Brown-Conrady Model
Definition
The standard parametric model for camera lens distortion, using radial coefficients k1, k2, k3 and tangential coefficients p1, p2. Camera calibration tools (OpenCV, Matlab Camera Calibrator, photogrammetric software) fit these coefficients from images of a known calibration target.
- Radial terms
- k1, k2, k3
- Tangential terms
- p1, p2
- Used by
- OpenCV, Matlab Camera Calibrator
- Application
- Camera calibration, lens distortion correction
- Field
- Digital image forensics
Common questions
Why does forensic image analysis use the Brown-Conrady coefficients?+
Matching the k1, k2, k3 and p1, p2 values recovered from a suspect camera against the distortion pattern in an evidentiary image can help confirm or exclude that camera as the capture device.
Can two different cameras produce nearly identical Brown-Conrady coefficients?+
Yes, cameras sharing the same lens design and sensor produce very similar coefficients, so the model narrows the likely source device but rarely uniquely identifies one physical unit on its own.
Related terms
- Colour Filter Array (CFA)
- A mosaic of spectrally selective filters placed over a camera sensor so that each photosite measures only one colour channel. The standard...
- Demosaicing
- The algorithm used to interpolate full-colour pixel values from the single-channel raw sensor data. Common methods include bilinear interpolation, gradient-based algorithms (AHD,...
- Device-Class Identification
- Establishing that an image was captured by one of a set of cameras sharing the same sensor architecture and firmware, rather than...
- Frequency-Domain CFA Spectrum
- When a demosaiced image is transformed into the frequency domain (using a 2D DFT), the CFA interpolation introduces periodic peaks at spatial...
- Radial Distortion
- A geometric lens aberration in which magnification changes with distance from the optical axis. Barrel distortion (negative radial coefficient k1) bows straight...