To realize automated quality inspection and pore size measurement in both the design and manufacturing of Ω shaped grooved heat pipes, this study proposes a computer vision-based approach for quality inspection and pore size measurement. Firstly, the scale in the collected image is identified and measured based on the HSV color space. Then, the grayscale image is subjected to preprocessing which includes Contrast-Limited Adaptive Histogram Equalization (CLAHE) and Gaussian filtering. The preprocessed image is subsequently binarized according to a binarization thresholding algorithm integrating a weighted fusion of the Triangle segmentation and Otsu’s method. Finally, Hough Circle Transform is applied to detect round pores in the binary image and measure their pixel size, which are converted to physical dimensions based on the scale identified on the HSV color space. Experimental results show that the proposed method accurately recognizes and measures the scale with an average error of only 0.604%. On the test set, it achieves the pore recognition accuracy of 95.59% and an average radius measurement error of 1.97%, demonstrating its potential to replace manual inspection and measurement in the design and manufacturing of Ω-shaped grooved heat pipes.

Computer vision,Ω-shaped Grooved Heat Pipe,Quality Inspection,Size Mesurement,Binarization Threshold