Electrical signature platform image recognition method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to an electrical signature platform image recognition method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: determining at least two target identification interfaces according to electrical signing work card types corresponding to a plurality of completed photos; according to the target identification service corresponding to each target identification interface, identifying the completed photos belonging to the same electrical signing work card type to obtain a candidate identification result; if the candidate recognition result shows that the unsafe factors exist in the photo, review confirmation information is obtained; and confirming the candidate recognition result according to the review confirmation information to obtain a target recognition result. The method can adapt to the multi-work-item, multi-target and high-concurrency detection scene of the maintenance electronic signature platform.

Description

Technical Field

[0001] The present application relates to the field of automated detection technology, and in particular to an electronic signature platform image recognition method, device, computer equipment, computer-readable storage medium and computer program product. Background Art

[0002] With the development of automated inspection technology, some static image recognition technologies have been applied in the field of aircraft maintenance to achieve automated inspection. However, existing static image recognition technologies generally only support localized operation on edge devices, usually relying on dedicated edge devices for inspection, and most of them are carried out in fixed scenarios. In this inspection scenario, the maintenance electronic signature platform requires a large number of tablets and different maintenance projects have different inspection targets. If each maintenance project requires training a corresponding image recognition model, there are problems such as difficult model deployment, slow local tablet inference, and low inference accuracy. Therefore, the current inspection technology cannot adapt to the maintenance electronic signature platform's multi-work project, multi-target, and high-concurrency inspection scenarios. Summary of the invention

[0003] Based on this, it is necessary to provide an image recognition method, device, computer equipment, computer-readable storage medium and computer program product for an electronic signature platform that can adapt to the multi-work items, multi-target and high-concurrency detection scenarios of the maintenance electronic signature platform in response to the above-mentioned technical problems.

[0004] In a first aspect, the present application provides an electronic signature platform image recognition method, comprising:

[0005] Determine at least two target recognition interfaces according to the types of electronically signed work cards corresponding to the plurality of completed photos;

[0006] According to the target recognition service corresponding to each target recognition interface, the completed photos belonging to the same type of electronically signed work card are recognized to obtain candidate recognition results;

[0007] If the candidate recognition result is that there are unsafe factors in the photo, obtaining review confirmation information;

[0008] The candidate recognition result is confirmed according to the review confirmation information to obtain a target recognition result.

[0009] In one embodiment, the step of determining at least two target identification interfaces according to the types of electronically signed work cards corresponding to the plurality of completed photos includes:

[0010] Determine at least two types of electronically signed work cards based on the plurality of completed photos;

[0011] The at least two target identification interfaces are determined according to at least two types of the electronic signature work cards.

[0012] In one embodiment, the target recognition service corresponding to each target recognition interface recognizes the completed photos belonging to the same type of electronically signed work card to obtain candidate recognition results, including:

[0013] Determining a target recognition model according to the target recognition service;

[0014] The completed photos belonging to the same type of electronically signed work card are identified according to the target recognition model to obtain the candidate recognition results.

[0015] In one embodiment, the method further comprises:

[0016] Obtaining training completion photos and remark information corresponding to the training completion photos;

[0017] Classify and store the training completion photos according to the type of the work card that uploaded the training completion photos to obtain a classification sample graph;

[0018] Performing data amplification on the classification sample graph to obtain an expanded sample graph;

[0019] The original recognition model is trained according to the expanded sample graph and the annotation information to obtain the target recognition model.

[0020] In one embodiment, the training of the original recognition model according to the expanded sample graph and the annotation information to obtain the target recognition model includes:

[0021] Annotating the expanded sample graph to obtain an annotated sample graph;

[0022] Automatically select parameters of the original recognition model according to the labeled sample graph to obtain a candidate recognition model;

[0023] The candidate recognition model is automatically tuned according to the annotation information to obtain the target recognition model.

[0024] In one embodiment, the method further comprises:

[0025] According to the target recognition service corresponding to each target recognition interface, the completed photos belonging to the same type of electronically signed work card are recognized to obtain recognition result label data;

[0026] An identification result graph is drawn based on the identification result label data and the completed photos belonging to the same type of electronically signed work card to obtain an identification result graph, and the identification result graph is displayed.

[0027] In a second aspect, the present application also provides an electronic signature platform image recognition device, including:

[0028] An interface determination module, used to determine at least two target recognition interfaces according to the types of electronically signed work cards corresponding to the multiple completed photos;

[0029] An identification module, used to identify the completed photos belonging to the same type of electronically signed work card according to the target identification service corresponding to each target identification interface, and obtain candidate identification results;

[0030] An acquisition module, used for acquiring review confirmation information if the candidate recognition result indicates that there are unsafe factors in the photo;

[0031] The confirmation module is used to confirm the candidate recognition result according to the review confirmation information to obtain the target recognition result.

[0032] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0033] Determine at least two target recognition interfaces according to the types of electronically signed work cards corresponding to the plurality of completed photos;

[0034] According to the target recognition service corresponding to each target recognition interface, the completed photos belonging to the same type of electronically signed work card are recognized to obtain candidate recognition results;

[0035] If the candidate recognition result is that there are unsafe factors in the photo, obtaining review confirmation information;

[0036] The candidate recognition result is confirmed according to the review confirmation information to obtain a target recognition result.

[0037] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0038] Determine at least two target recognition interfaces according to the types of electronically signed work cards corresponding to the plurality of completed photos;

[0039] According to the target recognition service corresponding to each target recognition interface, the completed photos belonging to the same type of electronically signed work card are recognized to obtain candidate recognition results;

[0040] If the candidate recognition result is that there are unsafe factors in the photo, obtaining review confirmation information;

[0041] The candidate recognition result is confirmed according to the review confirmation information to obtain a target recognition result.

[0042] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0043] Determine at least two target recognition interfaces according to the types of electronically signed work cards corresponding to the plurality of completed photos;

[0044] According to the target recognition service corresponding to each target recognition interface, the completed photos belonging to the same type of electronically signed work card are recognized to obtain candidate recognition results;

[0045] If the candidate recognition result is that there are unsafe factors in the photo, obtaining review confirmation information;

[0046] The candidate recognition result is confirmed according to the review confirmation information to obtain a target recognition result.

[0047] The above-mentioned electronic signature platform image recognition method, device, computer equipment, computer-readable storage medium and computer program product determine at least two target recognition interfaces according to the electronic signature work card types corresponding to multiple completed photos; identify the completed photos belonging to the same electronic signature work card type according to the target recognition service corresponding to each target recognition interface to obtain candidate recognition results; if the candidate recognition result is that there are unsafe factors in the photo, then obtain review confirmation information; confirm the candidate recognition result according to the review confirmation information to obtain the target recognition result. Therefore, by determining at least two target recognition interfaces according to the electronic signature work card types corresponding to multiple completed photos, different types of completed photos can be accurately assigned to different interface services for recognition, and then the completed photos belonging to the same electronic signature work card type are recognized through the target recognition service corresponding to each target recognition interface to obtain candidate recognition results, thereby realizing the electronic signature platform image recognition function with high concurrent recognition through the simultaneous processing of multiple target recognition services. Finally, the recognition accuracy is further improved by reviewing and confirming the completed photos with unsafe factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0049] FIG1 is a diagram showing an application environment of an electronic signature platform image recognition method in one embodiment;

[0050] FIG2 is a schematic diagram of a flow chart of an electronic signature platform image recognition method in one embodiment;

[0051] FIG3 is a schematic diagram of the labeling results of the electronic signature platform image recognition method in one embodiment;

[0052] FIG4 is a schematic diagram of a flow chart of an electronic signature platform image recognition method in one embodiment;

[0053] FIG5 is a diagram showing the recognition results of the electronic signature platform image recognition method in one embodiment;

[0054] FIG6 is a structural block diagram of an electronic signature platform image recognition device in one embodiment;

[0055] FIG. 7 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0057] The electronic signature platform image recognition method provided in the embodiment of the present application can be applied in the application environment shown in FIG. 1. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The server 104 determines at least two target recognition interfaces according to the electronic signature work card types corresponding to the multiple completed photos; the completed photos belonging to the same electronic signature work card type are recognized according to the target recognition service corresponding to each target recognition interface to obtain candidate recognition results; if the candidate recognition result is that there are unsafe factors in the photo, the review confirmation information is obtained; the candidate recognition result is confirmed according to the review confirmation information to obtain the target recognition result. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, projection devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The head mounted device may be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server 104 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0058] In an exemplary embodiment, as shown in FIG2 , a method for image recognition on an electronic signature platform is provided, which is described by taking the method applied to the server 104 in FIG1 as an example, and includes the following steps 202 to 208 . Among them:

[0059] Step 202, determining at least two target recognition interfaces according to the types of electronically signed work cards corresponding to the plurality of completed photos.

[0060] Among them, the completed photo refers to the work scene image uploaded by the staff after completing the corresponding work task to determine whether the corresponding work task has been successfully completed. It should be noted that the staff will use the electronic signature platform to sign the work card after completing the corresponding work task, that is, upload the work completion information through the electronic signature work card. In the process of uploading the work completion information, if a key step requires uploading the completed photo for image recognition, the electronic signature platform will give relevant prompts and wait for the staff to upload. Among them, the electronic signature work card type refers to the work task type corresponding to the completed photo, and each electronic signature work card type corresponds to a work task type.

[0061] In some embodiments, at least two target recognition interfaces can be determined based on the types of electronically signed work cards corresponding to multiple completed photos, so that different target recognition interfaces are called for different types of work to improve recognition efficiency and meet the technical requirements of high-concurrency recognition.

[0062] In some embodiments, at least two target identification interfaces are determined based on the electronic signature work card types corresponding to multiple completed photos, including: determining at least two electronic signature work card types based on multiple completed photos; determining at least two target identification interfaces based on at least two electronic signature work card types.

[0063] In some embodiments, the electronic signature work card type corresponding to each electronic signature work card uploaded for a completed photo can be determined based on multiple completed photos, and if there are repeated electronic signature work card types between multiple completed photos, the same interface is used for processing, and finally the non-duplicate electronic signature work card types corresponding to the multiple completed photos are screened out to obtain at least two electronic signature work card types.

[0064] In some embodiments, the target identification interface can use an API interface, and each target identification interface corresponds to a type of electronic signature card. Therefore, the corresponding target identification interface can be determined according to the type of electronic signature card, so as to accurately and quickly request the target identification service corresponding to the target identification interface. For example, the electronic signature card for installing a diffuser tube corresponds to the diffuser tube screw identification API interface.

[0065] Step 204, identifying the completed photos belonging to the same electronically signed work card type according to the target recognition service corresponding to each target recognition interface to obtain candidate recognition results.

[0066] Among them, completion photos belonging to the same electronically signed work card type are completion photos of the same type.

[0067] In some embodiments, completed photos belonging to the same electronically signed work card type are identified according to the target recognition service corresponding to each target recognition interface to obtain candidate recognition results, including: determining a target recognition model according to the target recognition service; and identifying completed photos belonging to the same electronically signed work card type according to the target recognition model to obtain candidate recognition results.

[0068] Among them, the target recognition model is a recognition model that is trained in advance and deployed on the online electronic signature platform. Each target recognition service corresponds to a target recognition model.

[0069] In some embodiments, a corresponding target recognition model is determined according to each of the multiple target recognition services, and the corresponding completed photos belonging to the same electronic signature work card type are recognized according to the multiple target recognition models to obtain multiple candidate recognition results, and the multiple candidate recognition results are sent to the corresponding electronic signature platform respectively.

[0070] In some embodiments, the method also includes: identifying completed photos belonging to the same electronically signed work card type according to the target recognition service corresponding to each target recognition interface to obtain recognition result label data; drawing a recognition result graph based on the recognition result label data and the completed photos belonging to the same electronically signed work card type to obtain a recognition result graph, and displaying the recognition result graph.

[0071] The recognition result label data is the recognition result data used to mark each recognition point.

[0072] In some embodiments, multiple recognition result label data are respectively used to label the corresponding completed photos of the same electronic signature work card type, to obtain a recognition result graph corresponding to each completed photo, and the recognition result graph is displayed. Specifically, the recognition result graph is shown in FIG3 .

[0073] Step 206: If the candidate recognition result indicates that there are unsafe factors in the photo, then obtain review confirmation information.

[0074] Among them, unsafe factors refer to the corresponding safety issues for each work task. If the target recognition model finds the existence of safety issues during the process of identifying the completed photos, the candidate recognition results will be set as having unsafe factors.

[0075] In some embodiments, if the candidate recognition result is that there are unsafe factors in the photo, it means that the current work task may still have safety hazards after completion. At this time, further investigation of this safety hazard is required. Therefore, a prompt review message is sent to another staff member responsible for verification to obtain further review confirmation information, thereby improving the accuracy of the final recognition result.

[0076] Step 208, confirming the candidate recognition result according to the review confirmation information to obtain the target recognition result.

[0077] The review confirmation information is confirmation information obtained after reviewing and identifying the corresponding completed pictures, which may be approval of completion or disapproval of completion.

[0078] In some embodiments, the candidate recognition result is confirmed based on the review confirmation information. If the review confirmation information is not in agreement with the completion, the target recognition result is set to recognition failure and please take a photo again; if the review confirmation information is in agreement with the completion, the target recognition result is set to the work task completed.

[0079] In the above-mentioned electronic signature platform image recognition method, at least two target recognition interfaces are determined according to the electronic signature work card types corresponding to multiple completed photos; the completed photos belonging to the same electronic signature work card type are recognized according to the target recognition services corresponding to each target recognition interface to obtain candidate recognition results; if the candidate recognition result is that there are unsafe factors in the photo, then the review confirmation information is obtained; the candidate recognition result is confirmed according to the review confirmation information to obtain the target recognition result. Therefore, by determining at least two target recognition interfaces according to the electronic signature work card types corresponding to multiple completed photos, different types of completed photos can be accurately assigned to different interface services for recognition, and then the completed photos belonging to the same electronic signature work card type are recognized through the target recognition services corresponding to each target recognition interface to obtain candidate recognition results, thereby realizing the electronic signature platform image recognition function with high concurrent recognition by processing multiple target recognition services at the same time. Finally, the recognition accuracy is further improved by reviewing and confirming the completed photos with unsafe factors.

[0080] In an exemplary embodiment, as shown in FIG4 , step A includes steps 402 to 404. Among them:

[0081] Step 402: Obtain the training completion photo and the remark information corresponding to the training completion photo.

[0082] Among them, the training completion photos and the corresponding remarks information of the training completion photos are uploaded using the picture remarks function of the electronic signature platform.

[0083] In some embodiments, after completing their work, multiple staff members upload the status of the completed work scene to the backend of the electronic signing platform in the form of picture annotations through the picture annotation function of the electronic signing platform, thereby obtaining completed training photos under different work tasks and the corresponding annotation information of the completed training photos, but not limited to this.

[0084] Step 404 , classify and store the training completion photos according to the type of the work card on which the training completion photos were uploaded, and obtain a classification sample graph.

[0085] Each work card type corresponds to the type of work performed by the staff.

[0086] In some embodiments, the work card type of the uploaded training completion photos is determined through the training completion photos, and then the training completion photos are classified and stored according to different work card types to obtain different types of classification sample images, and the classification sample images are classified and stored for subsequent classification training of the model training process.

[0087] Step 406, perform data augmentation on the classified sample graph to obtain an expanded sample graph.

[0088] Among them, data augmentation refers to expanding the amount of data by rotation, scaling, translation, adding random noise, etc.

[0089] In some embodiments, the classified sample images are first preprocessed, and sample images with more obvious corresponding features in each work task are screened out, and then data amplification is performed on the sample images to obtain expanded sample images, thereby increasing the diversity and quantity of the expanded sample images.

[0090] Step 408, training the original recognition model according to the expanded sample graph and the annotation information to obtain a target recognition model.

[0091] The remark information may be validation set data corresponding to each expanded sample diagram determined by actual maintenance experience.

[0092] In some embodiments, the original recognition model is trained according to the expanded sample graph and the annotation information to obtain a target recognition model, including: labeling the expanded sample graph to obtain a labeled sample graph; automatically selecting parameters of the original recognition model according to the labeled sample graph to obtain a candidate recognition model; and automatically tuning the candidate recognition model according to the annotation information to obtain a target recognition model.

[0093] In some embodiments, taking the task of installing diffuser tube screws as an example, the correctly installed diffuser tube screws can be labeled using the labeling tool of the one-stop AI platform to obtain a labeled sample image for subsequent model training.

[0094] Among them, the one-stop AI platform can adopt Huawei's modelarts platform. The specific annotation sample diagram can be shown in Figure 5. Each diffusion tube screw will be marked with a box. Of course, different types of work tasks use different annotation methods and the annotation content is also different, and is not limited to this.

[0095] In some embodiments, the target detection algorithm of the one-stop AI platform is used to train the original recognition model, and the parameters of the original recognition model are automatically selected by annotating sample images to obtain a candidate recognition model. The candidate recognition model is automatically tuned using the annotation information as a verification set to obtain a target recognition model, thereby realizing automatic learning function and quickly completing the training and online deployment of multiple target recognition models.

[0096] In this embodiment, by obtaining the training completion photos and the remark information corresponding to the training completion photos, the training completion photos are classified and stored according to the work card type of the uploaded training completion photos, a classification sample map is obtained, data is amplified on the classification sample map to obtain an expanded sample map, and the original recognition model is trained according to the expanded sample map and the remark information to obtain a target recognition model. It can reasonably and effectively combine the one-stop AI platform to adopt online training and online deployment. It only needs to train the model for different maintenance projects, and the model is deployed online at one time, which saves a lot of terminal deployment work, and the model iteration does not need to be repeatedly deployed.

[0097] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0098] Based on the same inventive concept, the embodiment of the present application also provides an electronic signature platform image recognition device for implementing the electronic signature platform image recognition method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more electronic signature platform image recognition device embodiments provided below can refer to the limitations of the electronic signature platform image recognition method above, and will not be repeated here.

[0099] In an exemplary embodiment, as shown in FIG6 , an electronic signature platform image recognition device is provided, comprising: an interface determination module 602 , an identification module 604 , an acquisition module 606 and a confirmation module 608 , wherein:

[0100] An interface determination module 602 is used to determine at least two target recognition interfaces according to the types of electronically signed work cards corresponding to the multiple completed photos;

[0101] The recognition module 604 is used to recognize the completed photos belonging to the same electronic signature work card type according to the target recognition service corresponding to each target recognition interface, and obtain a candidate recognition result;

[0102] An acquisition module 606 is used to acquire review confirmation information if the candidate recognition result indicates that there are unsafe factors in the photo;

[0103] The confirmation module 608 is used to confirm the candidate recognition result according to the review confirmation information to obtain the target recognition result.

[0104] In some embodiments, the interface determination module 602 is further used to determine at least two electronic signature work card types based on a plurality of completed photos; and determine at least two target identification interfaces based on the at least two electronic signature work card types.

[0105] In some embodiments, the recognition module 604 is also used to determine a target recognition model according to a target recognition service; and to recognize completion photos belonging to the same electronically signed work card type according to the target recognition model to obtain candidate recognition results.

[0106] In some embodiments, the device also includes: a training module, used to obtain training completion photos and annotation information corresponding to the training completion photos; classify and store the training completion photos according to the type of work card on which the training completion photos were uploaded to obtain a classification sample graph; perform data amplification on the classification sample graph to obtain an expanded sample graph; perform model training on the original recognition model based on the expanded sample graph and the annotation information to obtain a target recognition model.

[0107] In some embodiments, the training module is also used to annotate the expanded sample graph to obtain an annotated sample graph; automatically select parameters of the original recognition model based on the annotated sample graph to obtain a candidate recognition model; and automatically tune the candidate recognition model based on the annotation information to obtain a target recognition model.

[0108] In some embodiments, the device also includes: a display module, which is used to identify the completed photos belonging to the same electronic signature work card type according to the target recognition service corresponding to each target recognition interface to obtain recognition result label data; draw a recognition result graph based on the recognition result label data and the completed photos belonging to the same electronic signature work card type to obtain a recognition result graph, and display the recognition result graph.

[0109] In the above-mentioned electronic signature platform image recognition device, at least two target recognition interfaces are determined according to the electronic signature work card types corresponding to multiple completed photos; the completed photos belonging to the same electronic signature work card type are recognized according to the target recognition services corresponding to each target recognition interface to obtain candidate recognition results; if the candidate recognition result is that there are unsafe factors in the photo, then the review confirmation information is obtained; the candidate recognition result is confirmed according to the review confirmation information to obtain the target recognition result. Therefore, by determining at least two target recognition interfaces according to the electronic signature work card types corresponding to multiple completed photos, different types of completed photos can be accurately assigned to different interface services for recognition, and then the completed photos belonging to the same electronic signature work card type are recognized through the target recognition services corresponding to each target recognition interface to obtain candidate recognition results, thereby realizing the electronic signature platform image recognition function with high concurrent recognition by processing multiple target recognition services at the same time. Finally, the recognition accuracy is further improved by reviewing and confirming the completed photos with unsafe factors.

[0110] Each module in the above-mentioned electronic signature platform image recognition device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0111] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG7. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store target recognition service data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an image recognition method for an electronic signature platform is implemented.

[0112] Those skilled in the art will understand that the structure shown in FIG. 7 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0113] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps of the above-mentioned electronic signature platform image recognition method when executing the computer program.

[0114] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned electronic signature platform image recognition method are implemented.

[0115] In one embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned electronic signature platform image recognition method.

[0116] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0117] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0118] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0119] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An electronic signature platform image recognition method, characterized in that: The method comprises: Determine at least two target recognition interfaces according to the types of electronically signed work cards corresponding to the plurality of completed photos; According to the target recognition service corresponding to each target recognition interface, the completed photos belonging to the same type of electronically signed work card are recognized to obtain candidate recognition results; If the candidate recognition result is that there are unsafe factors in the photo, obtaining review confirmation information; The candidate recognition result is confirmed according to the review confirmation information to obtain a target recognition result.

2. The method according to claim 1, characterized in that The step of determining at least two target identification interfaces according to the types of electronically signed work cards corresponding to the plurality of completed photos includes: Determine at least two types of electronically signed work cards based on the plurality of completed photos; The at least two target identification interfaces are determined according to at least two types of the electronic signature work cards.

3. The method according to claim 1, characterized in that The target recognition service corresponding to each target recognition interface recognizes the completed photos belonging to the same type of electronically signed work card to obtain candidate recognition results, including: Determining a target recognition model according to the target recognition service; The completed photos belonging to the same type of electronically signed work card are identified according to the target recognition model to obtain the candidate recognition results.

4. The method according to claim 3, characterized in that: The method further comprises: Obtaining training completion photos and remark information corresponding to the training completion photos; Classify and store the training completion photos according to the type of the work card that uploaded the training completion photos to obtain a classification sample graph; Performing data amplification on the classification sample graph to obtain an expanded sample graph; The original recognition model is trained according to the expanded sample graph and the annotation information to obtain the target recognition model.

5. The method according to claim 4, characterized in that The performing model training on the original recognition model according to the expanded sample graph and the annotation information to obtain the target recognition model includes: Annotating the expanded sample graph to obtain an annotated sample graph; Automatically select parameters of the original recognition model according to the labeled sample graph to obtain a candidate recognition model; The candidate recognition model is automatically tuned according to the annotation information to obtain the target recognition model.

6. The method according to any one of claims 1 to 5, characterized in that: The method further comprises: According to the target recognition service corresponding to each target recognition interface, the completed photos belonging to the same type of electronically signed work card are recognized to obtain recognition result label data; An identification result graph is drawn based on the identification result label data and the completed photos belonging to the same type of electronically signed work card to obtain an identification result graph, and the identification result graph is displayed.

7. An electronic signature platform image recognition device, characterized in that: The device comprises: An interface determination module, used to determine at least two target recognition interfaces according to the types of electronically signed work cards corresponding to the multiple completed photos; An identification module, used to identify the completed photos belonging to the same type of electronically signed work card according to the target identification service corresponding to each target identification interface, and obtain candidate identification results; An acquisition module, used for acquiring review confirmation information if the candidate recognition result indicates that there are unsafe factors in the photo; The confirmation module is used to confirm the candidate recognition result according to the review confirmation information to obtain the target recognition result.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images