报告题目：Deep Denoising Prior-Aided Low-Rank Quaternion Color Image Restoration

专家姓名：何亮田，安徽大学

报告时间：2025年10月3日 11:00-12:00

报告地点： 格物楼3306室

报告摘要：Low-rank quaternion matrix representation has emerged as a powerful tool for color image restoration, demonstrating significant advantages in preserving the inherent correlation among color channels. While deep learning-based approaches often achieve superior restoration performance, they suffer from notable drawbacks, including a heavy reliance on large-scale training datasets, high computational costs, limited flexibility, and a lack of interpretability. To bridge the gap between model-driven and data-driven methods, this talk introduces a novel framework that embeds a deep denoiser as a powerful prior into a low-rank quaternion optimization model. This hybrid approach synergistically combines the interpretability and robustness of the model-based method with the expressive power of deep learning. Experimental results demonstrate that our proposed method achieves remarkable restoration quality, effectively leveraging the strengths of both paradigms. Finally, we will outline promising research directions for future work in this domain.

专家简介：安徽大学数学科学学院副教授，硕士生导师。于2018年在电子科技大学获取理学博士学位，主要研究方向是图像处理反问题的模型构建与优化算法研究。已在IEEE Transactions on Image Processing (TIP)、IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)、Information Sciences、Expert Systems with Applications等国际权威期刊，以及多媒体领域顶级会议ACM Multimedia (ACM MM)上发表高水平论文20余篇。作为项目负责人，主持了国家自然科学基金青年项目1项，并承担了多项安徽省科技厅、教育厅资助的科研项目。

初审｜缪吉飞

复审｜鲁学伟

终审｜杨汉春