Discorpy: algorithms and software for camera calibration and correction.

Area of Science:

• Optical Engineering
• Computer Vision
• Metrology

Background:

• Detector calibration is crucial for spatial accuracy in dimensional tomography, optical metrology, and computer vision.
• Existing methods often lack a balance of accuracy, robustness, ease of use, and ability to handle diverse distortions.
• Radial lens distortion is prevalent in high-resolution X-ray detector optics for synchrotron-based parallel-beam tomography.

Purpose of the Study:

• To develop and implement an approach for individually characterizing and correcting radial and perspective distortions.
• To achieve sub-pixel accuracy in detector calibration using a single optical target image.
• To provide a robust and user-friendly solution for lens distortion correction in scientific imaging.

Main Methods:

• Utilized a single calibration image to characterize both radial and perspective distortions.
• Developed algorithms to individually correct for optical center offset and lens distortions.
• Implemented the correction methods within the Discorpy software package.

Main Results:

• Successfully characterized and corrected both radial and perspective distortions from a single image.
• Achieved high accuracy in spatial measurements post-correction.
• Demonstrated the software's capability to handle common optical aberrations in detector systems.

Conclusions:

• The presented approach offers an effective solution for accurate camera and lens-based detector calibration.
• Discorpy software provides a robust tool for correcting complex distortions, enhancing spatial accuracy in various imaging applications.
• This method simplifies the calibration process while maintaining high precision for scientific imaging.