A Study on the Effectiveness of Spatial Filters on Thermal Image Pre-Processing and Correlation Technique for...

Area of Science:

• Materials Science
• Non-Destructive Testing
• Image Processing

Background:

• Thermal imaging is essential for structural health monitoring, detecting defects from aging or construction faults.
• Pre-processing and post-processing techniques are critical for accurate analysis of thermal image data.

Purpose of the Study:

• To assess the efficacy of spatial filters in pre-processing thermal images for defect detection.
• To evaluate a correlation technique for post-processing thermal image sequences to determine defect size.
• To apply these methods for non-destructive testing of steel structures.

Main Methods:

• Implemented linear filters (Gaussian, Window Averaging) and a non-linear filter (Median) for pulsed thermography image pre-processing.
• Assessed filter effectiveness using signal-to-noise ratio (SNR) as a quality metric.
• Applied a correlation technique to Gaussian-filtered images for defect size determination.

Main Results:

• All tested filters reduced impulse noise and improved thermal image quality.
• The Gaussian filter demonstrated superior performance over Window Averaging and Median filters in terms of SNR.
• The correlation technique successfully determined defect size from pre-processed thermal image sequences.

Conclusions:

• Spatial filters, particularly Gaussian, significantly enhance thermal image quality for structural health monitoring.
• Correlation techniques offer a viable method for rapid defect size measurement in steel structures.
• Accuracy of defect size measurement is influenced by thermography's detection limits and defect geometry.