Portable Hand Vein Finder System Based on Near-Infrared Illumination and Morphological Image Processing

Abstract

Vein detection technology has significant applications in many areas such as healthcare, security and aesthetics. In healthcare, vein visualization is highly necessary for making an accurate venipuncture. This work demonstrates a portable hand vein finder system using noninvasive infrared technology that can display the real-time venous image on a monitor. The skin surface is illuminated by a near-infrared (NIR) light source that contains 5 LEDs (850 nm) and 5 LEDs (940 nm). These LEDs are arranged alternately in a circular form. Under NIR light exposition, vein and surrounding tissues are distinguished due to their difference in the brightness and contrast. A 3 MP camera is used to capture the illuminated skin area, and the obtained signal is then transferred to a Raspberry Pi 4 for image processing. The input image is processed with median filtering and contrast-limited adaptive histogram equalization (CLAHE) for displaying on a monitor. After processing, venous images with much better contrast are observed. To project the venous patterns on the skin surface, local adaptive threshold and subsequent morphological image processing are used. After the first algorithm, the venous patterns are segmented out of the image. Then, morphological processing, which has rarely been widely used for vein visualization, further reduces the noise signal, such as hair on the hand thus clearer and sharper venous patterns are obtained. Our work provides an effective and low-cost solution for nursing staff in low and middle-income countries to perform a safe and accurate venipuncture.