Hybrid quantum-classical convolutional networks for robust denoising of quantum images in noisy systems
Year: 2025
Type: article
Abstract: Quantum imaging systems produce images with distinctive noise patterns that conventional denoising algorithms cannot effectively process. We present an innovative neural network architecture that merges quantum physics principles with deep learning to address this challenge. Our hybrid approach adapts standard image processing techniques to handle quantum-specific noise while preserving critical image features. Experimental validation demonstrates a consistent 12.6% improvement in output quality compared to existing methods, with efficient performance on standard computing hardware. Additionally, the model exhibits strong generalization capabilities, achieving robust performance across varying noise levels. This advancement represents an important step toward practical quantum imaging applications in fields ranging from medical diagnostics to secure communications.
Author: Ola Al-Ta’ani
Institution: Yarmouk University
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4410260943