Adaptive Denoising and Alignment Agents for Infrared Imaging
- 8 November 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 6, 1586-1591
Abstract
Clinical infrared imagers, such as those used to visualize subcutaneous vasculature, rely on the image-projection feedback loops. Specifically, these instruments embed model-based feedback algorithms to process the acquired ’invisible’ infrared data streams and then to project a visible copy of them back to the imaged region (e.g., the skin of the patient’s forearm). Being inherently noisy, the infrared frames and their projections are prone to misalignment, demanding frequent instrument tuning and recalibration. To address this challenge, we propose to reconsider the feedback loop entailed in such imagers from the standpoint of multi-agent deep learning. Both proposed agents – Denoiser and Aligner – are adaptive in that they continuously optimize the corresponding target value functions. Namely, the Denoiser learns the proper frequency decomposition of the acquired infrared data until the target segmentation metric is maximized; whereas, the Aligner learns the intensity fluctuations within the segmentation mask tuned by the Denoiser until its maximal overlap with the source infrared image. The idea is validated synthetically on a benchmark dataset and experimentally on a bench-top vein scanner in the lab, with the duet of agents proving efficient in handling the distortions and the misalignment.Keywords
Funding Information
- RFBR (21-51-12012)
This publication has 33 references indexed in Scilit:
- Augmented reality based real-time subcutaneous vein imaging systemBiomedical Optics Express, 2016
- Brain MR image denoising for Rician noise using pre-smooth non-local means filterBioMedical Engineering OnLine, 2015
- Denoising and deblurring of Fourier transform infrared spectroscopic imaging data2012
- A randomized trial of the Vein Viewer versus standard technique for placement of peripherally inserted central catheters (PICCs) in neonatesJournal of Perinatology, 2011
- Wavelet-based medical infrared image noise reduction using local model for signal and noise2011
- FSIM: A Feature Similarity Index for Image Quality AssessmentIEEE Transactions on Image Processing, 2011
- Infrared image denoising based on stationary wavelet transform2010
- A New Contourlet Transform with Sharp Frequency Localization2006
- Infrared imaging of subcutaneous veinsLasers in Surgery and Medicine, 2004
- An overlap invariant entropy measure of 3D medical image alignmentPattern Recognition, 1999