CN116310267A - Image processing device and processing method thereof - Google Patents
Image processing device and processing method thereof Download PDFInfo
- Publication number
- CN116310267A CN116310267A CN202211485433.1A CN202211485433A CN116310267A CN 116310267 A CN116310267 A CN 116310267A CN 202211485433 A CN202211485433 A CN 202211485433A CN 116310267 A CN116310267 A CN 116310267A
- Authority
- CN
- China
- Prior art keywords
- image processing
- mach
- optical
- zehnder modulator
- peak
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012545 processing Methods 0.000 title claims abstract description 45
- 238000003672 processing method Methods 0.000 title claims abstract description 13
- 230000003287 optical effect Effects 0.000 claims abstract description 89
- 230000010287 polarization Effects 0.000 claims abstract description 18
- 210000002569 neuron Anatomy 0.000 claims description 21
- 239000006096 absorbing agent Substances 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 5
- 229920006395 saturated elastomer Polymers 0.000 claims description 5
- 238000005265 energy consumption Methods 0.000 abstract description 5
- 238000004519 manufacturing process Methods 0.000 abstract description 3
- 230000004044 response Effects 0.000 description 7
- 238000013527 convolutional neural network Methods 0.000 description 6
- 238000000034 method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 210000005036 nerve Anatomy 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G02—OPTICS
- G02F—OPTICAL DEVICES OR ARRANGEMENTS FOR THE CONTROL OF LIGHT BY MODIFICATION OF THE OPTICAL PROPERTIES OF THE MEDIA OF THE ELEMENTS INVOLVED THEREIN; NON-LINEAR OPTICS; FREQUENCY-CHANGING OF LIGHT; OPTICAL LOGIC ELEMENTS; OPTICAL ANALOGUE/DIGITAL CONVERTERS
- G02F1/00—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics
- G02F1/01—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics for the control of the intensity, phase, polarisation or colour
- G02F1/21—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics for the control of the intensity, phase, polarisation or colour by interference
- G02F1/212—Mach-Zehnder type
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/061—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using biological neurons, e.g. biological neurons connected to an integrated circuit
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Nonlinear Science (AREA)
- Neurology (AREA)
- Computational Linguistics (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Optics & Photonics (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention relates to an image processing device and a processing method thereof, wherein the image processing device comprises a master laser, a first polarization controller, a Mach-Zehnder modulator, a first variable optical attenuator, an optical circulator and a slave laser which are connected in sequence; the image processing device further comprises an arbitrary waveform generator, and the output end of the arbitrary waveform generator is connected with the input end of the Mach-Zehnder modulator. The invention can simulate biological vision to realize image processing, so that machine vision is personified, the biological rationality of image processing is improved, the energy consumption of the device is reduced, the image processing rate is improved, and the production cost is reduced.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to an image processing apparatus and a processing method thereof.
Background
With the continuous development of technology, people have increasingly demanded target recognition, target search, voice recognition and other aspects, and the increase of data processing capacity, diversification of information, processing rapidity requirements and the like have brought new challenges. In recent years, in the photonic field, many different approaches have been proposed for hardware implementation of neuromorphic photonic platforms, mainly to achieve simulation of biological neuron dynamics through nonlinearities of optical materials or devices.
The method mainly uses a Convolutional Neural Network (CNN) in the aspect of image processing, the conventional scheme utilizes rich dynamics of a photon device to realize reproduction of the Convolutional Neural Network (CNN), and although the operation speed is improved to a certain extent, the inherent defects of the Convolutional Neural Network (CNN) still exist, such as large number of characteristic layers, large number of learning parameters and low biological rationality, thereby increasing system energy consumption and reducing efficiency.
Disclosure of Invention
The invention aims to solve the technical problems in the prior art and provide an image processing device and a processing method thereof, which can simulate biological vision to realize image processing, make machine vision personification, improve the biological rationality of image processing, reduce the energy consumption of the device, improve the image processing rate and reduce the production cost.
According to the technical scheme provided by the invention, the image processing device comprises a main laser, a first polarization controller, a Mach-Zehnder modulator, a first variable optical attenuator, an optical circulator and a slave laser which are connected in sequence; the image processing device also comprises an arbitrary waveform generator, wherein the output end of the arbitrary waveform generator is connected with the input end of the Mach-Zehnder modulator;
the main laser is used for generating an optical signal and transmitting the optical signal to the Mach-Zehnder modulator; the arbitrary waveform generator is used for generating an electric signal according to pixel information and biological receptive field characteristics in an image and transmitting the electric signal to the Mach-Zehnder modulator; the Mach-Zehnder modulator is used for mapping the optical signals into peak sequences with different weights according to the electrical signals so as to simulate biological receptive field characteristics and encode the peak sequences; the slave laser is used for integrating the peak sequences with different weights and displaying the segmented target image.
In one embodiment of the present invention, an optical isolator for preventing echo reflection of the optical signal is further included, and the optical isolator is connected between the mach-zehnder modulator and the first variable optical attenuator.
In one embodiment of the present invention, the optical system further comprises a second polarization controller and a second variable optical attenuator connected in sequence, wherein the second polarization controller and the second variable optical attenuator are positioned between the first variable optical attenuator and the optical circulator.
In one embodiment of the present invention, an optical isolator for preventing echo reflection of the optical signal is further included, and the optical isolator is connected between the mach-zehnder modulator and the first variable optical attenuator.
In one embodiment of the invention, the system further comprises an optical coupler for dividing the optical signal into two paths, a first photoelectric detector, a second photoelectric detector and an oscilloscope for displaying whether the peak sequence is responsive or not; the optical coupler is connected between the first variable optical attenuator and the second polarization controller; the input end of the first photoelectric detector is connected with the output end of the optical coupler, and the output end of the first photoelectric detector is connected with the oscilloscope; the input end of the second photoelectric detector is connected with the optical circulator, and the output end of the second photoelectric detector is connected with the oscilloscope.
In one embodiment of the invention, the slave laser is a two-segment laser with a saturated absorber.
In one embodiment of the invention, the mach-zehnder modulator maps the optical signal into a spike train by means of time encoding, the mach-zehnder modulator encoding the spike train in different time windows by means of time multiplexing.
In one embodiment of the invention, the arbitrary waveform generator generates the electrical signal according to a biosensing field characterization formula, the biosensing field characterization formula being:
wherein omega ij Representing the connection weight, omega, of the receptive field with the neuron i as the center of the input layer and the neuron j of the middle layer 0 Represents an initial connection weight, (d (I) i )-d(I j )) 2 Represents the distance, d, from neuron j to neuron i in the receptive field max Is constant.
The invention also comprises an image processing method which is applied to the image processing device, and the image processing method comprises the following steps:
generating an optical signal by the main laser and transmitting the optical signal to the Mach-Zehnder modulator;
generating an electrical signal by the arbitrary waveform generator and transmitting the electrical signal to the mach-zehnder modulator; wherein the electrical signal is generated according to pixel information in the image and biological receptive field characteristics;
mapping the optical signals into peak sequences with different weights according to the electric signals through the Mach-Zehnder modulator so as to simulate biological receptive field characteristics, and encoding the peak sequences;
and displaying the segmented target image through integrating the peak sequences with different weights from the laser.
In one embodiment of the invention, the segmented target image is displayed by setting a threshold in the slave laser, causing a spike sequence up to the threshold to respond.
In one embodiment of the invention, the mach-zehnder modulator maps the optical signal into a spike train by means of time encoding, the mach-zehnder modulator encoding the spike train in different time windows by means of time multiplexing.
1. According to the method, a receptive field characteristic formula is constructed, different weights are respectively given to the encoded spike sequences according to the receptive field characteristic formula, so that the connection strength between neurons which are closer to the center of the receptive field is higher, the simulated biological vision is used for realizing image processing, the machine vision is personified, and the biological rationality of the image processing is improved.
2. The secondary laser with the saturated absorber can better simulate the biological dynamics characteristic, can release higher speed peak sequences, has lower working threshold and working current, and further obtains higher information processing speed with lower energy consumption; the response condition of the peak sequence can be controlled by inputting a threshold value from the laser, so that the image processing device is simplified, and the production cost is reduced.
3. The invention adopts a time division multiplexing mode to encode the pixel information into different time windows of a single neuron, thereby realizing the simplification of the structure, and particularly reducing the implementation condition of hardware in practical application.
Drawings
In order that the invention may be more readily understood, a more particular description of the invention will be rendered by reference to specific embodiments thereof that are illustrated in the appended drawings.
FIG. 1 is a schematic diagram of an image processing apparatus of the present invention;
FIG. 2 is an oscilloscope display of a receptive field after processing a simple binary image in accordance with the invention;
FIG. 3 is a schematic diagram of a simple binary image processing result according to the present invention;
FIG. 4 is a schematic diagram of the complex binary image processing result according to the present invention.
Description of the specification reference numerals: 1-a primary laser; 2-a first polarization controller; 3-an arbitrary waveform generator; a 4-Mach-Zehnder modulator; a 5-optical isolator; 6-a first variable optical attenuator; 7-an optical coupler; 8-a second polarization controller; 9-a second variable optical attenuator; 10-optical circulator; 11-slave lasers; 12-a first photodetector; 13-a second photodetector; 14-oscilloscope.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and specific examples, which are not intended to be limiting, so that those skilled in the art will better understand the invention and practice it.
Referring to fig. 1, in order to simulate biological vision to realize image processing, to personify machine vision and improve biological rationality of image processing, an image processing apparatus of the present invention includes a master laser 1, a first polarization controller 2, a mach-zehnder modulator 4, a first variable optical attenuator 6, an optical circulator 10, and a slave laser 11 connected in this order; the image processing device further comprises an arbitrary waveform generator 3, wherein the output end of the arbitrary waveform generator 3 is connected with the input end of the Mach-Zehnder modulator 4;
wherein the main laser 1 is configured to generate an optical signal and transmit the optical signal to the mach-zehnder modulator 4; the arbitrary waveform generator 3 is configured to generate an electrical signal according to pixel information in an image and biological receptive field characteristics, and transmit the electrical signal to the mach-zehnder modulator 4; the Mach-Zehnder modulator 4 is configured to map the optical signal into a spike sequence with different weights according to the electrical signal, so as to simulate a biological receptive field characteristic, and encode the spike sequence; the slave laser 11 is used for integrating the peak sequences with different weights, and displaying the segmented target image.
In general, a receptive field of biological vision is composed of n×n neurons, and the closer to the center of the receptive field, the larger the contribution of the neurons to the image characteristics is, the larger the distance between the neurons located at the center of the receptive field and other neurons is, and the smaller the connection weight between the two neurons is. The invention adopts a mode of simulating biological receptive field characteristics to carry out segmentation processing on the image, and the biological rationality of the image processing is high.
As shown in fig. 1, a main laser 1 generates an optical signal and transmits the optical signal to the mach-zehnder modulator 4; the arbitrary waveform generator 3 converts pixel information in the image into an electric signal according to the biological receptive field characteristics, and transmits the electric signal to the Mach-Zehnder modulator 4; the optical signals are modulated and converted into peak sequences in the Mach-Zehnder modulator 4 to form a peak neural network, the peak sequences are endowed with different weights according to biological receptive field characteristics, compared with other nerve morphology systems, the peak neural network simulates peak reactions in biological neurons, and nerve morphology calculation can be realized in a more biological mode. The invention adopts the following lasers 11 to integrate the peak sequences with different weights, a response threshold value can be set in the following lasers 11, the peak sequences reaching the response threshold value can generate peak responses, and then the images after the responses are displayed in the following lasers 11. The secondary laser 11 adopted by the invention is a two-stage laser with a saturated absorber, the secondary laser 11 with the saturated absorber has the time integral characteristic, can better simulate the biological dynamics characteristic, can release peak sequences with higher speed, has lower working threshold and working current, further obtains higher information processing speed with lower energy consumption, and the specific condition and working principle of the secondary laser 11 are consistent with the prior art, and are not repeated here.
As shown in fig. 3, the image processing apparatus of the present invention processes a simple binary image, where (a), (b) is an original image, (a), (b) is an edge detection result, and (a), (b) is an image segmentation result obtained by adding a threshold to an output spike sequence so that a part of spikes do not respond, and thus the edge detection result is optimized.
Further, the present invention further comprises an optical isolator 5 for preventing echo reflection of the optical signal, said optical isolator 5 being connected between said mach-zehnder modulator 4 and the first variable optical attenuator 6.
Further, in order to process the complex binary image, the present invention further includes a second polarization controller 8 and a second variable optical attenuator 9 connected in sequence, where the second polarization controller 8 and the second variable optical attenuator 9 are located between the first variable optical attenuator 6 and the optical circulator 10.
Specifically, as shown in fig. 4, the segmentation results of the complex binary image are shown, where (a) is the original image and (b) - (f) are the segmentation results under different thresholds. It can be seen from the figure that the image processing apparatus of the present invention can not only process a simple binary image, but also process a complex binary image by adding a threshold to a peak sequence in the slave laser 11 after the second polarization controller 8 and the second variable optical attenuator 9 are provided. The function and operation of the second polarization controller 8 and the second variable optical attenuator 9 are consistent with the prior art and are well known to those skilled in the art.
Further, the invention also comprises an optical coupler 7 for dividing the optical signal into two paths, a first photoelectric detector 12, a second photoelectric detector 13 and an oscilloscope 14 for displaying whether the peak sequence responds or not; the optical coupler 7 is connected between the first variable optical attenuator 6 and the second polarization controller 8; the input end of the first photoelectric detector 12 is connected with the output end of the optical coupler 7, and the output end of the first photoelectric detector 12 is connected with the oscilloscope 14; the input end of the second photodetector 13 is connected with the optical circulator 10, and the output end of the second photodetector 13 is connected with the oscilloscope 14.
Specifically, as shown in fig. 2, the output from the laser 11 enters the oscilloscope 14 through the second photodetector 13, and then shows the response result of the spike sequence. The first photodetector 12 inputs the peak sequence which is not processed by the second polarization controller 8 and the second variable optical attenuator 9 to the oscilloscope 14, and can be compared with the peak sequence response result input by the second photodetector 13.
Further, in order to simplify the structure and reduce the hardware implementation condition, the mach-zehnder modulator 4 maps the optical signal into a spike sequence by using a time coding manner, and the mach-zehnder modulator 4 codes the spike sequence in different time windows by using a time division multiplexing manner.
Further, the arbitrary waveform generator 3 generates an electric signal according to a biological receptive field characteristic formula:
wherein omega ij Representing the connection weight, omega, of the receptive field with the neuron i as the center of the input layer and the neuron j of the middle layer 0 Represents an initial connection weight, (d (I) i )-d(I j )) 2 Represents the distance, d, from neuron j to neuron i in the receptive field max Is constant.
In summary, the present invention also includes an image processing method applied to the above image processing apparatus, where the image processing method includes:
generating an optical signal by the main laser 1 and transmitting the optical signal into the mach-zehnder modulator 4;
generating an electrical signal by the arbitrary waveform generator 3 and transmitting the electrical signal into the mach-zehnder modulator 4; wherein the electrical signal is generated according to pixel information in the image and biological receptive field characteristics;
mapping the optical signal into peak sequences with different weights according to the electric signal through the Mach-Zehnder modulator 4 so as to simulate biological receptive field characteristics and code the peak sequences;
the segmented target image is displayed by integrating the spike sequences of different weights from the laser 11.
It is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations and modifications of the present invention will be apparent to those of ordinary skill in the art in light of the foregoing description. It is not necessary here nor is it exhaustive of all embodiments. And obvious variations or modifications thereof are contemplated as falling within the scope of the present invention.
Claims (10)
1. An image processing apparatus, characterized in that: the system comprises a master laser, a first polarization controller, a Mach-Zehnder modulator, a first variable optical attenuator, an optical circulator and a slave laser which are connected in sequence; the image processing device also comprises an arbitrary waveform generator, wherein the output end of the arbitrary waveform generator is connected with the input end of the Mach-Zehnder modulator;
the main laser is used for generating an optical signal and transmitting the optical signal to the Mach-Zehnder modulator; the arbitrary waveform generator is used for generating an electric signal according to pixel information and biological receptive field characteristics in an image and transmitting the electric signal to the Mach-Zehnder modulator; the Mach-Zehnder modulator is used for mapping the optical signals into peak sequences with different weights according to the electrical signals so as to simulate biological receptive field characteristics and encode the peak sequences; the slave laser is used for integrating the peak sequences with different weights and displaying the segmented target image.
2. The image processing apparatus according to claim 1, wherein: and an optical isolator for preventing echo reflection of the optical signal, the optical isolator being connected between the Mach-Zehnder modulator and the first variable optical attenuator.
3. The image processing apparatus according to claim 1, wherein: the optical system further comprises a second polarization controller and a second variable optical attenuator which are connected in sequence, wherein the second polarization controller and the second variable optical attenuator are connected in sequence and are positioned between the first variable optical attenuator and the optical circulator.
4. An image processing apparatus according to claim 3, wherein: the system also comprises an optical coupler for dividing the optical signal into two paths, a first photoelectric detector, a second photoelectric detector and an oscilloscope for displaying whether the peak sequence is responded or not; the optical coupler is connected between the first variable optical attenuator and the second polarization controller; the input end of the first photoelectric detector is connected with the output end of the optical coupler, and the output end of the first photoelectric detector is connected with the oscilloscope; the input end of the second photoelectric detector is connected with the optical circulator, and the output end of the second photoelectric detector is connected with the oscilloscope.
5. The image processing apparatus according to claim 1, wherein: the slave laser is a two-stage laser with a saturated absorber.
6. The image processing apparatus according to claim 1, wherein: the Mach-Zehnder modulator maps the optical signal into a peak sequence in a time coding mode, and the Mach-Zehnder modulator codes the peak sequence in different time windows in a time multiplexing mode.
7. The image processing apparatus according to claim 1, wherein: the arbitrary waveform generator generates an electrical signal according to a biological receptive field characteristic formula, wherein the biological receptive field characteristic formula is as follows:
wherein omega ij Representing the connection weight, omega, of the receptive field with the neuron i as the center of the input layer and the neuron j of the middle layer 0 Represents an initial connection weight, (d (I) i )-d(I j )) 2 Represents the distance, d, from neuron j to neuron i in the receptive field max Is constant.
8. An image processing method applied to the image processing apparatus according to any one of claims 1 to 7, characterized in that the image processing method comprises:
generating an optical signal by the main laser and transmitting the optical signal to the Mach-Zehnder modulator;
generating an electrical signal by the arbitrary waveform generator and transmitting the electrical signal to the mach-zehnder modulator; wherein the electrical signal is generated according to pixel information in the image and biological receptive field characteristics;
mapping the optical signals into peak sequences with different weights according to the electric signals through the Mach-Zehnder modulator so as to simulate biological receptive field characteristics, and encoding the peak sequences;
and displaying the segmented target image through integrating the peak sequences with different weights from the laser.
9. The image processing method according to claim 8, characterized in that: and responding the peak sequence reaching the threshold by setting the threshold in the slave laser so as to display the segmented target image.
10. The image processing method according to claim 8, characterized in that: the Mach-Zehnder modulator maps the optical signal into a peak sequence in a time coding mode, and the Mach-Zehnder modulator codes the peak sequence in different time windows in a time multiplexing mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211485433.1A CN116310267A (en) | 2022-11-24 | 2022-11-24 | Image processing device and processing method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211485433.1A CN116310267A (en) | 2022-11-24 | 2022-11-24 | Image processing device and processing method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116310267A true CN116310267A (en) | 2023-06-23 |
Family
ID=86827519
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211485433.1A Pending CN116310267A (en) | 2022-11-24 | 2022-11-24 | Image processing device and processing method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116310267A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108197711A (en) * | 2017-12-13 | 2018-06-22 | 中国科学院自动化研究所 | The computational methods for paying attention to switching are felt based on class brain more |
| CN109952525A (en) * | 2016-09-02 | 2019-06-28 | 菲托尼克斯公司 | Method for being scanned along three-dimensional line and the method for scanning interested region by scanning multiple three-dimensional lines |
| WO2019183376A1 (en) * | 2018-03-21 | 2019-09-26 | The Board Of Trustees Of The Leland Stanford Junior University | Methods for temporal and spatial multiplexing of spatial light modulators and systems for same |
| CN112580737A (en) * | 2020-12-25 | 2021-03-30 | 航天科工网络信息发展有限公司 | Pulse neural network feature extraction method based on multi-scale feature fusion |
| CN114548393A (en) * | 2022-02-25 | 2022-05-27 | 太原理工大学 | High-speed information processing system for large-operation-space reserve pool calculation |
| CN114757344A (en) * | 2022-04-07 | 2022-07-15 | 苏州大学 | System and method for simulating biological neuron dynamics to realize logic operation |
| CN114970599A (en) * | 2022-04-01 | 2022-08-30 | 中国科学院深圳先进技术研究院 | Identification method and identification device for attention defect associated electroencephalogram signals and storage medium |
| CN115037374A (en) * | 2022-04-28 | 2022-09-09 | 苏州大学 | Broadband double-path polarization chaotic laser generation method and device |
-
2022
- 2022-11-24 CN CN202211485433.1A patent/CN116310267A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109952525A (en) * | 2016-09-02 | 2019-06-28 | 菲托尼克斯公司 | Method for being scanned along three-dimensional line and the method for scanning interested region by scanning multiple three-dimensional lines |
| CN108197711A (en) * | 2017-12-13 | 2018-06-22 | 中国科学院自动化研究所 | The computational methods for paying attention to switching are felt based on class brain more |
| WO2019183376A1 (en) * | 2018-03-21 | 2019-09-26 | The Board Of Trustees Of The Leland Stanford Junior University | Methods for temporal and spatial multiplexing of spatial light modulators and systems for same |
| CN112580737A (en) * | 2020-12-25 | 2021-03-30 | 航天科工网络信息发展有限公司 | Pulse neural network feature extraction method based on multi-scale feature fusion |
| CN114548393A (en) * | 2022-02-25 | 2022-05-27 | 太原理工大学 | High-speed information processing system for large-operation-space reserve pool calculation |
| CN114970599A (en) * | 2022-04-01 | 2022-08-30 | 中国科学院深圳先进技术研究院 | Identification method and identification device for attention defect associated electroencephalogram signals and storage medium |
| CN114757344A (en) * | 2022-04-07 | 2022-07-15 | 苏州大学 | System and method for simulating biological neuron dynamics to realize logic operation |
| CN115037374A (en) * | 2022-04-28 | 2022-09-09 | 苏州大学 | Broadband double-path polarization chaotic laser generation method and device |
Non-Patent Citations (4)
| Title |
|---|
| DONGHAO ZHENG 等: "Image segmentation method based on spiking neural network with adaptive synaptic weights", 《2019ICSIP》, 17 October 2019 (2019-10-17) * |
| PALLAVI GUPTA 等: "A Single-Neuron: Current Trends and Future Prospects", 《CELLS》, vol. 9, no. 6, 23 June 2020 (2020-06-23) * |
| 刘林云 等: "基于卷积神经网络的旁路密码分析综述", 《计算机科学》, vol. 49, no. 5, 11 January 2022 (2022-01-11) * |
| 刘远 等: "基于滤波反馈宽带平坦混沌信号的快速物理随机比特产生", 《物理学报》, vol. 71, no. 2, 2 August 2022 (2022-08-02) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021248868A1 (en) | Knowledge distillation-based compression method for pre-trained language model, and platform | |
| Wang et al. | Intelligent constellation diagram analyzer using convolutional neural network-based deep learning | |
| CN107342810B (en) | Deep learning Brilliant Eyes figure analysis method based on convolutional neural networks | |
| CN109376855B (en) | Optical neuron structure and neural network processing system comprising same | |
| Fan et al. | Joint optical performance monitoring and modulation format/bit-rate identification by CNN-based multi-task learning | |
| CN110932809B (en) | Fiber channel model simulation method, device, electronic equipment and storage medium | |
| Wang et al. | Adaptive demodulation technique for efficiently detecting orbital angular momentum (OAM) modes based on the improved convolutional neural network | |
| CN105787458A (en) | Infrared behavior identification method based on adaptive fusion of artificial design feature and depth learning feature | |
| CN111402866B (en) | Semantic recognition method and device and electronic equipment | |
| KR20190099380A (en) | A method for associating an ai device with a device based on a behavior pattern of a user and an apparatus therefor | |
| Wang et al. | Convolutional neural network-based deep learning for intelligent OSNR estimation on eye diagrams | |
| Tizikara et al. | Machine learning-aided optical performance monitoring techniques: A review | |
| CN116310267A (en) | Image processing device and processing method thereof | |
| CN111291881B (en) | Mode identification device and method based on cascade VCSELs photonic nerves | |
| Gao et al. | Joint baud-rate and modulation format identification based on asynchronous delay-tap plots analyzer by using convolutional neural network | |
| KR102567128B1 (en) | Enhanced adversarial attention networks system and image generation method using the same | |
| CN116188870A (en) | Steel surface defect image classification method based on pulse convolution neural network | |
| Xu et al. | Deep transfer learning based multi-impairment diagnosis for PAM-4 optical communication systems | |
| Chen et al. | Receptive Field-Based All-Optical Spiking Neural Network for Image Processing | |
| CN112767501B (en) | VCSEL-SA image identification system and method based on electric control stimulation | |
| Zhang et al. | Deep Learning-Based Intelligent Fibre Modulation Format Recognition in Optically Coupled Modes | |
| Zhu et al. | Multiple Stokes sectional plane image based modulation format recognition with a generative adversarial network | |
| CN117542085B (en) | Park scene pedestrian detection method, device and equipment based on knowledge distillation | |
| KR102417150B1 (en) | Method of recognizing user's motion and system therefor | |
| CN114050953B (en) | Method and system for identifying signal modulation format of coherent optical communication system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |