Image recognition processing method and device and terminal equipment

Abstract

The invention provides an image recognition processing method and device and terminal equipment, and is suitable for the technical field of data processing, and the method comprises the steps: carrying out the feature extraction processing of visual function information based on a preset visual function feature extraction vector, and obtaining a plurality of pieces of visual function feature extraction information; and based on a preset human body eye image recognition model, according to the multiple pieces of visual function feature extraction information and the multiple pieces of human body eye image information, calculating and obtaining multiple pieces of human body eye image recognition information. The adaptability of the generated visual function training image and individual differences is improved, and the pertinence and effectiveness of visual function training are improved.

Description

Technical Field

[0001] This application belongs to the field of data processing technology, and in particular relates to image recognition processing methods, apparatus and terminal equipment. Background Technology

[0002] With the rapid development of digital therapy in the field of visual health, visual function training has become an important means to improve visual function deficits such as amblyopia, strabismus, myopia, and eye strain.

[0003] In existing technologies, single or partial visual function information is obtained through specialized testing equipment, thereby outputting limited visual function assessment results or basic eye structure parameters, providing a reference for preliminary visual function training.

[0004] However, existing technologies cannot accurately reflect the relationship between human visual function and eye structure, which in turn affects the pertinence and effectiveness of visual function training programs. Summary of the Invention

[0005] In view of this, embodiments of this application provide an image recognition processing method, apparatus, and terminal device, aiming to solve the problem of insufficient accurate data support for visual function training in the prior art.

[0006] The first aspect of this application provides an image recognition processing method, including:

[0007] Acquire multiple visual function information and multiple human eye image information;

[0008] Based on multiple preset visual function feature extraction vectors, feature extraction processing is performed on the multiple visual function information to obtain multiple visual function feature extraction information.

[0009] Based on a preset human eye image recognition model, multiple human eye image recognition information is calculated according to the information extracted from multiple visual function features and multiple human eye image information.

[0010] A second aspect of this application provides an image recognition processing apparatus, comprising:

[0011] The information acquisition module is used to acquire multiple visual function information and multiple human eye image information;

[0012] The visual function feature extraction information generation module is used to perform feature extraction processing on the multiple visual function information based on multiple preset visual function feature extraction vectors to obtain multiple visual function feature extraction information.

[0013] The human eye image recognition information generation module is used to calculate multiple human eye image recognition information based on a preset human eye image recognition model, information extracted from multiple visual function features and multiple human eye image information.

[0014] A third aspect of this application provides a terminal device, the terminal device including a memory and a processor, the memory storing a computer program executable on the processor, the processor executing the computer program to implement the steps of the image recognition processing method described in the first aspect above.

[0015] A fourth aspect of this application provides a computer-readable storage medium, comprising: storing a computer program, wherein when executed by a processor, the computer program implements the steps of the image recognition processing method described in the first aspect above.

[0016] The beneficial effects of this application embodiment compared with the prior art are: this application deeply mines the key features related to training in visual function data, ensures the targeted analysis of visual functions such as human eye accommodation, effectively integrates visual function features with eye structure features, thereby accurately reflecting the overall functional state of the human eye, improving the adaptability of the generated visual function training images to individual differences, and improving the targeting and effectiveness of visual function training. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram illustrating the implementation flow of the image recognition processing method provided in Embodiment 1 of this application;

[0019] Figure 2 This is a schematic diagram illustrating the implementation flow of the image recognition processing method provided in Embodiment 2 of this application;

[0020] Figure 3 This is a schematic diagram illustrating the implementation flow of the image recognition processing method provided in Embodiment 3 of this application;

[0021] Figure 4 This is a schematic diagram illustrating the implementation flow of the image recognition processing method provided in Embodiment 4 of this application;

[0022] Figure 5 This is a schematic diagram illustrating the implementation flow of the image recognition processing method provided in Embodiment 5 of this application;

[0023] Figure 6 This is a schematic diagram of the structure of the image recognition processing device provided in the embodiments of this application;

[0024] Figure 7 This is a schematic diagram of the terminal device provided in the embodiments of this application. Detailed Implementation

[0025] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0026] To illustrate the technical solution described in this application, specific embodiments are provided below.

[0027] Figure 1 A flowchart illustrating the image recognition processing method provided in Embodiment 1 of this application is shown, and is described in detail below:

[0028] Step S101: Obtain multiple visual function information and multiple human eye image information.

[0029] In this embodiment, visual function information refers to a collection of various information related to human visual function, including positive and negative accommodation data and eye position fusion data, which reflect the human eye's accommodation ability and binocular coordination ability. This information can be obtained by using professional visual function testing equipment to test the human eye's accommodation ability, eye position, and fusion function. Human eye image information refers to multiple sets of human eye structure images captured by imaging equipment. This information can be obtained by capturing images of the human eye from different angles or states using high-resolution eye imaging equipment to clearly present the overall structure of the eyeball and key eye tissues.

[0030] In this embodiment, visual function information may include positive and negative accommodative function information of the human eye. Positive accommodative function information refers to information reflecting the eye's accommodative ability when focusing on near objects. This includes the accommodative amplitude, sensitivity, and accommodative limits achievable by the eye to clearly see near objects. This can be achieved by having the subject focus on near targets at different distances using specialized visual function testing equipment, and recording various parameters during the accommodative process. Negative accommodative function information refers to information reflecting the eye's ability to relax its accommodation when focusing on distant objects. This includes the speed, amplitude, and stability of the relaxation process when switching from focusing on a near object to focusing on a distant object. This can also be achieved by having the subject first focus on a near target and then quickly switch to a distant target using specialized visual function testing equipment, and recording various parameters during the relaxation process.

[0031] Step S102: Based on multiple preset visual function feature extraction vectors, feature extraction processing is performed on the multiple visual function information to obtain multiple visual function feature extraction information.

[0032] In this embodiment, the multiple preset visual function feature extraction vectors can be manually preset. Multiple visual function information can be matched with multiple preset visual function feature extraction vectors, so that each piece of visual function information corresponds to a suitable visual function feature extraction vector. Then, the suitable visual function feature extraction vector is used to perform targeted feature filtering and quantization transformation on the corresponding visual function information, extracting key feature parameters related to visual function training from the visual function information. Subsequently, the extracted key feature parameters are integrated and standardized to eliminate feature differences between different visual function information, finally obtaining multiple visual function feature extraction information to accurately reflect the core features of multiple visual function information.

[0033] Step S103: Based on the preset human eye image recognition model, multiple human eye image recognition information is calculated according to the information extracted from the multiple visual function features and the multiple human eye image information.

[0034] In this embodiment, the preset human eye image recognition model can be manually preset, a trained CNN model, a YOLO model, or a U-Net model. First, multiple visual function feature extraction information and multiple human eye image information can be input into the preset human eye image recognition model to complete the initial integration and verification of multi-dimensional input data, ensuring that the data format meets the model processing requirements. Then, image preprocessing is performed through the preset human eye image recognition model, including image noise reduction, contour enhancement, and key region segmentation, accurately locating relevant regions of the eye structure. Subsequently, the preprocessed multiple human eye image information and multiple visual function feature extraction information are correlated and fused to establish the correspondence between eye structure features and visual function features. Then, based on the fused comprehensive information, the preset human eye image recognition model calculates core indicators such as the eyeball axial ratio and refractive axial length to generate preliminary human eye image recognition results. Finally, the preliminary recognition results are verified and optimized to eliminate errors caused by data interference, ultimately obtaining multiple human eye image recognition information with the required accuracy.

[0035] In this embodiment, the preset human eye image recognition model can be trained by a large number of annotated human eye image sample data, corresponding visual function feature extraction information sample data, and human eye image recognition information sample data such as eyeball axial ratio and refractive eye axis that have been professionally tested and confirmed.

[0036] The image recognition processing method provided in this application deeply mines key features related to training in visual function data, ensuring the targeted analysis of visual functions such as human eye accommodation, effectively integrating visual function features with eye structure features, thereby accurately reflecting the overall functional state of the human eye, improving the adaptability of the generated visual function training images to individual differences, and enhancing the targeting and effectiveness of visual function training.

[0037] Figure 2 The flowchart illustrating the image recognition processing method provided in Embodiment 2 of this application is shown. Its difference from Embodiment 1 described above lies in:

[0038] Multiple preset visual function feature extraction vectors include preset visual function feature focusing vectors, preset visual function feature association matching vectors, and preset visual function feature load vectors.

[0039] Step S102 specifically includes:

[0040] Step S201: Convert the format of the multiple visual function information to generate multiple visual function vector information.

[0041] In this embodiment, the acquired multiple visual function information can be converted into a data format that meets the requirements of subsequent vector operations using the bag-of-words model in the Python library. This ensures that the multiple visual function information can be effectively operated with the preset various visual function feature extraction vectors, thereby generating standardized multiple visual function vector information.

[0042] Step S202: Multiply the multiple visual function vector information and the preset visual function feature focusing vector to calculate multiple visual function feature focusing information extraction vectors.

[0043] In this embodiment, the preset visual function feature focusing vector can be preset by humans. Multiple visual function vector information can be multiplied with this vector to accurately locate the core data dimension that is highly related to visual function training among multiple visual function vector information, thereby strengthening the signal strength of key features and generating multiple visual function feature focusing information extraction vectors.

[0044] Step S203: Multiply the multiple visual function vector information and the preset visual function feature association matching vector to calculate the multiple visual function feature association matching information extraction vector.

[0045] In this embodiment, the preset visual function feature association matching vector can be preset by humans. By multiplying multiple visual function vector information with this vector, the inherent correlation between different feature dimensions within multiple visual function vector information can be mined, thereby establishing the correlation mapping between features and generating multiple visual function feature association matching information extraction vectors.

[0046] Step S204: Multiply the multiple visual function vector information and the preset visual function feature load vector to calculate multiple visual function feature load information extraction vectors.

[0047] In this embodiment, the preset visual function feature load vector can be preset by humans. Multiple visual function vector information can be multiplied with this vector to quantify the influence weight of each feature dimension in the multiple visual function vector information on the overall visual function status assessment, thereby highlighting the role of high-weight features and generating multiple visual function feature load information extraction vectors.

[0048] Step S205: Multiply the multiple visual function feature focusing information extraction vectors and the multiple visual function feature association matching information extraction vectors to calculate multiple visual function feature focusing association vectors.

[0049] In this embodiment, multiple visual function feature focusing information extraction vectors and multiple visual function feature association matching information extraction vectors can be multiplied to fuse the focused key features and the relationships between features, thereby integrating them into a feature combination that is both targeted and relevant. This generates multiple visual function feature focusing association vectors to improve the completeness and relevance of visual function feature information.

[0050] Step S206: Multiply the multiple visual function feature focusing correlation vectors and multiple visual function feature load information extraction vectors to obtain multiple visual function feature extraction information.

[0051] In this embodiment, by multiplying multiple visual function feature focus association vectors with multiple visual function feature load information extraction vectors, the fused focus association features can be further filtered and strengthened by combining feature weights, thereby retaining the core features most valuable for visual function training, and thus generating multiple visual function feature extraction information.

[0052] The image recognition processing method provided in this application embodiment achieves layer-by-layer fusion and optimization of features through multiple rounds of vector multiplication, generating accurate and comprehensive visual function feature extraction information, effectively improving the targeting and accuracy of visual function feature extraction, enhancing the fusion effect of visual function features and eye structure features, improving the computational accuracy of multiple human eye image recognition information, and providing better data support for generating visual function training images adapted to individual differences, thereby improving the targeting and effectiveness of visual function training.

[0053] Figure 3 The flowchart illustrating the image recognition processing method provided in Embodiment 3 of this application is shown. Its difference from Embodiment 1 above lies in:

[0054] Multiple preset visual function feature extraction vectors include preset visual function feature dimension positioning vectors, preset visual function feature dimension comparison vectors, and preset visual function feature attribute matching vectors.

[0055] Step S102 specifically includes:

[0056] Step S301: Generate multiple visual function feature dimension positioning information based on the multiple visual function information and the preset visual function feature dimension positioning vector.

[0057] In this embodiment, the preset visual function feature dimension positioning vector can be manually set. It can be achieved by performing correlation analysis between multiple visual function information and the preset visual function feature dimension positioning vector, or by using a multiplication method to accurately locate key data dimensions closely related to visual function training from multiple visual function information, and then extracting and organizing these key data dimensions to generate multiple visual function feature dimension positioning information.

[0058] Step S302: Generate multiple visual function feature dimension comparison information based on the multiple visual function information and the preset visual function feature dimension comparison vector.

[0059] In this embodiment, the preset visual function feature dimension comparison vector can be manually preset. It can be achieved by comparing multiple visual function information with the preset visual function feature dimension comparison vector using a specific algorithm. This allows for in-depth analysis of the potential correlations between different feature dimensions within multiple visual function information. Furthermore, by sorting and summarizing these correlations, multiple visual function feature dimension comparison information is generated.

[0060] Step S303: Generate multiple visual function feature attribute matching information based on the multiple visual function information and the preset visual function feature attribute matching vector.

[0061] In this embodiment, the preset visual function feature attribute matching vector can be manually preset. It can be achieved by using a specialized matching algorithm to match multiple visual function information with the preset visual function feature attribute matching vector, which can accurately determine the attribute category of each feature in the multiple visual function information. Then, the attribute category is recorded and organized in detail to generate multiple visual function feature attribute matching information.

[0062] Step S304: Generate multiple visual function feature dimension matching information based on the multiple visual function feature dimension positioning information and the multiple visual function feature attribute matching information.

[0063] In this embodiment, multiple visual function feature dimension positioning information can be multiplied with multiple visual function feature attribute matching information, and the multiplication result can be used as multiple visual function feature dimension matching information.

[0064] Step S305: Generate multiple visual function feature comparison information based on the multiple visual function feature dimension comparison information and the multiple visual function feature dimension matching information.

[0065] In this embodiment, multiple visual function feature dimension comparison information can be multiplied with multiple visual function feature dimension matching information, and the multiplication result can be used as multiple visual function feature comparison information.

[0066] Step S306: The multiple visual function feature comparison information and the multiple visual function feature dimensional positioning information are added together to obtain multiple visual function feature enhancement information.

[0067] In this embodiment, multiple visual function feature comparison information and multiple visual function feature dimension positioning information can be used to enhance the signal strength of key features, while highlighting the role of important feature dimensions in the overall visual function features, so that the summation result can be used as multiple visual function feature comparison information.

[0068] Step S307: Based on the multiple visual function feature enhancement information, multiple visual function feature dimension comparison information, and multiple visual function feature attribute matching information, multiple visual function feature extraction information are obtained.

[0069] In this embodiment, multiple visual function feature enhancement information, multiple visual function feature dimension comparison information, and multiple visual function feature attribute matching information can be weighted and fused. The weights can be set manually, and the result of the weighted sum is used as the extracted information of multiple visual function features.

[0070] The image recognition processing method provided in this application realizes multi-dimensional analysis and feature extraction of visual function information, effectively improves the accuracy and comprehensiveness of visual function feature extraction, strengthens the fusion effect of visual function features and eye structure features, significantly improves the calculation accuracy of multiple human eye image recognition information, and provides higher quality and more accurate data support for generating visual function training images that are highly adapted to individual differences, thereby powerfully improving the pertinence and effectiveness of visual function training.

[0071] Figure 4 The flowchart illustrating the image recognition processing method provided in Embodiment 4 of this application is shown. The difference between this method and Embodiment 3 is that step S307 specifically includes:

[0072] Step S401: Generate multiple visual function feature enhancement comparison information based on the multiple visual function feature enhancement information and the multiple visual function feature dimension comparison information.

[0073] In this embodiment, multiple visual function feature enhancement information and multiple visual function feature dimension comparison information can be multiplied together. First, the key feature signals after enhancement are retained, and then the potential correlation between feature dimensions is combined for deep fusion, thereby enhancing the accuracy and recognizability of the associated features, and thus generating multiple visual function feature enhancement comparison information.

[0074] Step S402: Generate multiple visual function feature enhancement matching information based on the multiple visual function feature enhancement information and the multiple visual function feature attribute matching information.

[0075] In this embodiment, multiple visual function feature enhancement information and multiple visual function feature attribute matching information can be multiplied together. First, the enhanced key feature dimensions are retained, and then the corresponding association is made with the clear attribute categories of each feature. This ensures that the enhanced features and attribute information are accurately matched, thereby generating multiple visual function feature enhancement matching information.

[0076] Step S403: Based on the multiple visual function feature enhancement comparison information and the multiple visual function feature enhancement matching information, multiple visual function feature extraction information is obtained.

[0077] In this embodiment, multiple visual function feature enhancement comparison information and multiple visual function feature enhancement matching information can be weighted and fused. The weights can be set manually. First, the enhancement association attribute and enhancement matching attribute of the feature are integrated, and then the weights of the two types of feature information are balanced to eliminate feature redundancy and data interference, thereby generating multiple visual function feature extraction information.

[0078] The image recognition processing method provided in this application improves the accuracy and relevance of extracting information from multiple visual function features, enhances the fusion effect with multiple human eye image information, significantly improves the calculation accuracy of multiple human eye image recognition information, provides high-quality and accurate data support for generating visual function training images that are highly adapted to individual differences, and effectively improves the relevance and effectiveness of visual function training.

[0079] Figure 5 The flowchart illustrating the image recognition processing method provided in Embodiment 5 of this application is shown. The difference between this method and Embodiment 1 is that, after step S103, the method further includes:

[0080] Step S501: Based on the preset visual function training image generation model, visual function training image information is generated according to the multiple human eye image recognition information.

[0081] In this embodiment, the preset visual function training image generation model can be manually preset, and can be an LLM model, an AI model, or a GAN model. It can use multiple human eye image recognition information as input information to the preset visual function training image generation model, and then use the preset visual function training image generation model to plan and generate elements such as image content, structure, and color from multiple dimensions, thereby generating visual function training image information for output.

[0082] Following step S501, the method further includes:

[0083] Step S502: In response to the generation of the visual function training image information, acquire multiple post-training visual function information.

[0084] In this embodiment, after the visual function training image information is generated, it is necessary to obtain information on the changes in visual function status after training to provide a basis for subsequent optimization and adjustment. Professional visual function testing equipment can be used to conduct a comprehensive visual function test on the trainees who have undergone visual function training. First, positive and negative accommodation data testing can be performed, where the trainee fixates on targets at different distances. The equipment records various parameters of their accommodative ability when fixating on near objects, such as accommodative amplitude, accommodative sensitivity, and accommodative limits (positive accommodation data), and the speed, amplitude, and stability of accommodative relaxation when switching from fixating on near objects to fixating on distant objects (negative accommodation data). Then, eye position fusion data testing can be performed. Using specific targets and prisms, and following a standard testing procedure, the trainee's eye position at different distances, as well as key indicators such as fusion range, blur point, break point, and recovery point during binocular fusion, can be measured. The positive and negative accommodation data, eye position fusion data, and other visual function-related data obtained from the above tests can then be summarized and organized to generate multiple sets of post-training visual function information.

[0085] Step S503: The multiple training post-visual function information is treated as multiple visual function information, and the process returns to step S102.

[0086] In this embodiment, multiple post-training visual function information can be replaced with multiple new visual function information, which are then re-matched with multiple preset visual function feature extraction vectors. This ensures that each post-training visual function information corresponds to an appropriate visual function feature extraction vector. The appropriate visual function feature extraction vector then performs targeted feature filtering and quantization transformation on the corresponding post-training visual function information, mining new key visual function feature parameters after training. Subsequently, these newly extracted key feature parameters are integrated and standardized to eliminate feature differences between different post-training visual function information, thereby obtaining multiple visual function feature extraction information again. This information is used to continuously optimize and deeply mine visual function information, providing data that better reflects the actual training effect for subsequent processes such as calculating human eye image recognition information based on visual function feature extraction information and human eye image information, and generating visual function training image information. This continuously improves the adaptability of visual function training images to individual differences, and continuously enhances the targeting and effectiveness of visual function training.

[0087] The image recognition processing method provided in this application continuously iterates and updates visual function-related information, deeply explores the potential connection between visual function data and eye structure features, effectively improves the accuracy and comprehensiveness of visual function feature extraction, strengthens the fusion effect of visual function features and eye structure features, greatly improves the calculation accuracy of multiple human eye image recognition information, and provides higher quality and more accurate data support for generating visual function training images that are highly adapted to individual differences, thereby significantly improving the pertinence and effectiveness of visual function training.

[0088] Corresponding to the method in the above embodiments, Figure 6 A structural block diagram of an image recognition processing apparatus provided in an embodiment of this application is shown. For ease of explanation, only the parts related to the embodiments of this application are shown. Figure 6 The image recognition processing device in the example can be the execution subject of the image recognition processing method provided in the aforementioned embodiment 1.

[0089] Reference Figure 6 The image recognition processing device includes:

[0090] Information acquisition module 610 is used to acquire multiple visual function information and multiple human eye image information;

[0091] The visual function feature extraction information generation module 620 is used to perform feature extraction processing on the multiple visual function information based on multiple preset visual function feature extraction vectors to obtain multiple visual function feature extraction information.

[0092] The human eye image recognition information generation module 630 is used to calculate multiple human eye image recognition information based on a preset human eye image recognition model, according to the information extracted from the multiple visual function features and multiple human eye image information.

[0093] The process by which each module in the image recognition processing device provided in this application implements its respective function can be specifically referred to the foregoing. Figure 1 The description of Embodiment 1 shown will not be repeated here.

[0094] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0095] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0096] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0097] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0098] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance. It should also be understood that although the terms "first," "second," etc., are used in the text to describe various elements in some embodiments of this application, these elements should not be limited by these terms. These terms are merely used to distinguish one element from another. For example, a first table may be named a second table, and similarly, a second table may be named a first table, without departing from the scope of the various described embodiments. Both the first table and the second table are tables, but they are not the same table.

[0099] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0100] The image recognition processing method provided in this application embodiment can be applied to terminal devices such as mobile phones, tablets, wearable devices, vehicle devices, laptops, and super mobile personal computers. This application embodiment does not impose any restrictions on the specific type of terminal device.

[0101] For example, the terminal device may be a station in a WLAN, a cellular phone, a cordless phone, a session initiation protocol phone, a wireless local loop station, a personal digital processing device, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an in-vehicle device, a vehicle-to-everything (V2X) terminal, a computer, a laptop computer, a handheld communication device, a handheld computing device, a satellite wireless device, a wireless modem card, and / or other devices for communication over a wireless system, as well as next-generation communication systems, such as mobile terminals in 5G networks or mobile terminals in future evolved public terrestrial mobile networks.

[0102] Figure 7 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. For example... Figure 7 As shown, the terminal device 7 of this embodiment includes: at least one processor 70 ( Figure 7 (Only one is shown in the image) A memory 71 stores a computer program 72 that can run on the processor 70. When the processor 70 executes the computer program 72, it implements the steps in the various image recognition processing method embodiments described above, for example... Figure 1 Steps S101 to S103 are shown. Alternatively, when the processor 70 executes the computer program 72, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 6 The functions of modules 610 to 630 are shown.

[0103] The terminal device 7 can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor 70 and a memory 71. Those skilled in the art will understand that... Figure 7 This is merely an example of terminal device 7 and does not constitute a limitation on terminal device 7. It may include more or fewer components than shown, or combine certain components, or different components. For example, the terminal device may also include input transmitting devices, network access devices, buses, etc.

[0104] The processor 70 may be a central processing unit, or it may be other general-purpose processors, digital signal processors, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0105] In some embodiments, the memory 71 may be an internal storage unit of the terminal device 7, such as a hard disk or memory of the terminal device 7. The memory 71 may also be an external storage device of the terminal device 7, such as a plug-in hard disk or smart memory card equipped on the terminal device 7. Furthermore, the memory 71 may include both internal and external storage units of the terminal device 7. The memory 71 is used to store operating systems, applications, bootloaders, data, and other programs, such as the program code of the computer program. The memory 71 can also be used to temporarily store data that has been sent or will be sent.

[0106] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0107] This application also provides a terminal device, which includes at least one memory, at least one processor, and a computer program stored in the at least one memory and executable on the at least one processor. When the processor executes the computer program, it causes the terminal device to implement the steps in any of the above method embodiments.

[0108] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0109] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0110] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0111] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0112] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0113] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0114] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An image recognition processing method, characterized in that, include: Acquire multiple visual function information and multiple human eye image information; Based on multiple preset visual function feature extraction vectors, feature extraction processing is performed on the multiple visual function information to obtain multiple visual function feature extraction information. Based on a preset human eye image recognition model, multiple human eye image recognition information is calculated according to the information extracted from multiple visual function features and multiple human eye image information.

2. The image recognition processing method as described in claim 1, characterized in that, The visual function information includes positive accommodation function information and negative accommodation function information of the human eye.

3. The image recognition processing method as described in claim 1, characterized in that, Multiple preset visual function feature extraction vectors include preset visual function feature focusing vectors, preset visual function feature association matching vectors, and preset visual function feature load vectors. The step of performing feature extraction processing on the multiple visual function information based on multiple preset visual function feature extraction vectors to obtain multiple visual function feature extraction information specifically includes: The multiple visual function information is converted into a format to generate multiple visual function vector information; Multiple visual function vector information and preset visual function feature focusing vectors are multiplied together to calculate multiple visual function feature focusing information extraction vectors. The multiple visual function vectors and the preset visual function feature association matching vectors are multiplied together to calculate the multiple visual function feature association matching information extraction vectors. Multiple visual function vector information and preset visual function feature load vectors are multiplied together to calculate multiple visual function feature load information extraction vectors. Multiple visual function feature focusing information extraction vectors and multiple visual function feature association matching information extraction vectors are multiplied together to calculate multiple visual function feature focusing association vectors. The multiple visual function feature focusing correlation vectors and multiple visual function feature load information extraction vectors are multiplied together to obtain multiple visual function feature extraction information.

4. The image recognition processing method as described in claim 1, characterized in that, Multiple preset visual function feature extraction vectors include preset visual function feature dimension positioning vectors, preset visual function feature dimension comparison vectors, and preset visual function feature attribute matching vectors. The step of performing feature extraction processing on the multiple visual function information based on multiple preset visual function feature extraction vectors to obtain multiple visual function feature extraction information specifically includes: Based on the multiple visual function information and the preset visual function feature dimension positioning vector, multiple visual function feature dimension positioning information is generated. Based on the multiple visual function information and the preset visual function feature dimension comparison vector, multiple visual function feature dimension comparison information is generated. Based on the multiple visual function information and the preset visual function feature attribute matching vector, multiple visual function feature attribute matching information is generated. Based on the multiple visual function feature dimension positioning information and the multiple visual function feature attribute matching information, multiple visual function feature dimension matching information is generated. Based on the comparison information of multiple visual function feature dimensions and the matching information of multiple visual function feature dimensions, multiple visual function feature comparison information is generated; Multiple visual function feature enhancement information is obtained by adding the multiple visual function feature comparison information and the multiple visual function feature dimensional positioning information; Based on the multiple visual function feature enhancement information, multiple visual function feature dimension comparison information, and multiple visual function feature attribute matching information, multiple visual function feature extraction information is obtained.

5. The image recognition processing method as described in claim 4, characterized in that, The step of obtaining multiple visual function feature extraction information based on the multiple visual function feature enhancement information, multiple visual function feature dimension comparison information, and multiple visual function feature attribute matching information specifically includes: Based on the multiple visual function feature enhancement information and the multiple visual function feature dimension comparison information, multiple visual function feature enhancement comparison information is generated; Based on the multiple visual function feature enhancement information and the multiple visual function feature attribute matching information, multiple visual function feature enhancement matching information is generated; Based on the multiple visual function feature enhancement comparison information and the multiple visual function feature enhancement matching information, multiple visual function feature extraction information is obtained.

6. The image recognition processing method as described in claim 1, characterized in that, After the step of calculating multiple human eye image recognition information based on the multiple visual function feature extraction information and multiple human eye image information according to the preset human eye image recognition model, the method further includes: Based on a preset visual function training image generation model, visual function training image information is generated according to the recognition information of the multiple human eye images.

7. The image recognition processing method as described in claim 6, characterized in that, After the step of generating visual function training image information based on the multiple human eye image recognition information using the preset visual function training image generation model, the method further includes: In response to the generation of the visual function training image information, multiple post-training visual function information are acquired; The multiple trained post-visual function information is treated as multiple visual function information and returned to the step of performing feature extraction processing on the multiple visual function information based on multiple preset visual function feature extraction vectors to obtain multiple visual function feature extraction information.

8. An image recognition processing apparatus, characterized in that, include: The information acquisition module is used to acquire multiple visual function information and multiple human eye image information; The visual function feature extraction information generation module is used to perform feature extraction processing on the multiple visual function information based on multiple preset visual function feature extraction vectors to obtain multiple visual function feature extraction information. The human eye image recognition information generation module is used to calculate multiple human eye image recognition information based on a preset human eye image recognition model, information extracted from multiple visual function features and multiple human eye image information.

9. A terminal device, characterized in that, The terminal device includes a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images