Using a near-infrared camera to collect and analyze superficial blood vessel images of human forearm is a newly developed technical approach. However, due to individual differences, the forearm blood vessels in near-infrared image show blurred edges and weak blood vessel strength. In order to overcome the above problems, this paper uses a U-shaped symmetrical neural network structure that is more in line with the characteristics of this type of data, aiming to obtain a clear outline of blood vessels. First, image acquisition is performed using a self-designed device and a standard data set is produced. Second, we use the U-shaped constructed network for training and testing. Finally, the segmented image is evaluated using evaluation indicators such as mAcc and Dice. The experimental results show that our method can achieve effective segmentation of blood vessels with a small amount of data, providing support for the development of related medical technologies.
VU-Net: A Symmetric Network-Based Method for Near-Infrared Blood Vessel Image Segmentation
Lect. Notes Electrical Eng.
International Conference on Man-Machine-Environment System Engineering ; 2023 ; Beijing, China October 20, 2023 - October 23, 2023
2023-09-05
6 pages
Article/Chapter (Book)
Electronic Resource
English
The Near-Infrared Forearm Vessel Image Segmentation and Application Using Level Set
Springer Verlag | 2022
|Adaptive image enhancement for retinal blood vessel segmentation
IET Digital Library Archive | 2002
|Near-Infrared Vascular Image Enhancement Using Deep Convolutional Neural Network
Springer Verlag | 2022
|