Photon intellectual sensing neuron model, method and system

Abstract

The invention relates to the technical field of optical computing, in particular to a photon intelligent sensing computational neuron model, method and system. The model comprises a condensation module, a modulation module and an input module, and the input module is used for inputting preset incident light; the light condensation module is used for acquiring natural light of the first image in the target area and focusing the natural light into the modulation module; the modulation module is connected with the condensation module and the input module and is used for obtaining light focused by the condensation module and modulating the light with preset incident light to obtain a first modulation result. The spatial intensity image is processed through the light condensation module, the modulation module and the input module, coherence and interference effects are fully utilized, parallel processing of multiple features is realized, and thus the image recognition efficiency is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of optical computing technology, and in particular to a photonic intelligent computing neuron model, method and system. Background Art

[0002] With the rapid development of artificial intelligence and scientific computing, the complexity and scale of computing needs are also increasing. However, the performance of existing electronic circuits has reached saturation, limiting the potential to significantly improve signal processing speed and energy efficiency. Light has the advantages of high throughput and low latency during propagation. Based on this, performing image processing in parallel in the optical domain is seen as the key to breaking the bottleneck of existing electronic circuits. Summary of the invention

[0003] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.

[0004] To this end, the first purpose of the present disclosure is to propose a photon intelligent computing neuron model, which processes the spatial intensity image through a focusing module, a modulation module and an input module, makes full use of the coherence and interference effects, realizes the parallel processing of multiple features, and thus improves the image recognition efficiency. At the same time, the photon intelligent computing neuron model can be applied to edge computing, intelligent robots and other fields.

[0005] The second objective of the present disclosure is to provide an image classification method.

[0006] To achieve the above-mentioned purpose, the first aspect of the present disclosure proposes a photon intelligent computing neuron model, which includes a focusing module, a modulation module and an input module, wherein: The input module is used to input preset incident light; The focusing module is used to obtain natural light of the first image in the target area and focus the natural light into the modulation module; The modulation module is connected to the focusing module and the input module, and is used for acquiring the light focused by the focusing module and modulating it with the preset incident light to obtain a first modulation result.

[0007] Optionally, the modulation module is specifically used to modulate the acquired light focused by the focusing module with the same wavelength as the preset incident light to obtain a first modulation result.

[0008] Optionally, the acquiring light focused by the focusing module is modulated with the same wavelength as the preset incident light to obtain a first modulation result, including: The light focused by the focusing module is absorbed by the silicon waveguide to generate photogenerated carriers; Based on the high Q value and the refractive index of the photogenerated carriers, phase and / or amplitude modulation is performed on the same wavelength in the preset incident light to obtain a first modulation result.

[0009] Optionally, the photon intelligent computing neuron model is used for image classification.

[0010] To achieve the above-mentioned purpose, the second aspect of the present disclosure proposes an image classification method, including: Acquire natural light and preset incident light of a first image; Inputting the natural light of the first image and the preset incident light into the photon intelligent sensing neuron model array to obtain first modulation results corresponding to different wavelengths in the preset incident light; Combining the first modulation results corresponding to the same wavelength to obtain a second modulation result of the same wavelength in the preset incident light; Based on a second modulation result of the same wavelength in the preset incident light, a classification result of the first image is determined.

[0011] Optionally, the photon intelligent computing neuron model array is divided into different groups, and different groups of photon intelligent computing neuron models correspond to different wavelengths in the preset incident light.

[0012] Optionally, the step of inputting the natural light of the first image and the preset incident light into the photon intelligent sensing neuron model array to obtain first modulation results corresponding to different wavelengths in the preset incident light includes: The natural light of the first image and the preset incident light are input into the photon intelligent computing neuron model array, and each photon intelligent computing neuron model obtains a first modulation result of the corresponding wavelength based on the natural light of the first image and the preset incident light in the target area where the photon intelligent computing neuron model is located; Based on the first modulation result of the wavelength corresponding to each photon intelligent sensing neuron model, the first modulation results corresponding to different wavelengths in the preset incident light are obtained.

[0013] Optionally, the photon intelligent sensing neuron model includes a focusing module, a modulation module and an input module; the first modulation result of the corresponding wavelength is obtained based on the natural light of the first image in the target area where the photon intelligent sensing neuron model is located and the preset incident light, including: Inputting a preset incident light through the input module; Acquire the natural light of the first image in the target area where the photon intelligent sensing neuron model is located through a focusing module, and focus the natural light into the modulation module; The modulation module modulates the light focused by the focusing module and the preset incident light to obtain a first modulation result of a corresponding wavelength.

[0014] Optionally, determining the classification result of the first image based on the second modulation result of the same wavelength in the preset incident light includes: Based on the second modulation result of the same wavelength in the preset incident light, a corresponding light calculation result is obtained; Based on the correspondence between the features of different wavelengths in the light calculation result and the classification result, the classification result of the first image is determined.

[0015] Another object of the present invention is to provide an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the methods described in the above two aspects.

[0016] Another object of the present invention is to provide a computer storage medium, wherein the computer storage medium stores computer executable instructions; after the computer executable instructions are executed by a processor, the computer executes the methods described in the above two aspects.

[0017] In summary, the photon intelligent computing neuron model, method and system provided by the present disclosure processes the spatial intensity image through the focusing module, modulation module and input module, makes full use of the coherence and interference effects, realizes the parallel processing of multiple features, and thus improves the image recognition efficiency. At the same time, the photon intelligent computing neuron model can be applied to edge computing, intelligent robots and other fields.

[0018] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and / or additional aspects and advantages of the present disclosure will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 A schematic diagram of the structure of a photon intelligent sensing and computing neuron model provided in an embodiment of the present disclosure; Figure 2 A schematic diagram of a flow chart of an image classification method proposed in an embodiment of the present disclosure; Figure 3 A schematic diagram of the structure of an image classification system provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0020] Embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.

[0021] The present disclosure is described in detail below with reference to specific embodiments.

[0022] Figure 1 This is a schematic diagram of the structure of a photon intelligent computing neuron model provided by an embodiment of the present disclosure. Figure 1 As shown, the model includes an input module 101, a focusing module 102 and a modulation module 103, wherein: An input module 101 is used to input a preset incident light; A focusing module 102, used to obtain natural light of the first image in the target area and focus the natural light into the modulation module; The modulation module 103 is connected to the focusing module and the input module, and is used to obtain the light focused by the focusing module and modulate it with the preset incident light to obtain a first modulation result.

[0023] In one embodiment of the present disclosure, the above-mentioned photon intelligent computing neuron model can be applied to a variety of scenarios, such as image classification scenarios.

[0024] In one embodiment of the present disclosure, the modulation module is specifically used to modulate the light focused by the focusing module to obtain the same wavelength as the preset incident light to obtain a first modulation result.

[0025] In one embodiment of the present disclosure, the above-mentioned method of modulating the light focused by the focusing module with the same wavelength in the preset incident light to obtain a first modulation result may include: the light focused by the focusing module is absorbed by the silicon waveguide to generate photogenerated carriers, and based on the high Q value and the refractive index of the photogenerated carriers, the same wavelength in the preset incident light is phase and / or amplitude modulated to obtain the first modulation result.

[0026] Specifically, in one embodiment of the present disclosure, based on the high Q value of the resonant ring in the modulation module, the refractive index change caused by the carriers is amplified, thereby performing phase and / or amplitude modulation on the same wavelength in the preset incident light to obtain a first modulation result.

[0027] The photon intelligent computing neuron model of the disclosed embodiment processes the spatial intensity image through a focusing module, a modulation module and an input module, making full use of the coherence and interference effects to achieve parallel processing of multiple features, thereby improving the image recognition efficiency. At the same time, the photon intelligent computing neuron model can be applied to edge computing, intelligent robots and other fields.

[0028] In order to implement the above embodiment, Figure 2 The present disclosure also proposes an image classification method, which may include the following steps: Step 201, obtaining natural light and preset incident light of a first image; Step 202, inputting the natural light of the first image and the preset incident light into the photon intelligent sensing neuron model array to obtain the first modulation results corresponding to different wavelengths in the preset incident light; Step 203, combining the first modulation results corresponding to the same wavelength to obtain a second modulation result of the same wavelength in the preset incident light; Step 204: Determine a classification result of the first image based on a second modulation result of the same wavelength in the preset incident light.

[0029] In one embodiment of the present disclosure, the photon intelligent computing neuron model array is divided into different groups, and different groups of photon intelligent computing neuron models correspond to different wavelengths in the preset incident light.

[0030] In one embodiment of the present disclosure, a method for inputting natural light and preset incident light of a first image into a photon intelligent sensing neuron model array to obtain first modulation results corresponding to different wavelengths in the preset incident light may include the following steps: Step a: inputting the natural light and the preset incident light of the first image into the photon intelligent computing neuron model array, and each photon intelligent computing neuron model obtains a first modulation result of the corresponding wavelength based on the natural light and the preset incident light of the first image in the target area where the photon intelligent computing neuron model is located; Step b: based on the first modulation result of the wavelength corresponding to each photon intelligent sensing neuron model, obtain the first modulation results corresponding to different wavelengths in the preset incident light.

[0031] In one embodiment of the present disclosure, the above-mentioned photon intelligent computing neuron model includes a focusing module, a modulation module and an input module. In one embodiment of the present disclosure, the above-mentioned method for obtaining a first modulation result of a corresponding wavelength based on the natural light and the preset incident light of the first image in the target area where the photon intelligent computing neuron model is located may include the following steps: Step 1, inputting a preset incident light through an input module; Step 2: Obtain the natural light of the first image in the target area where the photon intelligent sensing neuron model is located through the focusing module, and focus the natural light into the modulation module; Step 3: The modulation module modulates the light focused by the focusing module with the preset incident light to obtain a first modulation result of the corresponding wavelength.

[0032] And, in one embodiment of the present disclosure, after obtaining the first modulation results corresponding to different wavelengths in the preset incident light through the above steps, the first modulation results corresponding to the same wavelength can be combined to obtain the second modulation result of the same wavelength in the preset incident light. Among them, the first modulation results corresponding to the same wavelength can be coherently combined to obtain the second modulation result of the same wavelength in the preset incident light.

[0033] Specifically, in one embodiment of the present disclosure, it is assumed that group A of photon intelligent computing neuron models processes wavelength 1, and group A of photon intelligent computing neuron models includes photon intelligent computing neuron model 1 and photon intelligent computing neuron model 2. Among them, when the photon intelligent computing neuron model 1 receives ambient light with an intensity of 1, the first modulation result of wavelength 1 is π, and when the photon intelligent computing neuron model 2 receives ambient light with an intensity of 1, the first modulation result of wavelength 1 is -π, then the first modulation results corresponding to wavelength 1 after the photon intelligent computing neuron model 1 and the photon intelligent computing neuron model 2 are coherently combined, and the second modulation result of wavelength 1 is that the light phase remains unchanged; when the photon intelligent computing neuron model 1 receives ambient light with an intensity of 1, the first modulation result of wavelength 1 is π, and when the photon intelligent computing neuron model 2 receives ambient light with an intensity of 0, the first modulation result of wavelength 1 is 0, then the first modulation results corresponding to wavelength 1 after the photon intelligent computing neuron model 1 and the photon intelligent computing neuron model 2 are coherently combined, and the second modulation result of wavelength 1 is that the light phase changes π. Based on this, the phase value of wavelength 1 can reflect the difference in the intensity of the ambient light received by the photon intelligent computing neuron model 1 and the photon intelligent computing neuron model 2.

[0034] In one embodiment of the present disclosure, after obtaining the second modulation result of the same wavelength in the preset incident light through the above steps, the classification result of the first image can be determined based on the second modulation result of the same wavelength in the preset incident light.

[0035] Specifically, in one embodiment of the present disclosure, the method for determining the classification result of the first image based on the second modulation result of the same wavelength in the preset incident light may include: obtaining the corresponding light calculation result based on the second modulation result of the same wavelength in the preset incident light, and determining the classification result of the first image based on the correspondence between the characteristics of different wavelengths in the light calculation result and the classification result.

[0036] In one embodiment of the present disclosure, the second modulation results of different wavelengths in the preset incident light are determined as the corresponding light calculation results. Also, in one embodiment of the present disclosure, the correspondence between the features of different wavelengths and the classification results can be set in advance through manual experience.

[0037] For example, in one embodiment of the present disclosure, assuming that wavelength 1 corresponds to image H, when the amplitude of wavelength 1 in the light calculation result is the largest, the classification result of the first image corresponding to the light calculation result is image H.

[0038] In the disclosed embodiment, the above-mentioned image classification method processes the spatial intensity image through the photon intelligent computing neuron model array, makes full use of the coherence and interference effects, realizes the parallel processing of multiple features, and thus improves the image recognition efficiency. At the same time, the photon intelligent computing neuron model can be applied to edge computing, intelligent robots and other fields.

[0039] In order to implement the above embodiment, Figure 3 The present disclosure also proposes an image classification system, which may include: An acquisition module 301 is used to acquire natural light and preset incident light of a first image; The first modulation module 302 is used to input the natural light of the first image and the preset incident light into the photon intelligent sensing neuron model array to obtain the first modulation results corresponding to different wavelengths in the preset incident light; The second modulation module 303 is used to combine the first modulation results corresponding to the same wavelength to obtain the second modulation result of the same wavelength in the preset incident light; The determination module 304 is used to determine the classification result of the first image based on the second modulation result of the same wavelength in the preset incident light.

[0040] In one embodiment of the present disclosure, the photon intelligent computing neuron model array is divided into different groups, and different groups of photon intelligent computing neuron models correspond to different wavelengths in the preset incident light.

[0041] In one embodiment of the present disclosure, the first modulation module is specifically configured to: The natural light and the preset incident light of the first image are input into the photon intelligent computing neuron model array, and each photon intelligent computing neuron model obtains a first modulation result of the corresponding wavelength based on the natural light and the preset incident light of the first image in the target area where the photon intelligent computing neuron model is located; Based on the first modulation result of the wavelength corresponding to each photon intelligent sensing neuron model, the first modulation results corresponding to different wavelengths in the preset incident light are obtained.

[0042] In one embodiment of the present disclosure, the above-mentioned photon intelligent computing neuron model includes a focusing module, a modulation module and an input module; the above-mentioned first modulation module is specifically used to: Inputting preset incident light through an input module; Acquire the natural light of the first image in the target area where the photon intelligent sensing neuron model is located through a focusing module, and focus the natural light into the modulation module; The modulation module modulates the light focused by the focusing module with the preset incident light to obtain a first modulation result of the corresponding wavelength.

[0043] In one embodiment of the present disclosure, the above-mentioned determination module is specifically used to: Based on a second modulation result of the same wavelength in the preset incident light, a corresponding light calculation result is obtained; The classification result of the first image is determined based on the correspondence between the features of different wavelengths in the light calculation result and the classification result.

[0044] In the disclosed embodiment, the above-mentioned image classification system processes the spatial intensity image through the photon intelligent computing neuron model array, makes full use of the coherence and interference effects, realizes the parallel processing of multiple features, and thus improves the image recognition efficiency. At the same time, the photon intelligent computing neuron model can be applied to edge computing, intelligent robots and other fields.

[0045] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this disclosure shall comply with the relevant laws and regulations and shall not violate public order and good morals.

[0046] It should be noted that personal information from users should be collected for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. In addition, such collection / sharing should be carried out after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign the agreement / authorization including authorization of relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others who have access to personal information data comply with its privacy policy and procedures.

[0047] The present disclosure anticipates providing implementation schemes for users to selectively block the use or access of personal information data. That is, the present disclosure anticipates providing hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, risks can be minimized by limiting data collection and deleting the data. In addition, when applicable, such personal information is de-identified to protect the privacy of the user.

[0048] The acquisition, transmission, storage, use, and processing of data in the technical solution disclosed in this disclosure are in compliance with the relevant provisions of national laws and regulations.

[0049] It should be noted that in the embodiments of the present disclosure, certain software, components, models and other existing solutions in the industry may be mentioned, which should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0050] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they contradict each other.

[0051] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, "plurality" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0052] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present disclosure belong.

[0053] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.

[0054] It should be understood that the various parts of the present disclosure can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0055] A person skilled in the art may understand that all or part of the steps in the above-mentioned embodiment method may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0056] In addition, each functional unit in each embodiment of the present disclosure may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0057] The storage medium mentioned above may be a read-only memory, a disk or an optical disk, etc. Although the embodiments of the present disclosure have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present disclosure. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present disclosure.

Claims

1. A photon intelligent computing neuron model, characterized in that: The model includes a focusing module, a modulation module and an input module, wherein: The input module is used to input preset incident light; The focusing module is used to obtain natural light of the first image in the target area and focus the natural light into the modulation module; The modulation module is connected to the focusing module and the input module, and is used for acquiring the light focused by the focusing module and modulating it with the preset incident light to obtain a first modulation result.

2. The model according to claim 1, characterized in that The modulation module is specifically used to modulate the acquired light focused by the focusing module with the same wavelength as the preset incident light to obtain a first modulation result.

3. The model according to claim 2, characterized in that The step of modulating the acquired light focused by the focusing module with the same wavelength as the preset incident light to obtain a first modulation result includes: The light focused by the focusing module is absorbed by the silicon waveguide to generate photogenerated carriers; Based on the high Q value and the refractive index of the photogenerated carriers, phase and / or amplitude modulation is performed on the same wavelength in the preset incident light to obtain a first modulation result.

4. The photon intelligent computing neuron model according to any one of claims 1 to 3, characterized in that: Used for image classification.

5. An image classification method, characterized in that: include: Acquire natural light and preset incident light of a first image; Inputting the natural light of the first image and the preset incident light into the photon intelligent sensing neuron model array to obtain first modulation results corresponding to different wavelengths in the preset incident light; Combining the first modulation results corresponding to the same wavelength to obtain a second modulation result of the same wavelength in the preset incident light; Based on a second modulation result of the same wavelength in the preset incident light, a classification result of the first image is determined.

6. The method according to claim 5, characterized in that The photon intelligent computing neuron model array is divided into different groups, and different groups of photon intelligent computing neuron models correspond to different wavelengths in the preset incident light.

7. The method according to claim 6, characterized in that The step of inputting the natural light of the first image and the preset incident light into the photon intelligent sensing neuron model array to obtain the first modulation results corresponding to different wavelengths in the preset incident light includes: The natural light of the first image and the preset incident light are input into the photon intelligent computing neuron model array, and each photon intelligent computing neuron model obtains a first modulation result of the corresponding wavelength based on the natural light of the first image and the preset incident light in the target area where the photon intelligent computing neuron model is located; Based on the first modulation result of the wavelength corresponding to each photon intelligent sensing neuron model, the first modulation results corresponding to different wavelengths in the preset incident light are obtained.

8. The method according to claim 7, characterized in that The photon intelligent sensing neuron model includes a focusing module, a modulation module and an input module; the first modulation result of the corresponding wavelength is obtained based on the natural light of the first image in the target area where the photon intelligent sensing neuron model is located and the preset incident light, including: Inputting a preset incident light through the input module; Acquire the natural light of the first image in the target area where the photon intelligent sensing neuron model is located through a focusing module, and focus the natural light into the modulation module; The modulation module modulates the light focused by the focusing module and the preset incident light to obtain a first modulation result of a corresponding wavelength.

9. The method according to claim 5, characterized in that The determining the classification result of the first image based on the second modulation result of the same wavelength in the preset incident light includes: Based on the second modulation result of the same wavelength in the preset incident light, a corresponding light calculation result is obtained; Based on the correspondence between the features of different wavelengths in the light calculation result and the classification result, the classification result of the first image is determined.

10. A computer storage medium, wherein: The computer storage medium stores computer executable instructions; after the computer executable instructions are executed by the processor, the method according to any one of claims 5 to 9 can be implemented.

Images