It is important to note that the gain parameter defined within the EMVA1288 standard is not to be confused with the gain found within the ‘analog to digital’ conversion process. The ratio between the analog signal value to digital grey scale value is commonly known as Gain and this value is measured in electrons per ADU. This measurement is referred to as the Signal, and the signal measurement in Figure 1 is represented by an arrow gage, while the error associated with this measurement is referred to as Read Noise or Temporal Dark Noise.įinally, Grey Scale is calculated by converting the signal value (in electrons) into a 16-bit Analog to Digital Units (ADU) pixel value. Once the pixel has finished collecting light, the charge in the well will be measured. Should the well receive more electrons than the saturation capacity, it will fail to store any further electrons. The number of electrons that may be stored within the well is referred to as the Well Depth or Saturation Capacity. This is because 3 electrons are generated when 6 photons “fall” on the sensor.Įlectrons are stored within the pixel prior to digitization. The sensor shown in Figure 1 has a QE of 50%, for example. The ratio of electrons generated during digitization to photons is referred to as Quantum Efficiency (QE). This article presents the measure of the efficiency of the conversion rather than focusing on how sensors achieve this, however. Pixel Size has a non-linear effect on the sensor’s ability to collect light because this must be squared to ascertain the light-sensitive area.ĭigitizing light involves a number of steps, the first of which is to convert the photons to electrons. Therefore this article will consider the number of photons to be a combination of exposure time and light intensity.

The number of photons observed by a pixel is dependent on exposure times and light intensity.

This specific type of noise is known as Shot Noise. The physics of light dictates that noise observed in the intensity of light is equivalent to the square root of the number of photons generated by the light source in question. Because a light source will generate photons at random intervals, there will be a degree of noise in the perceived intensity of the light. Light is comprised of discrete particles (photons), which are generated by a light source. It is important to first understand the noise inherent in the light itself. It will summarize key considerations around how an image sensor converts light into a digital image and ultimately dictates the sensor’s performance.įigure 1 uses the example of a single pixel to highlight these concepts. The first section of this article will explore the different aspects of the imaging performance of an imaging sensor. The standard also outlines how these measurements should be performed and how the results should be presented to ensure this is done in a unified manner. This article aims to answer these questions, outlining how best to compare and select cameras based on imaging performance data while following the EMVA1288 standard.ĮMVA1288 is a standard that defines which aspects of camera performance should be measured to ascertain its effectiveness. For example, how is signal to noise ratio different from dynamic range? What is quantum efficiency, and should this be measured at the peak or at a particular wavelength? The first step in this process is understanding what these various measurements mean. Comparing imaging performance of cameras such as temporal dark noise, quantum efficiency and saturation capacity is much more difficult, however. It is relatively straightforward to compare basic camera specifications such as interface, frame rate and resolution.

Camera calculate dark noise