講座名稱:Physics-Inspired Convolutional Neural Network for Solving Inverse Scattering Problems
講座時間:2019-09-24 9:00:00
講座地點:北校區(qū)西大樓III-412
講座人:陳旭東
講座人介紹:
陳旭東,IEEE電磁學(xué)會會士,分別于1999年和2001年獲得中國浙江大學(xué)電氣工程學(xué)士和碩士學(xué)位,2005年獲得美國麻省理工大學(xué)劍橋分校博士學(xué)位。自2005年起,他一直在新加坡國立大學(xué)工作。他發(fā)表了150篇期刊論文,根據(jù)ISI科學(xué)網(wǎng)(SCI)其總引文4900+。他撰寫了《電磁逆散射的計算方法》(Wiley IEEE,2018)一書。他的研究興趣主要包括電磁逆散射、傳感和數(shù)據(jù)融合、光學(xué)/紅外/微波掃描顯微鏡、光學(xué)加密和超材料。他在各種會議上組織了20多次關(guān)于逆散射和成像的會議。他是10余次會議組織委員會的成員,擔(dān)任總主席、TPC主席、獎勵委員會主席等。自2015年以來,他一直是《IEEE Transactions on Microwave Theory and Techniques》的副主編。陳博士是2010年國際科學(xué)聯(lián)合會青年科學(xué)家獎和IEEE ICCEM會議最佳論文獎的獲得者。
講座內(nèi)容:
The talk aims to solve a full-wave inverse scattering problem (ISP), which is a quantitative imaging problem, i.e., to reconstruct the permittivities of dielectric scatterers from the knowledge of measured scattering data. This talk proposes the convolution neural network (CNN) technique to solve full-wave ISPs. In order to make machine learning more powerful, a deep understanding of the corresponding forward problem is important. In solving ISP, the concept of induced current plays an essential role in the proposed CNN technique, which enables us to design the architecture of learning machine such that unnecessary computational effort spent in learning wave physics is minimized or avoided. Numerical simulations demonstrated that the proposed CNN scheme outperforms a brute-force application of CNN. The proposed deep learning inversion scheme is promising in providing quantitative images in real time.
主辦單位:發(fā)展規(guī)劃部