Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 56218 (05)
- University: Sharif University of Technology
- Department: Electrical Engineering
- Advisor(s): Khavasi, Amin
- Abstract:
- Optical computing is a new approach to the hardware implementation of devices that were previously implemented digitally and electronically. It has attracted a great deal of interest due to its benefits, which include high bandwidth, extensive internal connections, the possibility of parallel processing, high calculation speed, and low power consumption. Consequently, this type of implementation is regarded as an appropriate substrate for optical neural networks. Compact and low-power CMOS-compatible hardware can be used for on-chip optical neural networks (ONNs), enabling affordable and portable image classification solutions for applications like autonomous vehicles, healthcare, and optical communication. In this work, we propose a novel one-dimensional Optical Convolutional Neural Network (OCNN) architecture that significantly reduces the number of learn- able parameters required for an ONN. Our OCNN achieves an impressive accuracy of over 96% as a pattern classifier, utilizing only 90 learnable parameters, leading to a simpler structure compared to existing on-chip ONNs. Additionally, our OCNN demonstrates scalability and robustness, with an accuracy exceeding 89% in handwritten digit classification. The OCNN’s convolutional layer employs a lenslet 4f system for convolving desired kernels on input images, while an on-chip lens facilitates the desired Fourier Transform effortlessly. The subsequent layer consists of a sin gle metaline layer, implementing a fully connected layer. By parallelizing pre-trained OCNNs, an on-chip deep convolutional neural network (CNN) can be realized, where each OCNN functions as a separate kernel within a conventional CNN.
- Keywords:
- Analog Computing ; Optical Neural Networks ; Optical Convolutional Neural Network (OCNN) ; Metasurfaces ; Optical Computing ; Metalines ; Complementary Metal Oxide Semiconductor Technology (CMOS)
Digital Object List
محتواي کتاب
Bookmark