Intelligent audit method, audit system, electronic device and storage medium

Abstract

The present invention provides an intelligent audit method, an audit system, an electronic device and a storage medium, comprising the following steps: a training phase: training a deep learning model according to a received training image and a detection instruction corresponding to the training image, so that the deep learning model can identify the image features corresponding to the detection instruction; an audit phase: obtaining an audited image; calling the deep learning model to audit whether the audited image has the image features; if so, the audit result is qualified; if not, the audit result is unqualified. The present invention can reduce the audit workload and improve work efficiency by deploying an intelligent audit model.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to an intelligent audit method, an audit system, an electronic device and a storage medium. Background Art

[0002] The daily maintenance, upkeep, overhaul, and inspection of equipment are usually assigned to specific field operators in the form of work orders, and the field operators perform field operations according to the instructions of the work orders. Field operations are scattered at many points, with many types of tasks and many work steps. It is difficult to effectively supervise the field work site. Now, it is common to take photos of the work steps and important nodes, and then manually or machine review and cancel the order. However, when manually reviewing massive work orders, the human eye is easily fatigued, and the subtle differences in the on-site photos are difficult to distinguish with the naked eye. This is inefficient, time-consuming, and has a high rate of misjudgment. At present, there is a lack of simple and effective technical solutions to solve such problems. Summary of the invention

[0003] The purpose of the present invention is to provide an intelligent audit method, an audit system, an electronic device and a storage medium, which can reduce the audit workload and improve work efficiency by deploying an intelligent audit model.

[0004] To achieve the above-mentioned purpose, the present invention provides an intelligent audit method, comprising the following steps: training stage: training a deep learning model based on the received training images and the detection instructions corresponding to the training images, so that the deep learning model can recognize the image features corresponding to the detection instructions; audit stage: obtaining the audited image; calling the deep learning model to audit whether the audited image has the image features; if so, the audit result is qualified; if not, the audit result is unqualified.

[0005] Optionally, there are multiple deep learning models, and each of the deep learning models corresponds to an operation step.

[0006] Optionally, a plurality of different operation steps are sorted according to the received operation instructions.

[0007] Optionally, during the review stage, in two adjacent operation steps, if the review result of the deep learning model corresponding to the previous operation step is qualified, entering the next operation step is allowed, otherwise entering the next operation step is prohibited.

[0008] Optionally, there are multiple image features corresponding to each detection instruction. During the review stage, calling the deep learning model to review whether the review image has the image features specifically includes: selecting some of the image features from the multiple image features corresponding to the detection instruction based on the received prompt words; calling the deep learning model to review whether the review image has the image features in this part.

[0009] Optionally, the detection instruction includes an OCR recognition instruction, the image feature includes a text feature, and calling the deep learning model to review whether the audited image has the image feature specifically includes: calling the deep learning model to review whether the audited image has the text feature based on OCR recognition.

[0010] Optionally, after the audit result is unqualified, the deep learning model is retrained according to the received appeal instruction.

[0011] The present invention also provides an audit system, which is applied to any of the above-mentioned intelligent audit methods, including: a receiving module, used to receive the training image and / or the detection instruction corresponding to the training image; a training module, used to train the deep learning model; an acquisition module, used to acquire the audited image; and an execution module, used to call the deep learning model to audit whether the audited image has the image features.

[0012] The present invention also provides an electronic device, comprising: a memory storing a computer program; a processor communicatively connected to the memory, for executing any one of the intelligent audit methods described above when the computer program is called; and a display communicatively connected to the processor and the memory, for displaying a GUI interactive interface related to the intelligent audit method.

[0013] The present invention also provides a storage medium storing a computer program, which implements any of the above-mentioned intelligent audit methods when executed by a processor.

[0014] The intelligent audit method, audit system, electronic device and storage medium provided by the present invention have the following beneficial effects:

[0015] The present invention provides an intelligent audit method, comprising the following steps: training phase: training a deep learning model according to received training images and detection instructions corresponding to the training images, so that the deep learning model can identify the image features corresponding to the detection instructions; audit phase: obtaining the audited image; calling the deep learning model to audit whether the audited image has the image features; if so, the audit result is qualified; if not, the audit result is unqualified. With such a configuration, the present invention can realize the targeted deployment of deep learning models to realize intelligent audit using the training images and detection instructions provided by the user, and the user can deploy the intelligent audit model without coding, which greatly reduces the application threshold of artificial intelligence technology and the audit workload, and improves work efficiency.

[0016] The present invention also provides a retrieval system. Since the retrieval system and the intelligent audit method belong to the same inventive concept, the retrieval system can utilize the training images and detection instructions provided by the user to deploy deep learning models in a targeted manner to achieve intelligent audit. Users can deploy intelligent audit models without code, which greatly reduces the application threshold of artificial intelligence technology and the audit workload, and improves work efficiency.

[0017] The present invention also provides an electronic device. Since the electronic device and the intelligent audit method belong to the same inventive concept, the electronic device can utilize the training images and detection instructions provided by the user to deploy a deep learning model in a targeted manner to achieve intelligent audit. The user can deploy the intelligent audit model without code, which greatly reduces the application threshold of artificial intelligence technology and the audit workload, and improves work efficiency.

[0018] The present invention also provides a storage medium. Since the storage medium and the intelligent audit method belong to the same inventive concept, the storage medium can utilize the training images and detection instructions provided by the user to deploy deep learning models in a targeted manner to achieve intelligent audit. Users can deploy intelligent audit models without code, which greatly reduces the application threshold of artificial intelligence technology and the audit workload, and improves work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of a flow chart of an intelligent audit method provided in accordance with an embodiment of the present invention.

[0020] Figure 2 A schematic diagram of the block structure of an electronic device provided by an embodiment of the present invention.

[0021] The accompanying drawings are denoted as follows:

[0022] 101 - processor; 102 - communication interface; 103 - memory; 104 - communication bus; 105 - display. DETAILED DESCRIPTION

[0023] In order to make the purpose, advantages and features of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the drawings are all in a very simplified form and are not drawn to scale, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention. In addition, the structure shown in the drawings is often a part of the actual structure. In particular, the emphasis of each drawing is different, and sometimes different scales are used.

[0024] It should be understood that when an element or layer is referred to as "on...", "connected to" other elements or layers, it can be directly on other elements or layers, connected to other elements or layers, or can include intermediate elements or layers. On the contrary, when an element is referred to as "directly on...", "directly connected to" other elements or layers, intermediate elements or layers are not included. Although the terms first, second, third, etc. can be used to describe various elements, components, regions, layers and / or parts, these elements, components, regions, layers and / or parts should not be limited by these terms. These terms are only used to distinguish an element, component, region, layer or part from another element, component, region, layer or part. Therefore, without departing from the teachings of the present invention, the first element, component, region, layer or part discussed below can be expressed as a second element, component, region, layer or part. Spatial relationship terms such as "under...", "below", "below", "above...", "above", "above", etc., can be used here for the convenience of description to describe the relationship between an element or feature shown in the figure and other elements or features. It should be understood that, in addition to the orientation shown in the figure, the spatial relationship terms are intended to include different orientations of the devices in use and operation. For example, if the device in the accompanying drawings is turned over, then, the elements or features described as "under...", "below", "below" will be oriented to be "on" other elements or features. The device can be oriented in addition (rotated 90 degrees or other orientations) and the spatial descriptors used herein are interpreted accordingly. The purpose of the terms used herein is only to describe specific embodiments and is not intended to be a limitation of the present invention. When used herein, the singular forms of "one", "one" and "said / the" are also intended to include plural forms, unless the context clearly indicates another way. It should also be understood that the term "include" is used to determine the inclusion of features, steps, operations, elements and / or parts, but does not exclude the inclusion or addition of one or more other features, steps, operations, elements, parts and / or groups. When used herein, the term "and / or" includes any and all combinations of the relevant listed items.

[0025] The purpose of the present invention is to provide an intelligent audit method, an audit system, an electronic device and a storage medium, which can reduce the audit workload and improve work efficiency by deploying an intelligent audit model.

[0026] Please refer to Figure 1 , Figure 1 The following is a flow chart of an intelligent audit method provided by an embodiment of the present invention. Figure 1 To achieve the above purpose, the present invention provides an intelligent audit method, comprising the following steps:

[0027] Training phase:

[0028] Training a deep learning model according to the received training image and the detection instruction corresponding to the training image, so that the deep learning model can recognize the image features corresponding to the detection instruction;

[0029] Review stage:

[0030] Obtaining the images to be reviewed;

[0031] Calling the deep learning model to review whether the reviewed image has the image features;

[0032] If yes, the audit result is qualified;

[0033] If not, the audit result is unqualified.

[0034] With such a configuration, the present invention can utilize the training images and detection instructions provided by the user to deploy deep learning models in a targeted manner to achieve intelligent auditing. Users can deploy intelligent auditing models without coding, which greatly reduces the application threshold of artificial intelligence technology and the audit workload, and improves work efficiency.

[0035] It should be understood that the types of training images include pictures and videos, and there can be multiple deep learning models, each of which corresponds to an operation step. Further, multiple different operation steps can be sorted according to the received operation instructions. Specifically, the present invention can provide a visual interactive interface, and the user formulates the audit steps according to the specific business process and the content to be audited, and constructs the audit process by dragging arrows and rectangular boxes. Taking the water tank cleaning business as an example, the user sets the audit steps for cleaning the water tank in the interactive interface as: arrival (community gate), disinfectant, pre-shift education, safety briefing, house number, water outage notice, spraying disinfectant, turbidity, residual chlorine, and water tank lock. 10 steps, the user can enter the names of the above 10 steps in the rectangular box respectively, and control the business flow through arrow lines to complete the audit process setting. It is also possible to train the same deep learning model so that it can audit multiple operation steps at the same time.

[0036] On this basis, create a classification directory according to the review process created by the user, and upload pictures and videos to the corresponding classification directory in the visual interactive interface. Taking the water tank cleaning business as an example, create directories for the 10 categories mentioned in step 1, such as arrival, disinfectant, and pre-shift education, and upload pictures or videos to the corresponding directories. To achieve the best results, more than 50 photos can be uploaded for each category. In this way, the present invention can automatically build training tasks according to the review process and configuration parameters created by the user, and automatically save the best results after training. And realize automated deployment, users can operate visually throughout the process, and do not need to have professional AI knowledge.

[0037] During the audit stage, taking the water tank cleaning business as an example, when operating the "pre-shift education" step, the audit program is notified of the features to focus on by adding four labels: safety helmet, safety belt, boots, and name tag. The audit program will automatically detect the above four features for the "pre-shift education" step. If all five features are detected, the audit result is qualified, otherwise the audit result is judged to be unqualified.

[0038] In order to meet the constraints of the audit, during the audit stage, in two adjacent operation steps, if the audit result of the deep learning model corresponding to the previous operation step is qualified, then entering the next operation step is allowed, otherwise entering the next operation step is prohibited. It should be noted that if it is an application scenario of post-audit, it is not necessary to prohibit entering the next operation step, but the audit of all steps can be completed, and unqualified steps can be marked in the process.

[0039] Optionally, each of the detection instructions corresponds to multiple image features. In the review stage, calling the deep learning model to review whether the reviewed image has the image features specifically includes:

[0040] According to the received prompt word, select part of the image features from the plurality of image features corresponding to the detection instruction;

[0041] The deep learning model is called to review whether the reviewed image has the image features of the part.

[0042] Customers can fine-tune the model effect through natural language question-answering. Taking the water tank cleaning business as an example, in the step of spraying disinfectant, you can enter prompt words to describe the whole process, where a person wearing a blue helmet, white boots, and gloves is using a water gun to spray water for cleaning. The model is fine-tuned based on the details of the above description. The system will pay special attention to the color of the helmet and boots and the action of the person holding the water gun to spray water, thereby improving the classification accuracy.

[0043] Furthermore, the detection instruction includes an OCR recognition instruction, the image feature includes a text feature, and the calling of the deep learning model to review whether the audited image has the image feature specifically includes: calling the deep learning model to review whether the audited image has the text feature based on OCR recognition. In an exemplary embodiment, the present invention supports OCR recognition of text in the classification step with an accuracy rate of over 99%. Taking the water tank cleaning business as an example, OCR recognition must be performed in the three steps of "safety briefing", "house number", and "water outage notice". Based on the characteristics of the business, this system updates the dictionary of the traditional OCR model and conducts targeted training to improve the standardization and accuracy of single-step review.

[0044] Preferably, after the audit result is unqualified, the deep learning model is retrained according to the received appeal instruction. The present invention provides an appeal channel. Users who believe that the model has identified and classified the wrong pictures can file an appeal in the appeal channel. The system will increase the weight of the wrongly identified pictures, re-add them to the training set for training, and establish a positive feedback mechanism to make the system more accurate with use.

[0045] The present invention also provides an audit system, which is applied to any of the above-mentioned intelligent audit methods, comprising:

[0046] A receiving module, used for receiving the training image and / or a detection instruction corresponding to the training image;

[0047] A training module, used for training the deep learning model;

[0048] An acquisition module, used for acquiring the image to be reviewed;

[0049] An execution module is used to call the deep learning model to review whether the audited image has the image features.

[0050] Since the retrieval system and the intelligent audit method belong to the same inventive concept, the retrieval system can utilize the training images and detection instructions provided by the user to deploy deep learning models in a targeted manner to achieve intelligent audit. Users can deploy intelligent audit models without code, which greatly reduces the application threshold of artificial intelligence technology and the audit workload, and improves work efficiency.

[0051] Please refer to Figure 2 , Figure 2 FIG. 1 is a block diagram of an electronic device provided by an embodiment of the present invention. Figure 2 As shown, the present invention also provides an electronic device, the electronic device comprising:

[0052] A memory 103 storing a computer program;

[0053] A processor 101 is connected to the memory for executing any one of the above-mentioned intelligent audit methods when calling the computer program;

[0054] The display 105 is communicatively connected with the processor and the memory, and is used to display a GUI interaction interface related to the intelligent audit method.

[0055] Since the electronic device and the intelligent audit method belong to the same inventive concept, the electronic device can utilize the training images and detection instructions provided by the user to deploy deep learning models in a targeted manner to achieve intelligent audit. The user can deploy the intelligent audit model without code, which greatly reduces the application threshold of artificial intelligence technology and the audit workload, and improves work efficiency.

[0056] like Figure 2 As shown, the electronic device also includes a communication interface 102 and a communication bus 104, wherein the processor 101, the communication interface 102, and the memory 103 communicate with each other through the communication bus 104. The communication bus 104 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus 104 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The communication interface 102 is used for communication between the above-mentioned electronic device and other devices.

[0057] The processor 101 referred to in the present invention may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor 101 is the control center of the electronic device, and various interfaces and lines are used to connect various parts of the entire electronic device.

[0058] The memory 103 may be used to store the computer program. The processor 101 implements various functions of the electronic device by running or executing the computer program stored in the memory 103 and calling the data stored in the memory 103 .

[0059] The memory 103 may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0060] The present invention also provides a storage medium storing a computer program, which implements any of the above-mentioned intelligent audit methods when executed by a processor.

[0061] Since the storage medium and the intelligent audit method belong to the same inventive concept, the storage medium can utilize the training images and detection instructions provided by the user to deploy deep learning models in a targeted manner to achieve intelligent auditing. Users can deploy intelligent auditing models without code, which greatly reduces the application threshold of artificial intelligence technology and the audit workload, and improves work efficiency.

[0062] Since the storage medium provided by the present invention and the intelligent audit method described above belong to the same inventive concept, the storage medium provided by the present invention has all the advantages of the intelligent audit method described above, so the beneficial effects of the storage medium provided by the present invention will not be described one by one here.

[0063] The storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer hard disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this article, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, a device or a device or used in combination with it.

[0064] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program codes. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0065] It should also be noted that, although the present invention has been disclosed as a preferred embodiment, the above embodiment is not intended to limit the present invention. For any technician familiar with the art, without departing from the scope of the technical solution of the present invention, the technical content disclosed above can be used to make many possible changes and modifications to the technical solution of the present invention, or modified into equivalent embodiments of equivalent changes. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the scope of protection of the technical solution of the present invention.

[0066] It should also be understood that, unless otherwise specified or indicated, the terms "first", "second", "third", etc. in the specification are merely used to distinguish between the various components, elements, steps, etc. in the specification, and are not used to indicate the logical relationship or sequential relationship between the various components, elements, steps, etc.

[0067] It should also be recognized that the terms described herein are used only to describe specific embodiments and are not intended to limit the scope of the invention. It should be noted that the singular forms "a" and "an" used herein and in the appended claims include plural references unless the context clearly indicates otherwise. For example, a reference to "a step" or "a device" means a reference to one or more steps or devices, and may include secondary steps and secondary devices. All conjunctions used should be understood in the broadest sense. And, the word "or" should be understood to have the definition of a logical "or", rather than a logical "exclusive or", unless the context clearly indicates otherwise. In addition, the implementation of embodiments of the present invention may include performing selected tasks manually, automatically, or in combination.

Claims

1. An intelligent audit method, characterized in that: The following steps are involved: Training phase: Provide a visual interactive interface so that users can formulate audit steps according to specific business processes and audit content, and build audit processes by dragging arrows and rectangles; According to the received training images and the detection instructions corresponding to the training images, the deep learning model is trained so that the deep learning model can recognize the image features corresponding to the detection instructions, wherein the types of the training images include pictures and videos, and according to the review process created by the user, a classification directory is created, and a function of uploading pictures and videos to the corresponding classification directory in the visual interactive interface is provided; Review stage: Obtaining the images to be reviewed; Calling the deep learning model to review whether the reviewed image has the image features; If yes, the audit result is qualified; If not, the audit result is unqualified; There are multiple image features corresponding to each detection instruction. In the review stage, calling the deep learning model to review whether the reviewed image has the image features specifically includes: According to the received prompt word, select part of the image features from the plurality of image features corresponding to the detection instruction; The deep learning model is called to review whether the reviewed image has the image features of the part.

2. The intelligent audit method according to claim 1, characterized in that: There are multiple deep learning models, and each of the deep learning models corresponds to an operation step.

3. The intelligent audit method according to claim 2, characterized in that: According to the received operation instruction, a plurality of different operation steps are sorted.

4. The intelligent audit method according to claim 3, characterized in that: During the review stage, in two adjacent operation steps, if the review result of the deep learning model corresponding to the previous operation step is qualified, entering the next operation step is allowed, otherwise entering the next operation step is prohibited.

5. The intelligent audit method according to claim 1, characterized in that: The detection instruction includes an OCR recognition instruction, the image feature includes a text feature, and the calling of the deep learning model to review whether the reviewed image has the image feature specifically includes: The deep learning model is called to review whether the image being reviewed has the text features based on OCR recognition.

6. The intelligent audit method according to claim 1, characterized in that: After the audit result is unqualified, the deep learning model is retrained according to the received appeal instruction.

7. An audit system, applied to the intelligent audit method as claimed in any one of claims 1 to 6, characterized in that: include: A receiving module, used for receiving the training image and / or a detection instruction corresponding to the training image; A training module, used for training the deep learning model; An acquisition module, used for acquiring the image to be reviewed; An execution module is used to call the deep learning model to review whether the audited image has the image features.

8. An electronic device, characterized in that: include: a memory storing a computer program; A processor, communicatively connected to the memory, and configured to execute the intelligent audit method according to any one of claims 1 to 6 when calling the computer program; A display is communicatively connected to the processor and the memory, and is used to display a GUI interaction interface related to the intelligent audit method.

9. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the intelligent audit method as described in any one of claims 1 to 6 is implemented.

Images