Zaikun Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Zaikun Zhang has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Depeng Kong
State Key Laboratory of Ultrafast Optical Science and Technology, Xi'an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, Xi'an, China
3 shared publications Yi Geng
Xi’an Institute of Applied Optics, Xi’an 710065, People’s Republic of China
3 shared publicationsZhengshang Da
Xi’an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences (China)
3 shared publicationsHui Chen
2 shared publicationsRuiduo Wang
State Key Laboratory of Transient Optics and Photonics, Xi’an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, Xi’an 710119, People’s Republic of China
2 shared publicationsZhengquan He
2 shared publications
Latest Publications
- 24 November 2023
- Published by SPIE-Intl Soc Optical Eng
- Vol. 12935, 18-125
Compact lensless optoelectronic convolutional neural network for image classification
Recently, free-space optical neural networks (ONNs) have gained extensive interest as emerging machine learning platforms for implementing artificial intelligence tasks, such as image classification. Despite various optical implementations of electronic neural networks (ENNs), the bulky volume of optical components remains challenging to deploy edge devices, such as Internet of Things peripherals, wearable devices, and camera. To address this problem, we propose a compact lensless optoelectronic convolutional neural network (LOE-CNN) architecture with a lensless optical analog processor utilizing a single optimized diffractive phase mask (DPM) to perform convolution operations without Fourier lens. Comparing the processor with a commercially available NVIDIA A100 Tensor Core GPU in terms of speed and power, indicates the optical computing platform enables to replace the electronic processor in latency reduction and energy savings. Furthermore, we compare the LOE-CNN with two all-electronic neural networks (i.e., fully connected neural network [FC-NN] and convolutional neural network [CNN]) over the Modified National Institute of Standards and Technology (MNIST) dataset and Fashion-MNIST dataset, respectively, and demonstrate that the LOE-CNN can be functionally comparable to existing electronic counterparts in classification performance. My study not only opens up new application prospects for free-space ONNs based on compact lensless single-chip convolution processor, but also facilitates the development of ONNs-based smart devices.
- 24 November 2023
- Published by SPIE-Intl Soc Optical Eng
- Vol. 12935, 18-125
Compact lensless optoelectronic convolutional neural network for image classification
Recently, free-space optical neural networks (ONNs) have gained extensive interest as emerging machine learning platforms for implementing artificial intelligence tasks, such as image classification. Despite various optical implementations of electronic neural networks (ENNs), the bulky volume of optical components remains challenging to deploy edge devices, such as Internet of Things peripherals, wearable devices, and camera. To address this problem, we propose a compact lensless optoelectronic convolutional neural network (LOE-CNN) architecture with a lensless optical analog processor utilizing a single optimized diffractive phase mask (DPM) to perform convolution operations without Fourier lens. Comparing the processor with a commercially available NVIDIA A100 Tensor Core GPU in terms of speed and power, indicates the optical computing platform enables to replace the electronic processor in latency reduction and energy savings. Furthermore, we compare the LOE-CNN with two all-electronic neural networks (i.e., fully connected neural network [FC-NN] and convolutional neural network [CNN]) over the Modified National Institute of Standards and Technology (MNIST) dataset and Fashion-MNIST dataset, respectively, and demonstrate that the LOE-CNN can be functionally comparable to existing electronic counterparts in classification performance. My study not only opens up new application prospects for free-space ONNs based on compact lensless single-chip convolution processor, but also facilitates the development of ONNs-based smart devices.
- 25 May 2023
- Published by IOP Publishing in Journal of Physics D: Applied Physics
- Vol. 56 (35) , 355103
Compact lensless convolution processor for an optoelectronic convolutional neural network
To our knowledge, optical 4f systems have been widely used as a convolutional layer to perform convolutional computation in free-space optical neural networks (ONNs), which makes ONNs too bulky to be easily applied to miniaturized smart systems. Hence, we propose a compact lensless optoelectronic convolutional neural network (LOE-CNN) architecture in which a single optimized diffractive phase mask acts as an analog convolution processor to perform convolutional operation without a Fourier lens or lenslet array. We demonstrate that this LOE-CNN can be functionally comparable to existing electronic counterparts in classification performance, achieving a classification accuracy of 98.07% and 95% over the Modified National Institute of Standards and Technology dataset in simulation and experiment, respectively, which not only opens new application prospects for free-space ONNs based on a compact single-chip convolution processor, but also facilitates the development of ONN-based smart devices.
- 25 May 2023
- Published by IOP Publishing in Journal of Physics D: Applied Physics
- Vol. 56 (35) , 355103
Compact lensless convolution processor for an optoelectronic convolutional neural network
To our knowledge, optical 4f systems have been widely used as a convolutional layer to perform convolutional computation in free-space optical neural networks (ONNs), which makes ONNs too bulky to be easily applied to miniaturized smart systems. Hence, we propose a compact lensless optoelectronic convolutional neural network (LOE-CNN) architecture in which a single optimized diffractive phase mask acts as an analog convolution processor to perform convolutional operation without a Fourier lens or lenslet array. We demonstrate that this LOE-CNN can be functionally comparable to existing electronic counterparts in classification performance, achieving a classification accuracy of 98.07% and 95% over the Modified National Institute of Standards and Technology dataset in simulation and experiment, respectively, which not only opens new application prospects for free-space ONNs based on a compact single-chip convolution processor, but also facilitates the development of ONN-based smart devices.
- 13 August 2021
- Published by IOP Publishing in Applied Physics Express
- Vol. 14 (9) , 092002
Lensless multimode fiber imaging based on wavefront shaping
- 13 August 2021
- Published by IOP Publishing in Applied Physics Express
- Vol. 14 (9) , 092002
Lensless multimode fiber imaging based on wavefront shaping
- 23 September 2020
- Published by Optica Publishing Group in Optics Express
- Vol. 28 (20) , 30048-30062
Binary amplitude-only image reconstruction through a MMF based on an AE-SNN combined deep learning model
The obstacle of imaging through multimode fibers (MMFs) is encountered due to the fact that the inherent mode dispersion and mode coupling lead the output of the MMF to be scattered and bring about image distortions. As a result, only noise-like speckle patterns can be formed on the distal end of the MMF. We propose a deep learning model exploited for computational imaging through an MMF, which contains an autoencoder (AE) for feature extraction and image reconstruction and self-normalizing neural networks (SNNs) sandwiched and employed for high-order feature representation. It was demonstrated both in simulations and in experiments that the proposed AE-SNN combined deep learning model could reconstruct image information from various binary amplitude-only targets going through a 5-meter-long MMF. Simulations indicate that our model works effectively even in the presence of system noise, and the experimental results prove that the method is valid for image reconstruction through the MMF. Enabled by the spatial variability and the self-normalizing properties, our model can be generalized to solve varieties of other computational imaging problems.
- 23 September 2020
- Published by Optica Publishing Group in Optics Express
- Vol. 28 (20) , 30048-30062
Binary amplitude-only image reconstruction through a MMF based on an AE-SNN combined deep learning model
The obstacle of imaging through multimode fibers (MMFs) is encountered due to the fact that the inherent mode dispersion and mode coupling lead the output of the MMF to be scattered and bring about image distortions. As a result, only noise-like speckle patterns can be formed on the distal end of the MMF. We propose a deep learning model exploited for computational imaging through an MMF, which contains an autoencoder (AE) for feature extraction and image reconstruction and self-normalizing neural networks (SNNs) sandwiched and employed for high-order feature representation. It was demonstrated both in simulations and in experiments that the proposed AE-SNN combined deep learning model could reconstruct image information from various binary amplitude-only targets going through a 5-meter-long MMF. Simulations indicate that our model works effectively even in the presence of system noise, and the experimental results prove that the method is valid for image reconstruction through the MMF. Enabled by the spatial variability and the self-normalizing properties, our model can be generalized to solve varieties of other computational imaging problems.