In nailfold capillary imaging, tremors cause image blurring and prevent accurate measurement, while capillary bundle distributions can diminish vascular information. As a comprehensive solution, we combined the YOLOv8 detection model with a Laplacian clarity evaluation function for real-time tracking of focus changes during capillary imaging. Normalized cross-correlation template matching was used to calculate the optimal matching position of adjacent frames, determine the position with the highest vascular similarity, and obtain the optimal offset between frames for accurate registration. Compared with the original video, mean squared error (MSE) decreased by 64.5%, peak signal-to-noise ratio (PSNR) increased by 3.88 dB, and structural similarity index (SSIM) increased by 4.7% in zooming experiments. MSE decreased by 57.9%, PSNR increased by 3.99 dB, and SSIM increased by 3.19% in fixed-focus experiments. Compared with state-of-the-art registration algorithms, our proposal provides better overall performance across three evaluation indicators, achieving the best registration effect for capillary images.

In nailfold video recordings, the micro-shaking of the hand is amplified and interferes with physician observations and parameter measurement. We developed a fast and accurate registration method for large-field-of-view nailfold video images. Nailfold videos are first represented in the YCrCb color space, with the Cb spatial component replacing the original grayscale image to reduce sensitivity to illumination. The projection variance of each row/column is employed to improve registration accuracy and processing speed. The method was compared with Origin GrayDrop, feature point matching, DUT, and Adobe Premiere Pro in terms of the peak signal-to-noise ratio, structural similarity index, and mean squared error. The peak signal-to-noise ratio and structural similarity index are enhanced, and the mean squared error is reduced compared to the original projection method. The proposed method is faster than the comparison methods and provides the best combination of registration accuracy and fast processing for nailfold video image registration.