A Novel Method for Motion Blur Detection and Quantification Using Signal Analysis on a Controlled Empirical Image...

Area of Science:

• Computer Vision
• Image Processing
• Motion Analysis

Background:

• Motion blur significantly degrades image quality in single-frame imaging.
• Quantitative validation of motion blur is challenging due to the lack of ground-truth motion parameters in real-world images.

Purpose of the Study:

• To present an interpretable, measure-first framework for detecting, localizing, and quantifying motion blur.
• To enable quantitative validation of motion blur by estimating motion parameters from images.
• To operate under a validated condition of one-dimensional horizontal uniform motion.

Main Methods:

• The framework analyzes image rows as 1D spatial signals.
• It employs Movement Artifact Position Estimation (MAPE) using scanline self-similarity.
• Reference Origin Point Estimation (ROPE) and Movement Artifact Quantification (MAQ) are used for localization and blur magnitude summarization.

Main Results:

• MAPE achieved 70-90% detection rates across velocities.
• ROPE localized reference origins with 97-99% detection accuracy.
• An empirical mapping from MAQ to velocity achieved R² = 0.9900, enabling calibrated velocity estimates.

Conclusions:

• The developed framework provides a reliable method for quantifying motion blur and estimating velocity in controlled environments.
• The study clarifies the empirical boundaries of the current controlled single-marker regime, especially concerning additive noise.
• The approach is validated for one-dimensional horizontal uniform motion and does not claim generalization to complex scenarios.