An effective method for detecting dorsal hand veins utilising near-infrared imaging technology
Keywords:adaptive thresholding, feature extraction, image processing, near-infrared, vein detection
Intravenous access for blood collection and other related therapies is one of the most frequently practiced procedures in the modern medical system. The procedure requires complex training and experience, as it might cause dangerous nerve damage and subcutaneous bleeding. This paper proposes a dorsal hand vein detection method utilising the near-infrared (NIR) imaging device to segment and visualise the subcutaneous vein patterns on the skin directly. Applying NIR light has received substantial attention because of its non-invasive and revealing substantially more information than the visible one. The proposed method is divided into the low- and high-level processes. The captured image is smoothed and enhanced to make the vein patterns clearer in the low-level process. The pre-processed image is then segmented step by step to extract the vein features and eliminate the pseudo-vein regions precisely. Lastly, the detected veins are thinned to reduce the thickness and projected back onto the acquired image in the high-level process. The proposed method performs effectively in detecting the clear dorsal hand veins through the experiment with a processing time of 0.61s for the high-resolution image.
Received 23 April 2019; accepted 26 June 2019
How to Cite
This work is licensed under a Attribution-NonCommercial-NoDerivatives 4.0 International