All-optical depth diffraction neural network system and method based on spatial partially coherent light
A neural network and coherent light technology, applied in the field of all-optical deep diffraction neural network system, can solve problems such as incoherent signal processing, achieve fast machine learning tasks, low power consumption, and realize the effect of complex applications
Active Publication Date: 2021-09-07
TSINGHUA UNIV
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] The above-mentioned all-optical diffraction deep neural network must be calculated for coherent light, but most light sources in natural scenes are spatially partially coherent light, so the current all-optical diffraction deep neural network cannot directly process partially coherent signals in natural scenes
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0028] Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary and are intended to explain the present invention and should not be construed as limiting the present invention.
[0029] The following describes the plenoptic depth diffraction neural network system and method based on spatial partially coherent light according to the embodiments of the present invention with reference to the accompanying drawings. Diffraction neural network system.
[0030] figure 1 It is a schematic structural diagram of an all-optical depth diffraction neural network system based on spatial partially coherent light according to an embodiment of the present invention.
[0031] Such as figure 1 As shown, the plenoptic de...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses an all-optical depth diffraction neural network system and method based on spatial partially coherent light, comprising: a conversion module for converting an input spatial partially coherent optical signal into a coherent optical signal; an all-optical deep diffraction neural network module , used for transforming, extracting and compressing the coherent optical signal; the information acquisition module, used for receiving the output signal of the all-optical deep diffraction neural network module, and generating the processing result of the spatially partially coherent optical signal according to the output signal. The system can expand the application field of the all-optical deep-diffraction neural network, and enable the all-optical deep-diffractive neural network to better complete more complex machine learning tasks, especially to complete natural scene image recognition processing and computing tasks.
Description
technical field [0001] The invention relates to the technical fields of optoelectronic computing and machine learning, in particular to an all-optical deep diffraction neural network system and method based on spatial partially coherent light. Background technique [0002] Deep learning uses multilayer artificial neural networks implemented in computers to digitally learn information from data and perform advanced tasks with performance comparable to or better than humans. Recently, examples where deep learning has made significant progress in the field of machine learning include medical image analysis, speech recognition, image classification, and more. The traditional deep learning network is implemented based on circuits, and its running speed is limited by electrical devices such as CPU and GPU in the circuit, which has problems such as slow running speed, low computing efficiency, and huge energy consumption. Currently, there is an all-optical diffractive deep neural ...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06N3/067G06N20/00G02F1/35G02F1/355G02F1/01
CPCG02F1/0136G02F1/3515G02F1/3551G06N3/067G06N3/0675G06N20/00
Inventor 谢浩林星周天贶严涛吴嘉敏戴琼海
Owner TSINGHUA UNIV