journal article·2024-12-18

Advanced Image Classification Using a Differential Diffractive Network with “Learned” Structured Illumination

Jiajun Zhang, Shuyan Zhang, Weijie Shi, Yong Hu, Zheng-Gao Dong, Jiaqi Li, Weibing Lu

ACS Photonics· Volume 11· Issue 12
Publisher page

Abstract

As a new optical machine learning framework, the diffractive deep neural network (D 2 NN) has attracted much attention due to its advantages such as low power consumption, parallel computing, and fast execution speed. Here, we demonstrate a new optical neural network design of a differential D 2 NN with structured illumination. In this scheme, the illumination patterns participate in the training process of the network and are optimized by an end-to-end technique. With the application of differential detection, the non-negativity constraint in a diffractive neural network can be alleviated. The test results show that this network architecture can achieve 97.63 and 88.10% classification accuracies on the MNIST and Fashion-MNIST data sets using only one diffractive layer, which exceeds the effect achieved by the five-layer traditional D 2 NN. Moreover, this network architecture can achieve a comprehensive improvement over a traditional D 2 NN in the challenging classification problems of tiny samples and samples blocked by occlusions. Compared with the traditional D 2 NN, this scheme innovatively uses the illumination patterns as new degrees of freedom in system design, which can effectively improve classification ability and reduce the space complexity of the optical neural network.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Neural Networks and Reservoir Computing · Advanced Memory and Neural Computing · Optical Network Technologies

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Key Laboratory of Quantum Materials and Devices of Ministry of Education, School of Physics, Southeast University, Nanjing 211189, China

Center for Flexible RF Technology, State Key Lab of Millimeter waves, School of Information Science and Engineering, Southeast University, Nanjing 210096, China

Funding

China National Funds for Distinguished Young Scientists · Fundamental Research Funds for the Central Universities · National Natural Science Foundation of China

References (56)

Show source-provided references
1. 10.1038/nphoton.2017.9310.1038/nphoton.2017.93 ↗
2. 10.1038/s41586-020-2973-610.1038/s41586-020-2973-6 ↗
3. 10.1186/s43074-021-00026-010.1186/s43074-021-00026-0 ↗
4. 10.1038/s41467-024-45982-w10.1038/s41467-024-45982-w ↗
5. 10.1364/OPTICA.5.00086410.1364/OPTICA.5.000864 ↗
6. 10.1126/science.aat808410.1126/science.aat8084 ↗
7. 10.1117/1.AP.4.6.06400210.1117/1.AP.4.6.064002 ↗
8. 10.1038/s41377-024-01482-610.1038/s41377-024-01482-6 ↗
9. 10.1186/s43593-022-00012-410.1186/s43593-022-00012-4 ↗
10. 10.1021/acsphotonics.1c0136510.1021/acsphotonics.1c01365 ↗
11. 10.1186/s43593-024-00067-510.1186/s43593-024-00067-5 ↗
12. 10.1364/OL.39404610.1364/OL.394046 ↗
13. 10.1186/s43593-022-00021-310.1186/s43593-022-00021-3 ↗
14. 10.1038/s41467-024-49304-y10.1038/s41467-024-49304-y ↗
15. 10.1364/OE.46313710.1364/OE.463137 ↗
16. 10.1038/s41467-020-20268-z10.1038/s41467-020-20268-z ↗
17. 10.1002/adma.20230899310.1002/adma.202308993 ↗
18. 10.1364/PRJ.43291910.1364/PRJ.432919 ↗
19. 10.1038/s41377-020-0303-210.1038/s41377-020-0303-2 ↗
20. 10.1038/s41598-022-11331-410.1038/s41598-022-11331-4 ↗
21. 10.1002/lpor.20230136710.1002/lpor.202301367 ↗
22. 10.1038/s41377-023-01116-310.1038/s41377-023-01116-3 ↗
23. 10.1038/s41377-020-00446-w10.1038/s41377-020-00446-w ↗
24. 10.1515/nanoph-2022-061510.1515/nanoph-2022-0615 ↗
25. 10.1038/s41377-020-00439-910.1038/s41377-020-00439-9 ↗
26. 10.1103/PhysRevLett.123.02390110.1103/PhysRevLett.123.023901 ↗
27. 10.1021/acsphotonics.0c0158310.1021/acsphotonics.0c01583 ↗
28. 10.1364/OL.44.00518610.1364/OL.44.005186 ↗
29. 10.1515/nanoph-2020-029110.1515/nanoph-2020-0291 ↗
30. 10.1038/s41467-023-42556-010.1038/s41467-023-42556-0 ↗
31. 10.1364/OL.52064210.1364/OL.520642 ↗
32. 10.1364/PRJ.47453510.1364/PRJ.474535 ↗
33. 10.1364/PRJ.51384510.1364/PRJ.513845 ↗
34. 10.1364/PRJ.51353710.1364/PRJ.513537 ↗
35. 10.1364/AO.43573810.1364/AO.435738 ↗
36. Wirth-Singh, A.; Xiang, J.; Choi, M.; Fröch, J. E.; Huang, L.; Colburn, S.; Shlizerman, E.; Majumdar, A. Compressed Meta-Optical Encoder for Image Classification, 2024. ArXiv ArXiv:2406.06534. 10.48550/arXiv.2406.06534.10.48550/arXiv.2406.06534 ↗
37. 10.1038/s41566-024-01450-x10.1038/s41566-024-01450-x ↗
38. 10.1038/s41467-022-32117-210.1038/s41467-022-32117-2 ↗
39. 10.1364/OE.39237010.1364/OE.392370 ↗
40. 10.1117/1.AP.1.4.04600110.1117/1.AP.1.4.046001 ↗
41. 10.1016/j.cossms.2024.10114410.1016/j.cossms.2024.101144 ↗
42. 10.1002/adma.20220639910.1002/adma.202206399 ↗
43. 10.1002/adma.20231178510.1002/adma.202311785 ↗
44. 10.1002/adma.20220852010.1002/adma.202208520 ↗
45. 10.1038/s41928-022-00719-910.1038/s41928-022-00719-9 ↗
46. 10.1038/s41377-022-00844-210.1038/s41377-022-00844-2 ↗
47. 10.1002/lpor.20230027210.1002/lpor.202300272 ↗
48. 10.1117/1.APN.2.6.06600610.1117/1.APN.2.6.066006 ↗
49. 10.1038/s41566-021-00796-w10.1038/s41566-021-00796-w ↗
50. 10.1021/acs.nanolett.0c0110510.1021/acs.nanolett.0c01105 ↗
51. 10.1038/s41566-023-01170-810.1038/s41566-023-01170-8 ↗
52. 10.1364/OPTICA.6.00113210.1364/OPTICA.6.001132 ↗
53. 10.1364/PRJ.41596410.1364/PRJ.415964 ↗
54. 10.1038/s41586-023-06558-810.1038/s41586-023-06558-8 ↗
55. 10.1038/s41467-024-50677-310.1038/s41467-024-50677-3 ↗
56. 10.1038/s41467-023-42984-y10.1038/s41467-023-42984-y ↗