Concept
In the Camera Monitor System (CMS), distortion refers to the deviation of the imaging result from the ideal projection model, and its definition varies depending on the ideal reference.
To adapt to engineering applications, it is mainly divided into two categories:

• TV Distortion: Characterizes the overall curvature of the image with a single reference value, serving as a practical metric for quickly assessing the system's distortion level.
  
• Geometric Radial Distortion: Uses ideal rectilinear projection (i.e., the ideal pinhole projection where straight lines in the real scene remain straight in the image) as the reference. It is quantified by calculating the ratio of the radial height deviation Δh to the ideal value $h_{ideal}$, which can precisely characterize the local distortion properties in different field of view regions of the CMS.
  

This article mainly introduces the test method for CMS geometric radial distortion.

Calculation Method:

Distortion schematic diagram (1 monitor border; 4 image center; 7 ideal projection distance; 8 actual projection distance; 9 deviation between actual and ideal projection height)

The calculation formula for geometric radial distortion is:

$\text{RadialDistortion} = \frac{\Delta h}{h_{ideal}} = \frac{h_{monitor} - h_{ideal}}{h_{ideal}} = \frac{h_{monitor}}{h_{ideal}} - 1$

where: $h_{monitor}$ is the measured radial height of the image; $h_{ideal}$ is the unobservable theoretical radial height of the image starting from the optical center.

Measured image radial height $h_{monitor}$:

$h_{monitor}(i) = \frac{1}{2}\left[ w_{monitor}(i) - w_{monitor}(i_{centre}) \right]$

where: $w_{monitor}(i)$: the pixel width of the checkerboard at target point $i$ in the image; $i_{centre}$: the image center (position aligned with the optical axis).

Ideal image radial height $h_{ideal}$:
Based on the linear characteristics of ideal rectilinear projection, a least squares linear fitting is performed on the measured $h_{monitor}$ data to obtain the ideal linear growth curve. The values of the corresponding points on this curve are $h_{ideal}$ (unobservable theoretical values).

Standard Specifications: (GB 15084 4.3.2.12)
For Class I, II, and III CMS, the maximum distortion within the specified minimum field of view, relative to linear or pinhole projection, shall be no greater than 20%.

Test Method:
Test Environment:

Equipment layout reference	Actual scene on the monitor side

Camera side: Light source color temperature 6500K±1500K, illuminance 800±80lx;
Monitor side: Dark environment with illuminance below 10lx.

Main Equipment Used:
Automated Chart Switching Bracket Test System, Multi-CCT LED Fill Light Source (Visible and NIR), Reflective Checkerboard Test Chart, 2D Imaging Luminance Meter, RIQA-CMS Image Quality Analysis Software, Monitor and Luminance Meter Mobile Stand

Operating Procedures:
Camera Side:

1. Switch the chart to the checkerboard;
2. Turn on the chart fill light source and adjust the light source parameters so that the illuminance on the chart is 800lx and the color temperature is 6500K;
3. Adjust the camera so that its optical axis is perpendicular to the chart;
4. Adjust the distance between the camera and the chart so that the squares fill the monitor screen symmetrically;
5. Use a laser rangefinder or tape measure to measure the adjusted distance between the camera and the chart.

Monitor Side:

1. Ensure the reference lens of the imaging luminance meter is perpendicular to the monitor;
2. Turn off other light sources so that the ambient illuminance on the monitor side is below 10lx;
3. Capture images using the 2D imaging luminance meter.

Analysis and Interpretation:
Use the RIQA Software (an image quality software independently developed by Yanding) to analyze the test samples.
The measured distortion rate is <20%, therefore, the geometric distortion of this sample meets the requirements.