Industrial production digital image man-machine interaction method and system

Abstract

The invention is suitable for the field of digital images, and provides an industrial production digital image man-machine interaction method and system, and the method comprises the following steps: building personal digital images of employees based on a factory management system; when the two employee terminals establish communication connection, calling the corresponding personal digital image and sending the personal digital image to the two employee terminals; a real-time voice signal of an employee is obtained through an employee terminal and is preprocessed, so that the accuracy of voice content is improved; and forwarding and playing the real-time voice signal based on the personal digital image of the employee terminal of the opposite side. Firstly, the personal digital image is established according to a real person, and the played content is also transmitted in real time, so that good use experience can be achieved both in daily communication between employees and in communication guidance of a technical level.

Description

Technical Field

[0001] The present invention relates to the field of digital images, and in particular to a method and system for human-computer interaction of industrially produced digital images. Background Art

[0002] In industrial production, human-machine interaction is a very important component, which involves the information exchange and control between workers and mechanical equipment. It plays a key role in improving efficiency, safety and flexibility.

[0003] Traditional human-computer interaction is carried out in the form of text or voice on terminal devices, and the introduction of digital image technology can greatly improve the user experience. Compared with traditional text interfaces, digital images are more friendly and can enhance communication effects through expressions, movements and sounds, making interactions more humane.

[0004] In the prior art, digital images in industrial production are generally used to assist in controlling mechanical equipment, fault detection, operation guidance, etc. However, in actual situations, digital images are mostly standard models. No matter how the pronunciation and language are improved, the real experience is insufficient during interaction. Therefore, a human-computer interaction method and system for digital images in industrial production are extracted to solve the above problems. Summary of the invention

[0005] In view of the deficiencies in the prior art, the purpose of the present invention is to provide an industrial production digital image human-computer interaction method and system to solve the problems existing in the above-mentioned background technology.

[0006] The present invention is implemented as follows: a method for industrial production of digital image human-computer interaction, the method comprising the following steps:

[0007] Establish personal digital images of employees based on the factory management system, with a one-to-one correspondence between personal digital images and individual employees;

[0008] When two employee terminals establish a communication connection, the corresponding personal digital image is retrieved and sent to the two employee terminals. The retrieved personal digital image is in a cross-corresponding state with the two employee terminals. The employee terminals are human-computer interaction devices used by employees when performing production;

[0009] Acquire employees' real-time voice signals through employee terminals and pre-process them to improve the accuracy of voice content;

[0010] The real-time voice signal is relayed and played based on the personal digital image of the other party's employee terminal, so that the communication between employees can be played in real time with the personal digital image.

[0011] As a further solution of the present invention: the step of establishing the personal digital image of the employee based on the factory management system specifically includes:

[0012] Retrieve the employee's file information based on the factory management system and select the employee's photo from the file information;

[0013] Based on the person's photo, a face detection algorithm is used to detect and crop the face area image, and at the same time, the facial feature points are aligned, and the feature points refer to the characteristic organs of the face;

[0014] Generate a 3D facial reconstruction model of the employee based on the face area image using a deep learning model, wherein the deep learning model is used to recognize the geometric structure, texture and expression of the face;

[0015] A human body model is constructed based on the factory's surveillance images and integrated with a 3D facial reconstruction model to obtain a personal digital image.

[0016] As a further solution of the present invention: the step of constructing a human body model based on the factory monitoring image specifically includes:

[0017] Use the factory's monitoring equipment to identify specific employees and obtain relevant monitoring images;

[0018] Extracting a plurality of key frames from the surveillance image, wherein the key frames include images of the individual employee in different postures and angles;

[0019] The pose estimation model is used to traverse the key frames for pose estimation, so as to extract the key points of the human body in the key frames;

[0020] Generate a human body model based on the pose estimation results of each key frame;

[0021] A 3D clothing modeling tool is used to construct a clothing model that is closest to the monitoring image and apply it to the human body model.

[0022] As a further solution of the present invention: when two employee terminals establish a communication connection, the step of retrieving the corresponding personal digital image and sending it to the two employee terminals specifically includes:

[0023] Obtaining employee status information through an employee terminal, wherein the status information is the employee's current working status;

[0024] According to the status information, it is converted into instructions that match the digital image and applied to the personal digital image, and it is displayed in the function platform to determine whether the employee is in a state where communication is possible;

[0025] When an employee selects an idle personal digital image, a communication request is sent to the corresponding employee terminal through the platform;

[0026] When the request is approved, a communication connection between the two employee terminals is established, and the personal digital image is cross-sent to the two employee terminals.

[0027] As a further solution of the present invention: the step of obtaining the real-time voice signal of the employee through the employee terminal and preprocessing it specifically includes:

[0028] The acquired real-time speech signal is subjected to noise reduction processing, wherein the noise reduction processing includes frequency domain filtering, adaptive filtering and deep learning noise reduction;

[0029] When there is an echo in the voice collection environment, an echo cancellation algorithm is used to remove the echo of the real-time voice signal;

[0030] Adjust the overall volume of the real-time speech signal and perform dereverberation processing.

[0031] Another object of the present invention is to provide an industrial production digital image human-computer interaction system, the system comprising:

[0032] The digital image building module builds the personal digital image of employees based on the factory management system, and the personal digital image corresponds to the individual employee on a one-to-one basis;

[0033] A communication determination module, used for retrieving the corresponding personal digital image and sending it to the two employee terminals when two employee terminals establish a communication connection, wherein the retrieved personal digital image is in a cross-corresponding state with the two employee terminals, and the employee terminals are human-computer interaction devices used by employees when performing production;

[0034] The voice processing module is used to obtain the real-time voice signals of employees through the employee terminals and pre-process them to improve the accuracy of the voice content;

[0035] The real-time transmission module relays and plays the real-time voice signal based on the personal digital image of the other party's employee terminal, so that the communication between employees can be played in real time with the personal digital image.

[0036] As a further solution of the present invention: the digital image creation module includes:

[0037] The information retrieval unit retrieves the employee's file information based on the factory management system and selects the employee's personal photos from the file information;

[0038] A face detection unit detects and crops a face area image based on a person's photo using a face detection algorithm, and aligns facial feature points, which refer to characteristic organs of a person's face;

[0039] A model building unit, used to generate a 3D facial reconstruction model of the employee based on the face area image using a deep learning model, wherein the deep learning model is used to recognize the geometric structure, texture and expression of the face;

[0040] The overall model building unit constructs a human body model based on the factory's surveillance images and integrates the 3D facial reconstruction model to obtain a personal digital image.

[0041] As a further solution of the present invention: the overall model building unit includes:

[0042] The image acquisition subunit is used to identify specific individual employees through the factory's monitoring equipment and obtain relevant monitoring images;

[0043] A key frame extraction subunit is used to extract multiple key frames from the monitoring image, wherein the key frames include images of the individual employee in different postures and angles;

[0044] A pose estimation subunit, used for traversing key frames using a pose estimation model to perform pose estimation, and for extracting key points of a human body in the key frames;

[0045] A human body model generation subunit, used to generate a human body model according to the pose estimation result of each key frame;

[0046] The additional construction subunit is used to use a 3D clothing modeling tool to construct a clothing model that is closest to the monitoring image and apply it to the human body model.

[0047] As a further solution of the present invention: the communication determination module includes:

[0048] A status acquisition unit, used to acquire employee status information through an employee terminal, wherein the status information is the employee's current working status;

[0049] The status synchronization unit is used to convert the status information into instructions that conform to the digital image and apply them to the personal digital image, and display them in the function platform to determine whether the employee is in a state where communication is possible;

[0050] The communication request subunit is used to send a communication request to the corresponding employee terminal through the platform when an employee selects a personal digital image in an idle state;

[0051] The communication establishment subunit is used to establish a communication connection between two employee terminals when the request is approved, and to cross-send the personal digital image to the two employee terminals.

[0052] As a further solution of the present invention: the speech processing module includes:

[0053] A noise reduction processing unit, used to perform noise reduction processing on the acquired real-time voice signal, wherein the noise reduction processing includes frequency domain filtering, adaptive filtering and deep learning noise reduction;

[0054] An echo processing unit, used to remove the echo of the real-time voice signal by using an echo cancellation algorithm when there is an echo in the voice collection environment;

[0055] The volume adjustment unit is used to adjust the overall volume of the real-time voice signal and perform dereverberation processing.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] The present invention will utilize the factory management system commonly used in industrial production to establish personal digital images that conform to the appearance of employees. These personal digital images will be directly used for communication between employees. By utilizing the interactive characteristics of existing digital image technology, when two employees establish a communication connection, the other party's personal digital image will be displayed on the employee terminal. When the two are communicating, the personal digital image can play the voice of both parties in real time in a relay manner. First of all, the personal digital image is established based on the real person, and the played content is also transmitted in real time, so that whether it is daily communication between employees or communication guidance at the technical level, a good user experience can be obtained. Relying on the characteristics of digital image broadcasting, it will not affect some operations of employees at work, and the actual use effect is very significant. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 The present invention is a flowchart of a human-computer interaction method for industrial production of digital images.

[0059] Figure 2 The present invention is a flowchart of establishing an employee's personal digital image based on a factory management system in an industrial production digital image human-computer interaction method.

[0060] Figure 3 The present invention is a flowchart of constructing a human body model based on factory monitoring images in a human-computer interaction method for industrial production digital images.

[0061] Figure 4 The present invention is a flow chart of retrieving the corresponding personal digital image and sending it to two employee terminals in an industrial production digital image human-computer interaction method.

[0062] Figure 5 The present invention is a flowchart of obtaining and preprocessing the real-time voice signals of employees through employee terminals in a human-computer interaction method for industrial production digital images.

[0063] Figure 6 It is a structural diagram of an industrial production digital image human-computer interaction system.

[0064] Figure 7 This is a structural diagram of a digital image building module in an industrial production digital image human-computer interaction system.

[0065] Figure 8 A structural schematic diagram of a unit for establishing an overall model in an industrial production digital image human-computer interaction system.

[0066] Fig. 9 The schematic diagram is a structural diagram of a communication determination module in an industrial production digital image human-computer interaction system.

[0067] Fig.10 The schematic diagram is a structural diagram of a speech processing module in an industrial production digital image human-computer interaction system. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solution and advantages 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 understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0069] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0070] like Figure 1 As shown, an embodiment of the present invention provides an industrial production digital image human-computer interaction method, characterized in that the method comprises the following steps:

[0071] S100, establishes the personal digital image of employees based on the factory management system, and the personal digital image corresponds to the individual employee one by one;

[0072] S200, when two employee terminals establish a communication connection, retrieve the corresponding personal digital image and send it to the two employee terminals, the retrieved personal digital image and the two employee terminals are in a cross-corresponding state, and the employee terminals are human-computer interaction devices used by employees when performing production;

[0073] S300, obtaining the real-time voice signal of the employee through the employee terminal and preprocessing it to improve the accuracy of the voice content;

[0074] S400, the real-time voice signal is relayed and played based on the personal digital image of the other party's employee terminal, so that the communication between employees can be played in real time with the personal digital image.

[0075] It should be noted that general digital images are generated according to standards. When conducting human-computer interaction, it is difficult for people to perceive the characteristics of familiar characters from the digital images, and it is difficult to obtain a real sense of use. The digital image of the present invention is established by collecting employee information based on the factory management system, so that the generated personal digital image can meet some characteristics of the employees, making it a replaceable communication object, thereby ensuring a real experience of interaction. The factory management system is a comprehensive information system for managing and optimizing factory production processes, resource allocation and operation management. It integrates various hardware and software technologies, such as a large number of monitoring equipment, personnel management systems, etc.

[0076] The present invention will utilize the factory management system commonly used in industrial production to establish personal digital images that conform to the appearance of employees. These personal digital images will be directly used for communication between employees. By utilizing the interactive characteristics of existing digital image technology, when two employees establish a communication connection, the other party's personal digital image will be displayed on the employee terminal. When the two are communicating, the personal digital image can play the voice of both parties in real time in a relay manner. First of all, the personal digital image is established based on the real person, and the played content is also transmitted in real time, so that whether it is daily communication between employees or communication guidance at the technical level, a good user experience can be obtained. Relying on the characteristics of digital image broadcasting, it will not affect some operations of employees at work, and the actual use effect is very significant.

[0077] like Figure 2 As shown in FIG. 1 , as a preferred embodiment of the present invention, the step of establishing the personal digital image of an employee based on the factory management system specifically includes:

[0078] S101, retrieve the employee's file information based on the factory management system, and select the employee's photo from the file information;

[0079] S102, detecting and cropping a face region image based on the person photo using a face detection algorithm, and aligning facial feature points, wherein the feature points refer to characteristic organs of the person's face;

[0080] S103, generating a 3D facial reconstruction model of the employee based on the face region image using a deep learning model, wherein the deep learning model is used to recognize the geometric structure, texture and expression of the face;

[0081] S104, constructing a human body model based on the factory's surveillance images and integrating the 3D facial reconstruction model to obtain a personal digital image.

[0082] In the embodiment of the present invention, generally speaking, there are files of management personnel in the factory, and the files also contain photos of employees. Such photos are the most authentic personal information. Relevant facial features can be extracted by analyzing the photos. When generating a 3D facial reconstruction model of the employee, tools such as DECA and Pix2Face can be used. They can generate a 3D facial model based on a facial picture, and also include facial expressions and some details. However, it should be noted that the photos of the people in the file information are generally half-length photos, so only an accurate facial model can be obtained, but a model of the human torso cannot be obtained. Although the existing technology can infer the posture and body structure of the person through posture estimation, it is not accurate enough. At this time, the monitoring images of the employees as a whole collected by the monitoring equipment in the factory are used. The monitoring images can be obtained through cameras installed in the factory, and of course, they can also be some access control devices for identity verification.

[0083] like Figure 3 As shown, as a preferred embodiment of the present invention, the step of constructing a human body model based on the monitoring image of the factory specifically includes:

[0084] S114, identifying specific individual employees through the factory's monitoring equipment and obtaining relevant monitoring images;

[0085] S124, extracting a plurality of key frames from the surveillance image, wherein the key frames include images of the individual employee in different postures and angles;

[0086] S134, using a pose estimation model to traverse the key frames to perform pose estimation, so as to extract key points of the human body in the key frames;

[0087] S144, generating a human body model according to the pose estimation result of each key frame;

[0088] S154, using a 3D clothing modeling tool to construct a clothing model that is closest to the clothing model in the monitoring image and applying it to the human body model.

[0089] In the embodiment of the present invention, key frames are extracted from the surveillance images, and these frames should contain different postures and angles of the employees for subsequent 3D reconstruction. A video analysis tool can be used to automatically select representative frames, or manually select them. If the video quality is poor (such as blur, low resolution, noise, etc.), image enhancement techniques (such as super-resolution, denoising, sharpening, etc.) can be used to improve the image quality, and image segmentation techniques (such as Rembg, DeepLabV3, etc.) can be used to remove the background and extract the outline of the person to ensure that the subsequent processing is only for the person itself. Then, the key points of the human body (such as shoulders, elbows, wrists, knees, etc.) are extracted from each frame. In order to ensure that the posture of the person between different frames is consistent, a posture tracking algorithm (such as SORT, DeepSort) or a tracking technology based on the optical flow method (Optical Flow) can be used to ensure the continuity of the posture of the same person in different frames. It should be noted that if the surveillance image contains videos from different perspectives, multi-view stereoscopic vision (MVS, Multi-View Stereo) technology can be used to combine information from different perspectives to reconstruct a more complete human model. In addition, the clothing model can also be replaced with reference to the actual situation to make the personal digital image more appropriate.

[0090] like Figure 4 As shown, as a preferred embodiment of the present invention, when two employee terminals establish a communication connection, the step of retrieving the corresponding personal digital image and sending it to the two employee terminals specifically includes:

[0091] S201, obtaining employee status information through an employee terminal, where the status information is the employee's current working status;

[0092] S202, converting the status information into instructions that match the digital image and applying them to the personal digital image, and displaying them in the function platform to determine whether the employee is in a state where communication is possible;

[0093] S203, when an employee selects a personal digital image in an idle state, a communication request is sent to a corresponding employee terminal through the platform;

[0094] S204, when the request is approved, a communication connection between the two employee terminals is established, and the personal digital image is cross-sent to the two employee terminals.

[0095] In the embodiment of the present invention, the personal digital image can not only be used to participate in the process of human-computer interaction, but the status of the personal digital image displayed on the employee terminal can also be linked to the status of the employee. For example, if an employee is in a busy working state, other personnel can see the personal digital image of the employee on the platform as an animation of working in progress. This can intuitively display the status of the employee. When two employees establish a communication connection, the other party's personal digital image will be displayed on their respective employee terminals, and the personal digital image can be used as a carrier for interaction.

[0096] like Figure 5 As shown, as a preferred embodiment of the present invention, the step of obtaining the real-time voice signal of the employee through the employee terminal and preprocessing it specifically includes:

[0097] S301, performing noise reduction processing on the acquired real-time voice signal, wherein the noise reduction processing includes frequency domain filtering, adaptive filtering and deep learning noise reduction;

[0098] S302, when there is an echo in the voice collection environment, using an echo cancellation algorithm to remove the echo of the real-time voice signal;

[0099] S303, adjusting the overall volume of the real-time voice signal and performing dereverberation processing.

[0100] In the embodiment of the present invention, background noise is first removed, such as fan sound, keyboard tapping sound, machine running sound, etc., and the voice signal is converted from the time domain to the frequency domain through Fourier transform, and then high-frequency noise or low-frequency hum is removed in the frequency domain, and then the filter parameters are dynamically adjusted according to the change of environmental noise to remove noise in real time, and finally a model based on deep learning (such as DNN, CNN, RNN, etc.) is used to separate clean speech from the voice signal. These models usually require a large amount of noisy speech and clean speech data for training, and echo cancellation and volume adjustment are conventional technical means, so they will not be described in detail.

[0101] like Figure 6 As shown, the embodiment of the present invention further provides an industrial production digital image human-computer interaction system, characterized in that the system includes:

[0102] A digital image building module 100 builds a personal digital image of an employee based on the factory management system, and the personal digital image corresponds to the individual employee on a one-to-one basis;

[0103] The communication determination module 200 is used to retrieve the corresponding personal digital image and send it to the two employee terminals when the two employee terminals establish a communication connection. The retrieved personal digital image is in a cross-corresponding state with the two employee terminals. The employee terminals are human-computer interaction devices used by employees when performing production;

[0104] The voice processing module 300 is used to obtain the real-time voice signal of the employee through the employee terminal and pre-process it to improve the accuracy of the voice content;

[0105] The real-time transmission module 400 relays and plays the real-time voice signal based on the personal digital image of the other party's employee terminal, so that the communication between employees can be played in real time with the personal digital image.

[0106] In the embodiment of the present invention, the present invention will utilize the factory management system commonly used in industrial production to establish personal digital images that conform to the appearance of employees. These personal digital images will be directly used for communication between employees. Utilizing the interactive characteristics of existing digital image technology, when two employees establish a communication connection, the other party's personal digital image will be displayed on the employee terminal. When the two are communicating, the personal digital image can play the voice of both parties in real time in a relay manner. First of all, the personal digital image is established according to the real person, and the played content is also transmitted in real time, so that whether it is daily communication between employees or communication guidance at the technical level, a good user experience can be obtained, and relying on the characteristics of digital image broadcasting, it will not affect some operations of employees at work.

[0107] like Figure 7 As shown, as a preferred embodiment of the present invention, the digital image building module 100 includes:

[0108] The information retrieval unit 101 retrieves the employee's file information based on the factory management system, and selects the employee's photo from the file information;

[0109] A face detection unit 102 detects and crops a face region image based on a person's photo using a face detection algorithm, and aligns facial feature points, where the feature points refer to characteristic organs of a person's face;

[0110] A model building unit 103 is used to generate a 3D facial reconstruction model of the employee based on the face area image using a deep learning model, wherein the deep learning model is used to recognize the geometric structure, texture and expression of the face;

[0111] The overall model building unit 104 builds a human body model based on the factory's surveillance images and integrates the 3D facial reconstruction model to obtain a personal digital image.

[0112] like Figure 8 As shown, as a preferred embodiment of the present invention, the overall model building unit 104 includes:

[0113] The image acquisition subunit 114 is used to identify specific individual employees through the factory's monitoring equipment and acquire relevant monitoring images;

[0114] A key frame extraction subunit 124 is used to extract a plurality of key frames from the surveillance image, wherein the key frames include images of the individual employee in different postures and angles;

[0115] A posture estimation subunit 134, used for traversing key frames using a posture estimation model to perform posture estimation, and for extracting key points of a human body in the key frames;

[0116] A human body model generating subunit 144, configured to generate a human body model according to the pose estimation result of each key frame;

[0117] The additional construction subunit 154 is used to use a 3D clothing modeling tool to construct a clothing model that is closest to the monitoring image and apply it to the human body model.

[0118] like Fig. 9 As shown, as a preferred embodiment of the present invention, the communication determination module 200 includes:

[0119] A status acquisition unit 201 is used to acquire employee status information through an employee terminal, where the status information is the employee's current working status;

[0120] The status synchronization unit 202 is used to convert the status information into instructions that conform to the digital image and apply it to the personal digital image, and display it in the function platform to determine whether the employee is in a state that can communicate;

[0121] The communication request subunit 203 is used to send a communication request to the corresponding employee terminal through the platform when an employee selects a personal digital image in an idle state;

[0122] The communication establishing subunit 204 is used to establish a communication connection between two employee terminals when the request is approved, and to cross-send the personal digital image to the two employee terminals.

[0123] like Fig.10 As shown, as a preferred embodiment of the present invention, the speech processing module 300 includes:

[0124] A noise reduction processing unit 301 is used to perform noise reduction processing on the acquired real-time speech signal, wherein the noise reduction processing includes frequency domain filtering, adaptive filtering and deep learning noise reduction;

[0125] The echo processing unit 302 is used to remove the echo of the real-time voice signal by using an echo cancellation algorithm when there is an echo in the voice collection environment;

[0126] The volume adjustment unit 303 is used to adjust the overall volume of the real-time speech signal and perform dereverberation processing. The above only describes in detail the preferred embodiments of the present invention, and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0127] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. 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 performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0128] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can 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).

[0129] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the disclosure in the specification and examples. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A human-computer interaction method for industrial production digital images, characterized in that: The method comprises the following steps: Establish personal digital images of employees based on the factory management system, with a one-to-one correspondence between personal digital images and individual employees; When two employee terminals establish a communication connection, the corresponding personal digital image is retrieved and sent to the two employee terminals. The retrieved personal digital image is in a cross-corresponding state with the two employee terminals. The employee terminals are human-computer interaction devices used by employees when performing production; Acquire employees' real-time voice signals through employee terminals and pre-process them to improve the accuracy of voice content; The real-time voice signal is relayed and played based on the personal digital image of the other party's employee terminal, so that the communication between employees can be played in real time with the personal digital image.

2. The industrial production digital image human-computer interaction method according to claim 1, characterized in that: The step of establishing the employee's personal digital image based on the factory management system specifically includes: Retrieve the employee's file information based on the factory management system and select the employee's photo from the file information; Based on the person's photo, a face detection algorithm is used to detect and crop the face area image, and at the same time, the facial feature points are aligned, and the feature points refer to the characteristic organs of the face; Generate a 3D facial reconstruction model of the employee based on the face area image using a deep learning model, wherein the deep learning model is used to recognize the geometric structure, texture and expression of the face; A human body model is constructed based on the factory's surveillance images and integrated with a 3D facial reconstruction model to obtain a personal digital image.

3. The industrial production digital image human-computer interaction method according to claim 2, characterized in that: The step of constructing a human body model based on the factory monitoring images specifically includes: Use the factory's monitoring equipment to identify specific employees and obtain relevant monitoring images; Extracting a plurality of key frames from the surveillance image, wherein the key frames include images of the individual employee in different postures and angles; The pose estimation model is used to traverse the key frames for pose estimation, so as to extract the key points of the human body in the key frames; Generate a human body model based on the pose estimation results of each key frame; A 3D clothing modeling tool is used to construct a clothing model that is closest to the monitoring image and apply it to the human body model.

4. The industrial production digital image human-computer interaction method according to claim 1, characterized in that: When two employee terminals establish a communication connection, the step of retrieving the corresponding personal digital image and sending it to the two employee terminals specifically includes: Obtaining employee status information through an employee terminal, wherein the status information is the employee's current working status; According to the status information, it is converted into instructions that match the digital image and applied to the personal digital image, and it is displayed in the function platform to determine whether the employee is in a state where communication is possible; When an employee selects an idle personal digital image, a communication request is sent to the corresponding employee terminal through the platform; When the request is approved, a communication connection between the two employee terminals is established, and the personal digital image is cross-sent to the two employee terminals.

5. The industrial production digital image human-computer interaction method according to claim 1, characterized in that: The step of obtaining the real-time voice signal of the employee through the employee terminal and preprocessing it specifically includes: The acquired real-time speech signal is subjected to noise reduction processing, wherein the noise reduction processing includes frequency domain filtering, adaptive filtering and deep learning noise reduction; When there is an echo in the voice collection environment, an echo cancellation algorithm is used to remove the echo of the real-time voice signal; Adjust the overall volume of the real-time speech signal and perform dereverberation processing.

6. An industrial production digital image human-computer interaction system, characterized in that: The system comprises: The digital image building module builds the personal digital image of employees based on the factory management system, and the personal digital image corresponds to the individual employee on a one-to-one basis; A communication determination module, used for retrieving the corresponding personal digital image and sending it to the two employee terminals when two employee terminals establish a communication connection, wherein the retrieved personal digital image is in a cross-corresponding state with the two employee terminals, and the employee terminals are human-computer interaction devices used by employees when performing production; The voice processing module is used to obtain the real-time voice signals of employees through the employee terminals and pre-process them to improve the accuracy of the voice content; The real-time transmission module relays and plays the real-time voice signal based on the personal digital image of the other party's employee terminal, so that the communication between employees can be played in real time with the personal digital image.

7. The industrial production digital image human-computer interaction system according to claim 6, characterized in that: The digital image building module includes: The information retrieval unit retrieves the employee's file information based on the factory management system and selects the employee's personal photos from the file information; A face detection unit detects and crops a face area image based on a person's photo using a face detection algorithm, and aligns facial feature points, which refer to characteristic organs of a person's face; A model building unit, used to generate a 3D facial reconstruction model of the employee based on the face area image using a deep learning model, wherein the deep learning model is used to recognize the geometric structure, texture and expression of the face; The overall model building unit constructs a human body model based on the factory's surveillance images and integrates the 3D facial reconstruction model to obtain a personal digital image.

8. The industrial production digital image human-computer interaction system according to claim 7, characterized in that: The overall model building unit comprises: The image acquisition subunit is used to identify specific individual employees through the factory's monitoring equipment and obtain relevant monitoring images; A key frame extraction subunit is used to extract multiple key frames from the monitoring image, wherein the key frames include images of the individual employee in different postures and angles; A pose estimation subunit, used for traversing key frames using a pose estimation model to perform pose estimation, and for extracting key points of a human body in the key frames; A human body model generation subunit, used to generate a human body model according to the pose estimation result of each key frame; The additional construction subunit is used to use a 3D clothing modeling tool to construct a clothing model that is closest to the monitoring image and apply it to the human body model.

9. The industrial production digital image human-computer interaction system according to claim 6, characterized in that: The communication determination module comprises: A status acquisition unit, used to acquire employee status information through an employee terminal, wherein the status information is the employee's current working status; The status synchronization unit is used to convert the status information into instructions that conform to the digital image and apply them to the personal digital image, and display them in the function platform to determine whether the employee is in a state where communication is possible; The communication request subunit is used to send a communication request to the corresponding employee terminal through the platform when an employee selects a personal digital image in an idle state; The communication establishment subunit is used to establish a communication connection between two employee terminals when the request is approved, and to cross-send the personal digital image to the two employee terminals.

10. The industrial production digital image human-computer interaction system according to claim 6, characterized in that: The speech processing module comprises: A noise reduction processing unit, used to perform noise reduction processing on the acquired real-time voice signal, wherein the noise reduction processing includes frequency domain filtering, adaptive filtering and deep learning noise reduction; An echo processing unit, used to remove the echo of the real-time voice signal by using an echo cancellation algorithm when there is an echo in the voice collection environment; The volume adjustment unit is used to adjust the overall volume of the real-time voice signal and perform dereverberation processing.

Images