Computational imaging is a multidisciplinary field that combines optics, signal processing, and mathematical algorithms to enhance the capabilities of imaging systems. Traditional imaging techniques often face limitations in terms of resolution, signal-to-noise ratio, and speed. Computational imaging approaches address these challenges by leveraging advanced algorithms to reconstruct high-quality images from incomplete or indirect measurements.

The emergence of deep learning (DL) has revolutionized computational imaging, exhibiting remarkable performance in tasks such as image reconstruction, classification, object detection and segmentation, and super-resolution. By leveraging the power of deep learning, computational imaging systems have witnessed significant improvements in terms of image quality, processing speed, and overall versatility. The integration of deep learning into computational imaging has opened up exciting possibilities across various applications.

The objective of this research topic is to highlight the most recent advancements, applications, and challenges in deep learning-enhanced computational imaging. It aims to provide a comprehensive overview of the state-of-the-art techniques, methodologies, and applications that harness the power of deep learning to enhance computational imaging performance. This includes the deep learning model for image preprocessing, reconstruction and analysis, the newly developed deep learning model, and the applications of deep learning-based techniques.

We invite researchers to submit research articles and reviews focusing on, but not limited to, the following topics:

1. DL-based method for addressing ill-posed reverse problems
2. Application of DL techniques in various aspects of image analysis, such as denoising, super-resolution, image classification, segmentation, registration
3. Reinforcement learning to optimize imaging acquisition protocols
4. Novel approaches that Integrate physical knowledge into neural network model, enhancing the interpretability and performance of DL-based methods
5. Advancements in DL architectures for improved computational imaging outcomes, such as transfer learning, continual learning, and self-supervised learning
6. Performance analysis of DL-based method, including assessments of uncertainty and generalization capabilities.

Submit your idea

Impact

Advanced Optical Technologies

Submit your idea

Impact