Your browser doesn't support javascript.
loading
Phase imaging with an untrained neural network.
Wang, Fei; Bian, Yaoming; Wang, Haichao; Lyu, Meng; Pedrini, Giancarlo; Osten, Wolfgang; Barbastathis, George; Situ, Guohai.
Afiliación
  • Wang F; 1Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, 201800 Shanghai, China.
  • Bian Y; 2Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences, 100049 Beijing, China.
  • Wang H; 1Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, 201800 Shanghai, China.
  • Lyu M; 2Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences, 100049 Beijing, China.
  • Pedrini G; 1Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, 201800 Shanghai, China.
  • Osten W; 2Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences, 100049 Beijing, China.
  • Barbastathis G; 1Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, 201800 Shanghai, China.
  • Situ G; 2Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences, 100049 Beijing, China.
Light Sci Appl ; 9: 77, 2020.
Article en En | MEDLINE | ID: mdl-32411362
Most of the neural networks proposed so far for computational imaging (CI) in optics employ a supervised training strategy, and thus need a large training set to optimize their weights and biases. Setting aside the requirements of environmental and system stability during many hours of data acquisition, in many practical applications, it is unlikely to be possible to obtain sufficient numbers of ground-truth images for training. Here, we propose to overcome this limitation by incorporating into a conventional deep neural network a complete physical model that represents the process of image formation. The most significant advantage of the resulting physics-enhanced deep neural network (PhysenNet) is that it can be used without training beforehand, thus eliminating the need for tens of thousands of labeled data. We take single-beam phase imaging as an example for demonstration. We experimentally show that one needs only to feed PhysenNet a single diffraction pattern of a phase object, and it can automatically optimize the network and eventually produce the object phase through the interplay between the neural network and the physical model. This opens up a new paradigm of neural network design, in which the concept of incorporating a physical model into a neural network can be generalized to solve many other CI problems.
Palabras clave

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Idioma: En Revista: Light Sci Appl Año: 2020 Tipo del documento: Article País de afiliación: China Pais de publicación: Reino Unido

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Idioma: En Revista: Light Sci Appl Año: 2020 Tipo del documento: Article País de afiliación: China Pais de publicación: Reino Unido