Near-infrared vascular images play an important role in the diagnosis and treatment of vascular diseases. However, near-infrared vascular images often have problems such as low image quality and unclear vascular patterns. To solve these problems, we propose a Deep Convolutional Neural Network (DCNN) auto-encoder for image enhancement to enhance vascular structures and suppress non-vascular structures. We also collect a datasets of 156 images for the training and validation testing of the model; and further we use the full-reference image quality assessment metrics, i.e., Peak Signal to Noise Ratio (PSNR) and Structural SIMilarity (SSIM) to quantitatively evaluate the image enhancement effect of this model. The experimental results show: compared with the traditional image enhancement algorithm, the enhanced image quality of the Residual Convolutional Auto-Encoder (RCAE) model is better and more similar to the original image.
Near-Infrared Vascular Image Enhancement Using Deep Convolutional Neural Network
Lect. Notes Electrical Eng.
International Conference on Man-Machine-Environment System Engineering ; 2022 ; Beijing, China October 21, 2022 - October 23, 2022
2022-08-21
7 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Deep adaptive convolutional neural network for near infrared and thermal face recognition
British Library Conference Proceedings | 2022
|British Library Conference Proceedings | 2022
|