主講人：黃忠億 清華大學(xué)教授

時(shí)間：2025年11月21日14：30

地點(diǎn)：三號(hào)樓301室

舉辦單位：數(shù)理學(xué)院

主講人介紹：黃忠億，清華大學(xué)數(shù)學(xué)科學(xué)系長(zhǎng)聘教授、博士生導(dǎo)師，從事計(jì)算數(shù)學(xué)與科學(xué)工程計(jì)算方面的研究。2020年獲國(guó)家杰出青年基金資助，2013年獲優(yōu)秀青年基金資助。在多尺度數(shù)學(xué)物理問(wèn)題的建模、分析和數(shù)值模擬等方面取得了一系列重要?jiǎng)?chuàng)新性成果，并成功應(yīng)用于材料科學(xué)、流體力學(xué)、圖像處理、金融數(shù)學(xué)、人工智能、信息論等領(lǐng)域。在 Mathematics of Computation, Numerische Mathematik, SIAM 系列雜志, Journal of Computational Physics 等國(guó)際頂尖雜志和IEEE等國(guó)際會(huì)議上發(fā)表論文百余篇，受到國(guó)際同行好評(píng)。

內(nèi)容介紹：Traditional image segmentation methods, such as variational models based on PDEs, offer strong mathematical interpretability and precise boundary modeling, but often suffer from sensitivity to parameter settings and high computational costs. In contrast, deep learning models such as UNet—which is relatively lightweight in parameters—excel in automatic feature extraction but lack theoretical interpretability and require extensive labeled data. To harness the complementary strengths of both paradigms, we propose Variational Model based Tailored UNet (VM_TUNet), a novel hybrid framework that integrates the phase field model with the deep learning backbone of UNet which combines the interpretability and edge-preserving properties of variational methods with the adaptive feature learning of neural networks. Experimental results on benchmark datasets demonstrate that VM_TUNet achieves superior segmentation accuracy and dice score compared to traditional deep learning methods, particularly in challenging scenarios which requires fine boundary delineation.