Key metrics:

  • Observed & real relative luminance
  • Contrast-to-noise ratio (CNR)
  • Tonal contrast gain (TCG)
  • Contrast dynamic range (CDR)

This setup can saturate any sensor!

The image sensor sector is booming, especially in new markets such as autonomous cars. Such applications require higher performance — for example, a very high dynamic range. However, obtaining repeatable and rigorous measurements of these new capabilities is very complex.

We at DXOMARK follow these improvements by participating on scientific committees such as IEEE and ISO. In close collaboration with the industry and the P2020 committee, we have developed a dynamic range measurement setup that can measure 150 dB or higher in one frame, thus meeting the requirement of the standard.

Our setup consists of 4 light panels whose intensity and spectrum can be controlled separately. High-quality test patterns complement the light panels and allow for high-precision measurements. The brightest panel can reach 1.5 million cd/m², saturating all sensors currently on the market.

General specifications:

  • Size: 1200 x 1200 x 800 mm​
  • Power supply: 100–240 V, 50–60Hz
  • Total maximum power: 1800 W

Technical specifications:

  • 4 separated light panels
  • Dynamic range: 0.002 to 1 500 000 cd/m²​
  • Max luminance: adjustable between 40 000 cd/m² and 1 500 000 cd/m²
  • CCT: 5600 K
  • CRI > 95

Key features:

  • Up to 177 dB
  • Flicker-free (DC-driven)
  • Computer control​ (DMX or Arnet)
  • Full automatic measurements
  • Support for RAW images up to 32 bits/pixels
  • Reference: AHDR150_001

Measurement descriptions

Observed & real relative luminance:

To check lens limitations due to glare, this metric shows the sensor linear response compared to the true luminance of the chart.

Contrast-to-noise ratio (CNR):

CNR is computed on all successive patches in the image captured by the camera, and is compared with the limit under which the camera cannot see (0 dB).

Tonal contrast gain (TCG):

TCG measures the contrast fidelity. In the case of nonlinear images, this metric is a better indicator about the quality of contrast reproduction.

Related industries