Documentation – Pilot v25.2

Image Coordinate Systems

About

Different software/programming languages may use different image coordinate systems [1, 2]. Although there are many possible image coordinate systems, four common ones have been defined within IEEE Std 2020-2024 [2]. The labeling of the image coordinate system facilitates interfacing between software/programming languages that use different image coordinate systems.

Factors

The following are factors of an image coordinate system.

Factor Description Example(s)
Axis Ordering The ordering of the pair of coordinates

x, y (row, column); y, x (column, row)

Reference Pixel Index The index of the reference pixel

0; 1

Reference Pixel Location The location of the reference pixel

Lower Left; Upper Left; Other

Positive x-axis Direction The direction within the image that the x-coordinate increases

Left; Right

Positive y-axis Direction The direction within the image that the y-coordinate increases

Up; Down

Integer Pixel Location The within-pixel location of the integer index

Pixel center; Other

Downsampling The amount of downsampling performed

None; 2×2; Other

Cropping  The amount of cropping performed

No Cropping; Cropping

Use within Imatest

Unless otherwise specified, Imatest uses an IEEE Std 2020 [2] Type IV image coordinate system.

Image Coordinate Systems

  IEEE Std 2020 [2] Type I IEEE Std 2020 [2] Type II IEEE Std 2020 [2] Type III IEEE Std 2020 [2] Type IV
Axis Ordering x, y (row, column) x, y (row, column) x, y (row, column) x, y (row, column)
Reference Pixel Index 0 0 1 1
Reference Pixel Location Lower Left Upper Left Lower Left Upper Left
Positive x-axis Direction Right Right Right Right
Positive y-axis Direction Up Down Up Down
Integer Pixel Location Pixel Center Pixel Center Pixel Center Pixel Center
Downsampling None None None None
Cropping None None None None
Example Usage   OpenCV  

MATLAB*

Imatest

* The MATLAB axis ordering is column, row.

Examples

The following examples illustrate various image coordinate systems for a 6-column by 4-row image. The reference pixel is marked in gray.

  
IEEE Std 2020 [2] Type I IEEE Std 2020 [2] Type II
IEEE Std 2020 [2] Type III IEEE Std 2020 [2] Type IV

References

[1] P. Romanczyk, “Validation methods for geometric camera calibration”  in Proc. IS&T Int’l. Symp. on Electronic Imaging: Autonomous Vehicles and Machines,  2020,  pp 150-1 – 150-6,  https://doi.org/10.2352/ISSN.2470-1173.2020.16.AVM-150

[2] IEEE Std 2020-2024. “IEEE Standard for Automotive System Image Quality”. DOI: 10.1109/IEEESTD.2024.10935839