Detection of particle motion using image processing with particular emphasis on rolling motion.

Area of Science:

• Granular physics
• Image analysis
• Geotechnical engineering

Background:

• Image processing is crucial for analyzing granular systems near surfaces.
• Applications include sediment flow and bedload transport studies.
• Accurate particle detection is essential for understanding granular dynamics.

Purpose of the Study:

• To review image-processing techniques for single and multiple particle detection.
• To demonstrate the application of Canny edge and Hough transforms.
• To assess the accuracy of these methods for particle position and motion analysis.

Main Methods:

• Review of established image-processing algorithms.
• Exemplary application of Canny edge detection for feature identification.
• Utilizing Hough transform for precise particle localization and motion tracking.
• Detailed algorithmic step-by-step description.

Main Results:

• Canny edge and Hough transforms significantly enhance particle recognition reliability.
• Successful detection of both particle positions and rotatory motion demonstrated.
• Algorithm validated through application to diverse experimental examples.
• Quantitative accuracy assessment of the developed methods.

Conclusions:

• Image processing, particularly with Canny edge and Hough transforms, offers a reliable approach for granular system analysis.
• The presented methods accurately capture particle dynamics, including rotation.
• Detailed algorithmic descriptions and accuracy discussions provide a valuable resource for researchers.