Perspective transformation based-initial value estimation for the speckle control points matching in an out-of-focus...

Area of Science:

• Computer Vision
• Metrology
• Image Processing

Background:

• Traditional camera calibration methods using chessboard or circular markers struggle with accuracy in defocused or noisy images.
• Digital Image Correlation (DIC) offers noise robustness for speckle pattern matching, making it suitable for low-quality imaging conditions.
• Accurate initial value estimation is critical for DIC speckle control point matching, but challenging with significant pixel scale differences or out-of-focus images.

Purpose of the Study:

• To develop an efficient initial value estimation method for speckle control point matching using DIC.
• To address the challenges of inaccurate initial DIC values caused by differing pixel scales and out-of-focus images.
• To enhance camera calibration accuracy and measurement precision in low-quality imaging scenarios.

Main Methods:

• A novel approach based on perspective transformation for initial value estimation in DIC speckle pattern matching.
• Detection of speckle region corners in reference and target images.
• Coarse matching of neighborhood points using fixed subset searching after transforming the target image to match the reference image's pixel scale.
• Inverse perspective transformation to obtain initial values for DIC.

Main Results:

• The proposed method demonstrates higher camera calibration accuracy compared to chessboard and circular marker methods.
• Measurement precision is improved over speckle pattern calibration methods employing Scale-Invariant Feature Transform (SIFT)-based initial value estimation.
• The method effectively handles challenges related to differing physical pixel scales and out-of-focus target images.

Conclusions:

• The perspective transformation-based initial value estimation method significantly enhances DIC speckle control point matching for camera calibration.
• This technique offers a more robust and accurate solution for camera calibration, particularly when dealing with degraded image quality.
• The findings suggest a promising advancement for metrology applications requiring precise camera calibration under adverse imaging conditions.