temporal dark noise, quantisation noise and spatial non-uniformities. Processing such as demosaicing, denoising, sharpening and compression can change the noise characteristics of the output image.
For the relationship between ISO, gain and read noise, see our guide to camera ISO performance.
Always state whether SNR is measured on raw sensor data, a processed image, a region of interest or a task-specific output. Values measured at different stages of the imaging pipeline should not be compared as if they were equivalent.
Dynamic range and its measurement
Under EMVA 1288, dynamic range is defined as the ratio between the saturation signal and the absolute sensitivity threshold. It is commonly expressed as a ratio, in decibels or in bits. The result depends on the standard’s measurement definitions and the camera operating point. It should not be replaced by an arbitrary SNR threshold chosen after the measurements.
Dynamic range in photographic practice may also be discussed in exposure stops, but a meaningful comparison still needs a declared method, threshold and processing stage. Manufacturer figures obtained using different methods or modes may not be directly comparable.
What changes the measured result?
- Exposure and illumination: photon statistics affect SNR, while saturation limits the upper end of the usable range.
- Readout mode and gain: conversion gain, read noise and analogue-to-digital conversion can change the response.
- Temperature and exposure duration: dark current and its noise contribution can become more significant under some conditions.
- Processing: spatial and temporal denoising can improve apparent noise while changing detail and noise correlation.
- Operating mode: binning, cropping, frame rate and sensor readout settings may change the effective performance.
Repeatable sensor-characterisation workflow
- Record camera and sensor model, firmware, lens or optical configuration, pixel format, bit depth, gain, exposure time and temperature.
- Follow the selected standard’s dark-frame and uniform-illumination procedures.
- Capture sufficient frames and exposure levels to estimate signal response, temporal noise and saturation behaviour.
- Record the analysis software, calibration settings, region selection and any excluded measurements.
- Report the measured quantities and operating conditions together; do not detach a single headline number from its test context.
- Repeat the measurement after material firmware, sensor-mode or processing changes.
Interpreting SNR in computational workflows
Multi-frame denoising and HDR fusion can improve the appearance or measurable SNR of a processed output under suitable conditions. The result depends on frame alignment, motion, exposure strategy and the processing pipeline. It does not change the sensor’s intrinsic characterisation result under the standard measurement conditions. Keep sensor characterisation separate from the performance of a complete computational imaging system.
For a practical low-light example, see our article on computational night photography.
Conclusion
SNR helps describe noise relative to signal, while dynamic range describes the span between a defined upper and lower measurement limit. A reliable comparison states the standard, measurement stage, operating mode and environmental conditions. “Dynamic Range 2.0” should not be presented as a recognised standard or as a replacement for conventional dynamic-range measurement.