Visualization method and system based on AI image discrimination, terminal and storage medium

Abstract

The invention belongs to the technical field of AI image recognition, and particularly relates to a visualization method and system based on AI image discrimination, a terminal and a storage medium, and the method comprises the steps: S1, obtaining the actual type of a target object, and carrying out the regional processing of the target object according to the actual type; s2, acquiring actual view information of each sub-region through AI imaging equipment; s3, performing optimization preprocessing on the actual view information to obtain preprocessed view information; s4, extracting and processing the preprocessed view information to obtain actual image features; and S5, comparing the preprocessed view information with the actual image features through an AI image recognition technology to obtain a comparison result. According to the method, a large amount of complex image data can be automatically identified and analyzed, the need of manual intervention is reduced, the data processing efficiency is remarkably improved, and the learning and adaptive capacity of the system to new conditions is also enhanced.

Description

Technical Field

[0001] The present invention belongs to the field of AI image recognition technology, and in particular relates to a visualization method, system, terminal and storage medium based on AI image recognition. Background Art

[0002] Artificial Intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. Artificial Intelligence is a branch of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robots, speech recognition, image recognition, natural language processing and expert systems. Since the birth of artificial intelligence, the theory and technology of artificial intelligence have become increasingly mature, and the application fields have continued to expand. It can be imagined that the scientific and technological products brought by artificial intelligence in the future will be the "container" of human wisdom. Artificial intelligence can simulate the information process of human consciousness and thinking.

[0003] However, some existing technologies are difficult to capture high-level features of complex target images in actual applications, and require a lot of manual operation, resulting in limited recognition accuracy and thus low operating efficiency. Summary of the invention

[0004] The purpose of the present invention is to provide a visualization method based on AI image recognition to address the deficiencies of the prior art and to solve any of the above-mentioned technical problems.

[0005] In order to achieve the above object, the present invention adopts the following technical solution:

[0006] A visualization method based on AI image recognition, comprising:

[0007] S1. Obtaining the actual type of the target object, and performing regional processing on the target object according to the actual type;

[0008] S2. Acquire the actual view information of each sub-area through AI imaging equipment;

[0009] S3, optimizing preprocessing of the actual view information to obtain preprocessed view information;

[0010] S4, extracting and processing the preprocessed view information to obtain actual image features;

[0011] S5. Compare the preprocessed view information and the actual image features by using AI image recognition technology to obtain a comparison result.

[0012] Preferably, the step of obtaining the actual type of the target object and performing regional processing on the target object according to the actual type includes:

[0013] Determine whether the actual species of the target object is a biological species or a non-biological species;

[0014] According to the biological species or non-biological species combined with the previously registered core data information, the area is divided into at least one important sub-region and a non-important sub-region.

[0015] Preferably, the step of acquiring actual view information of each sub-region by using an AI imaging device includes:

[0016] Driving the three-dimensional moving device to move so that the AI ​​imaging device can cyclically move in horizontal, vertical and longitudinal directions;

[0017] Using the three-dimensional mobile device to collect dynamic views in the three-dimensional space and static views in the two-dimensional space of each sub-region;

[0018] The actual view information is obtained by comparing and verifying the dynamic view and the static view.

[0019] Preferably, the step of optimizing preprocessing the actual view information to obtain preprocessed view information includes:

[0020] Inputting the collected actual view information into a preset physical model;

[0021] The physical model preprocesses the spectral image to obtain preprocessed view information; wherein the preprocessing includes correcting geometric deformation of the image, wavelength correction, and removing the influence of the atmosphere or water vapor.

[0022] Preferably, the step of extracting and processing the preprocessed view information to obtain actual image features comprises:

[0023] A convolutional neural network deep learning model is used to perform feature extraction on the preprocessed view information to obtain extracted image features, so as to obtain the actual image features.

[0024] In which, training is performed on the large-scale data set, and high-level features in the actual image features of the target object are learned through CNN.

[0025] Preferably, the convolutional neural network calculation formula is as follows:

[0026] Convolution operation: Assume that the input image is I, the convolution kernel (filter) is K, and the output feature map is O; the sizes of the input image and convolution kernel are W×H and F×F respectively, the stride is S, and the padding is P; then the convolution operation is calculated as follows:

[0027] O(i, j) = ∑[I*K(i, j)]

[0028] Among them, O(i, j) is a pixel value on the output feature map, I*K(i, j) represents the result of element-by-element multiplication and addition of the input image I and the convolution kernel K at position (i, j); this operation moves on the entire image with a step size of S until the entire image is traversed.

[0029] Preferably, the step of comparing the preprocessed view information and the actual image features by using AI image recognition technology to obtain a comparison result includes:

[0030] Comparing the preprocessed view information with the reference image in the database using AI image recognition technology to form a first result;

[0031] Compare the actual image features with reference features of the reference image in the database using AI image recognition technology to form a second result;

[0032] When the first result is that the two data are consistent and the second result is that the two data are consistent, feedback information that the state of the target object is normal is generated; and the AI ​​imaging device is turned off;

[0033] When the first result is that the data of the two are inconsistent and / or the second result is that the data of the two are inconsistent, feedback information indicating that the state of the target object is abnormal is generated, and the maintenance manager is notified at the same time.

[0034] The present invention also discloses a visualization system based on AI image recognition, which is used to execute the visualization method based on AI image recognition; wherein the visualization system based on AI image recognition comprises:

[0035] An acquisition module, the acquisition module is used to obtain the actual type of the target object, and perform regional processing on the target object according to the actual type; and acquire the actual view information of each sub-region through the AI ​​imaging device;

[0036] an optimization module, the optimization module being used to optimize preprocessing of the actual view information to obtain preprocessed view information;

[0037] A processing module, the processing module is used to extract and process the pre-processed view information to obtain actual image features;

[0038] A comparison module is used to compare the preprocessed view information and the actual image features through AI image recognition technology to obtain a comparison result.

[0039] The present invention also discloses a visualization terminal based on AI image recognition, comprising: a memory, a processor, and a visualization program based on AI image recognition stored in the memory and executable on the processor, wherein the visualization program based on AI image recognition implements the steps of the visualization method based on AI image recognition when executed by the processor.

[0040] The present invention also discloses a storage medium, on which a visualization program based on AI image recognition is stored. When the visualization program based on AI image recognition is executed by a processor, the steps of the visualization method based on AI image recognition are implemented.

[0041] The beneficial effect of the present invention is that the technical solution first obtains the actual type of the target object, and performs regional processing on the target object according to the actual type, so that corresponding important regional parts can be divided according to different types to ensure the pertinence and comprehensiveness of data acquisition; then the actual view information of each sub-region is acquired through the AI ​​imaging device, so as to obtain more comprehensive data, and when an error occurs in the acquisition of a certain area, the area can be re-acquired separately, which is beneficial to improving data acquisition and subsequent recognition efficiency; then the actual view information is optimized and pre-processed to obtain pre-processed view information, so that image quality and accuracy can be improved; then the pre-processed view information is extracted and processed to obtain actual image features, so that the actual image features can be quickly and accurately acquired, which is beneficial to the accuracy and efficiency of subsequent recognition; finally, the pre-processed view information and the actual image features are compared by AI image recognition technology to obtain a comparison result; thus, a large amount of complex image data can be automatically identified and analyzed, the need for manual intervention is reduced, the efficiency of data processing is significantly improved, and the system's ability to learn and adapt to new situations is enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The following will refer to the attached Figures 1 to 3 To describe the features, advantages and technical effects of exemplary embodiments of the present invention.

[0043] Figure 1 A flowchart of a visualization method based on AI image recognition according to an embodiment of the present invention;

[0044] Figure 2 A structural block diagram of a visualization system based on AI image recognition according to an embodiment of the present invention;

[0045] Figure 3This is a structural block diagram of a visualization terminal based on AI image recognition according to an embodiment of the present invention.

[0046] In the figure: 1001 - processor; 1002 - communication bus; 1003 - user interface; 1004 - network interface; 1005 - memory. DETAILED DESCRIPTION

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions.

[0048] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.

[0049] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0050] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and multiple situations exist alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0051] In the description of the embodiments of the present application, unless otherwise clearly specified and limited, technical terms such as "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0052] The present invention proposes a visualization method, system, terminal and storage medium based on AI image recognition.

[0053] like Figure 3 As shown, Figure 3 It is a schematic diagram of the terminal structure of the hardware operating environment involved in the embodiment of the present invention.

[0054] The terminal of the embodiment of the present invention can be a PC, or it can be a smart phone, a tablet computer, an e-book reader, an MP3 (Moving Picture Experts Group Audio Layer III) player, an MP4 (Moving Picture Experts Group Audio Layer IV) player, a portable computer, or other portable terminal devices with display function.

[0055] like Figure 3 As shown, the terminal may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0056] Optionally, the terminal may also include a camera, an RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi module, and the like. Among them, sensors include light sensors, motion sensors, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display screen according to the brightness of the ambient light, and the proximity sensor may turn off the display screen and / or backlight when the mobile terminal is moved to the ear. As a type of motion sensor, the gravity acceleration sensor can monitor the magnitude of acceleration in all directions (generally three axes), and can monitor the magnitude and direction of gravity when stationary. It can be used for applications that identify the posture of the mobile terminal (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; of course, the mobile terminal may also be equipped with other sensors such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, etc., which will not be repeated here.

[0057] Those skilled in the art will appreciate that the terminal structure shown in the figure does not constitute a limitation on the terminal, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0058] Reference Figure 1 One embodiment of the present invention provides a visualization method based on AI image recognition, a system, a terminal and a storage medium, such as Figure 1 As shown, the visualization method based on AI image recognition includes:

[0059] S1. Obtaining the actual type of the target object, and performing regional processing on the target object according to the actual type;

[0060] S2. Acquire the actual view information of each sub-area through AI imaging equipment;

[0061] S3, optimizing preprocessing of the actual view information to obtain preprocessed view information;

[0062] S4, extracting and processing the preprocessed view information to obtain actual image features;

[0063] S5. Compare the preprocessed view information and the actual image features by using AI image recognition technology to obtain a comparison result.

[0064] The technical solution of the present invention first obtains the actual type of the target object, and performs regional processing on the target object according to the actual type, so that corresponding important regional parts can be divided according to different types to ensure the pertinence and comprehensiveness of data acquisition; then the actual view information of each sub-region is acquired through the AI ​​imaging device, so as to obtain more comprehensive data, and when an error occurs in the acquisition of a certain area, the area can be re-acquired separately, which is beneficial to improving data acquisition and subsequent recognition efficiency; then the actual view information is optimized and pre-processed to obtain pre-processed view information, so that image quality and accuracy can be improved; then the pre-processed view information is extracted and processed to obtain actual image features, so that the actual image features can be quickly and accurately acquired, which is beneficial to the accuracy and efficiency of subsequent recognition; finally, the pre-processed view information and the actual image features are compared by AI image recognition technology to obtain a comparison result; thus, a large amount of complex image data can be automatically identified and analyzed, the need for manual intervention is reduced, the efficiency of data processing is significantly improved, and the system's ability to learn and adapt to new situations is enhanced.

[0065] Specifically, in some implementations, in S1, the step of obtaining the actual type of the target object and performing region-based processing on the target object according to the actual type includes:

[0066] Determine whether the actual type of the target object is a biological type or a non-biological type (such as a device type);

[0067] According to the biological species or non-biological species combined with the previously registered core data information, the area is divided into at least one important sub-region and a non-important sub-region.

[0068] That is to say, when the actual type of the target object is a biological species, in order to ensure the comprehensiveness of the data, it is necessary to perform 360-degree image recognition of the entire biological species without viewing angles to ensure the correctness of subsequent identification and comparison, thereby reducing unnecessary troubleshooting time. When the actual type of the target object is a non-biological species, such as mechanical equipment, only the core part of the non-biological species needs to be image recognized in a 360-degree non-viewing manner to ensure the correctness of subsequent identification and comparison, thereby reducing unnecessary data collection and improving operational efficiency.

[0069] Specifically, in some implementations, in S2, the step of acquiring actual view information of each sub-region by using an AI imaging device includes:

[0070] Driving the three-dimensional moving device to move so that the AI ​​imaging device can cyclically move in horizontal, vertical and longitudinal directions;

[0071] Using the three-dimensional mobile device to collect dynamic views in the three-dimensional space and static views in the two-dimensional space of each sub-region;

[0072] The actual view information is obtained by comparing and verifying the dynamic view and the static view.

[0073] That is to say, the three-dimensional moving device includes a horizontal moving mechanism (such as a horizontal moving lead screw or a horizontal moving cylinder), a vertical moving mechanism (such as a vertical moving lead screw or a vertical moving cylinder) and a longitudinal moving mechanism (longitudinal moving cylinder or a longitudinal moving cylinder); so that the AI ​​imaging device can perform cyclical motion along the horizontal, vertical and longitudinal directions; then at least one group (preferably three groups) of dynamic views and static views of each area are collected, and finally the dynamic view and the static view are compared and verified to obtain the actual view information; thereby verifying the authenticity and accuracy of the static view acquisition through the dynamic view, thereby reducing the need for manual intervention, significantly improving the efficiency of data processing, and enhancing the system's ability to learn and adapt to new situations.

[0074] Specifically, in some implementations, in S3, the step of optimizing preprocessing the actual view information to obtain preprocessed view information includes:

[0075] Inputting the collected actual view information into a preset physical model;

[0076] The physical model preprocesses the spectral image to obtain preprocessed view information; wherein the preprocessing includes correcting the geometric deformation of the image, wavelength correction, removing the influence of the atmosphere or water vapor, etc., to obtain more accurate spectral information. The model is designed based on optical principles and spectral characteristics and is used for image preprocessing and correction.

[0077] By preprocessing the actual view information, interference factors are eliminated, errors are reduced, and the image analysis and recognition capabilities are improved. For example, the actual view information of a dynamic scene is input into the HysiCAD (Hyperspectral Imaging Computer Aided Design) physical model for processing. The HysiCAD model is based on the principles of optics and spectroscopy and can perform geometric correction and wavelength correction on spectral images.

[0078] Specifically, in some implementations, in S4, the step of extracting and processing the pre-processed view information to obtain actual image features includes:

[0079] Using a convolutional neural network deep learning model, extracting features from the preprocessed view information to obtain extracted image features, thereby obtaining the actual image features. The actual image features at least include contour features, shape features, texture features, color features, color distribution features, dynamic image sequence features, and surrounding environment background features of the target object;

[0080] Wherein, by training on the large-scale data set, high-level features in the actual image features of the target object are learned through CNN;

[0081] Furthermore, the convolutional neural network calculation formula is as follows:

[0082] Convolution operation: Assume that the input image is I, the convolution kernel (filter) is K, and the output feature map is O; the sizes of the input image and convolution kernel are W×H and F×F respectively, the stride is S, and the padding is P; then the convolution operation is calculated as follows:

[0083] O(i, j) = ∑[I*K(i, j)]

[0084] Among them, O(i, j) is a pixel value on the output feature map, I*K(i, j) represents the result of element-by-element multiplication and addition of the input image I and the convolution kernel K at position (i, j); this operation moves on the entire image with a step size of S until the entire image is traversed.

[0085] Specifically, in some embodiments, in S5, the step of comparing the preprocessed view information and the actual image features by using AI image recognition technology to obtain a comparison result includes:

[0086] Comparing the preprocessed view information with the reference image in the database using AI image recognition technology to form a first result;

[0087] Compare the actual image features with reference features of the reference image in the database using AI image recognition technology to form a second result;

[0088] When the first result is that the two data are consistent and the second result is that the two data are consistent, feedback information that the state of the target object is normal is generated; and the AI ​​imaging device is turned off; so as to achieve the effect of saving resources and protecting the device;

[0089] When the first result is that the two data are inconsistent and / or the second result is that the two data are inconsistent, feedback information of abnormal state of the target object is generated, and the maintenance manager is notified at the same time; so as to realize fixed-point and targeted rescue maintenance operations.

[0090] The present invention also provides a visualization system based on AI image recognition.

[0091] Specifically, Figure 3 As shown, the visualization system based on AI image recognition includes:

[0092] The acquisition module 620 is used to obtain the actual type of the target object, and perform regional processing on the target object according to the actual type; and acquire the actual view information of each sub-region through the AI ​​imaging device;

[0093] An optimization module 610, the optimization module 610 is used to optimize preprocessing of the actual view information to obtain preprocessed view information;

[0094] A processing module 640, the processing module 640 is used to extract and process the pre-processed view information to obtain actual image features;

[0095] The comparison module 630 is used to compare the pre-processed view information and the actual image features through AI image recognition technology to obtain a comparison result.

[0096] In addition, an embodiment of the present invention further provides a computer-readable storage medium, on which a visualization program based on AI image recognition is stored. When the visualization program based on AI image recognition is executed by a processor, the following operations are implemented:

[0097] Obtaining the actual type of the target object, and performing regional processing on the target object according to the actual type;

[0098] The actual view information of each sub-area is acquired through AI imaging equipment;

[0099] Optimizing preprocessing of the actual view information to obtain preprocessed view information;

[0100] Extracting and processing the preprocessed view information to obtain actual image features;

[0101] The preprocessed view information and the actual image features are compared by using AI image recognition technology to obtain a comparison result.

[0102] Furthermore, the step of obtaining the actual type of the target object and performing regional processing on the target object according to the actual type includes:

[0103] Determine whether the actual type of the target object is a biological type or a non-biological type (such as a device type);

[0104] According to the biological species or non-biological species combined with the previously registered core data information, the area is divided into at least one important sub-region and a non-important sub-region.

[0105] Furthermore, the step of acquiring actual view information of each sub-region by using an AI imaging device includes:

[0106] Driving the three-dimensional moving device to move so that the AI ​​imaging device can cyclically move in horizontal, vertical and longitudinal directions;

[0107] Using the three-dimensional mobile device to collect dynamic views in the three-dimensional space and static views in the two-dimensional space of each sub-region;

[0108] The actual view information is obtained by comparing and verifying the dynamic view and the static view.

[0109] Furthermore, the step of optimizing preprocessing the actual view information to obtain preprocessed view information includes:

[0110] Inputting the collected actual view information into a preset physical model;

[0111] The physical model preprocesses the spectral image to obtain preprocessed view information; wherein the preprocessing includes correcting geometric deformation of the image, wavelength correction, and removing the influence of the atmosphere or water vapor.

[0112] Furthermore, the step of extracting and processing the preprocessed view information to obtain actual image features includes:

[0113] A convolutional neural network deep learning model is used to perform feature extraction on the preprocessed view information to obtain extracted image features, so as to obtain the actual image features.

[0114] In which, training is performed on the large-scale data set, and high-level features in the actual image features of the target object are learned through CNN.

[0115] Furthermore, the convolutional neural network calculation formula is as follows:

[0116] Convolution operation: Assume that the input image is I, the convolution kernel (filter) is K, and the output feature map is O; the sizes of the input image and convolution kernel are W×H and F×F respectively, the stride is S, and the padding is P; then the convolution operation is calculated as follows:

[0117] O(i, j) = ∑[I*K(i, j)]

[0118] Among them, O(i, j) is a pixel value on the output feature map, I*K(i, j) represents the result of element-by-element multiplication and addition of the input image I and the convolution kernel K at position (i, j); this operation moves on the entire image with a step size of S until the entire image is traversed.

[0119] Furthermore, the step of comparing the preprocessed view information and the actual image features by using AI image recognition technology to obtain a comparison result includes:

[0120] Comparing the preprocessed view information with the reference image in the database using AI image recognition technology to form a first result;

[0121] Compare the actual image features with reference features of the reference image in the database using AI image recognition technology to form a second result;

[0122] When the first result is that the two data are consistent and the second result is that the two data are consistent, feedback information that the state of the target object is normal is generated; and the AI ​​imaging device is turned off;

[0123] When the first result is that the data of the two are inconsistent and / or the second result is that the data of the two are inconsistent, feedback information indicating that the state of the target object is abnormal is generated, and the maintenance manager is notified at the same time.

[0124] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode includes only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

[0125] According to the disclosure and teaching of the above description, those skilled in the art to which the present invention belongs can also change and modify the above embodiment. Therefore, the present invention is not limited to the above specific embodiment, and any obvious improvement, replacement or modification made by those skilled in the art on the basis of the present invention belongs to the protection scope of the present invention. In addition, although some specific terms are used in this specification, these terms are only for the convenience of description and do not constitute any limitation to the present invention.

Claims

1. A visualization method based on AI image recognition, characterized in that: include: S1. Obtaining the actual type of the target object, and performing regional processing on the target object according to the actual type; S2. Acquire the actual view information of each sub-area through AI imaging equipment; S3, optimizing preprocessing of the actual view information to obtain preprocessed view information; S4, extracting and processing the preprocessed view information to obtain actual image features; S5. Compare the preprocessed view information and the actual image features by using AI image recognition technology to obtain a comparison result.

2. The visualization method based on AI image recognition according to claim 1, characterized in that: The step of obtaining the actual type of the target object and performing regional processing on the target object according to the actual type includes: Determine whether the actual species of the target object is a biological species or a non-biological species; According to the biological species or non-biological species combined with the previously registered core data information, the area is divided into at least one important sub-region and a non-important sub-region.

3. The visualization method based on AI image recognition according to claim 1, characterized in that: The step of acquiring actual view information of each sub-region by using an AI imaging device includes: Driving the three-dimensional moving device to move so that the AI ​​imaging device can cyclically move in horizontal, vertical and longitudinal directions; Using the three-dimensional mobile device to collect dynamic views in the three-dimensional space and static views in the two-dimensional space of each sub-region; The actual view information is obtained by comparing and verifying the dynamic view and the static view.

4. The visualization method based on AI image recognition according to claim 1, characterized in that: The step of optimizing and preprocessing the actual view information to obtain preprocessed view information includes: Inputting the collected actual view information into a preset physical model; The physical model preprocesses the spectral image to obtain preprocessed view information; wherein the preprocessing includes correcting geometric deformation of the image, wavelength correction, and removing the influence of the atmosphere or water vapor.

5. The visualization method based on AI image recognition according to claim 1, characterized in that: The step of extracting and processing the preprocessed view information to obtain actual image features comprises: A convolutional neural network deep learning model is used to perform feature extraction on the preprocessed view information to obtain extracted image features, so as to obtain the actual image features. In which, training is performed on the large-scale data set, and high-level features in the actual image features of the target object are learned through CNN.

6. The visualization method based on AI image recognition according to claim 5, characterized in that: The convolutional neural network calculation formula is as follows: Convolution operation: Assume that the input image is I, the convolution kernel (filter) is K, and the output feature map is O; the sizes of the input image and convolution kernel are W×H and F×F respectively, the stride is S, and the padding is P; then the convolution operation is calculated as follows: O(i, j) = ∑[I*K(i, j)] Among them, O(i, j) is a pixel value on the output feature map, I*K(i, j) represents the result of element-by-element multiplication and addition of the input image I and the convolution kernel K at position (i, j); this operation moves on the entire image with a step size of S until the entire image is traversed.

7. The visualization method based on AI image recognition according to claim 1, characterized in that: The step of comparing the preprocessed view information and the actual image features by using the AI ​​image recognition technology to obtain a comparison result includes: Comparing the preprocessed view information with the reference image in the database using AI image recognition technology to form a first result; Compare the actual image features with reference features of the reference image in the database using AI image recognition technology to form a second result; When the first result is that the two data are consistent and the second result is that the two data are consistent, feedback information that the state of the target object is normal is generated; and the AI ​​imaging device is turned off; When the first result is that the data of the two are inconsistent and / or the second result is that the data of the two are inconsistent, feedback information indicating that the state of the target object is abnormal is generated, and the maintenance manager is notified at the same time.

8. A visualization system based on AI image recognition, characterized in that: Used to execute the visualization method based on AI image recognition according to any one of claims 1 to 7 above; wherein the visualization system based on AI image recognition comprises: An acquisition module, the acquisition module is used to obtain the actual type of the target object, and perform regional processing on the target object according to the actual type; and acquire the actual view information of each sub-region through the AI ​​imaging device; an optimization module, the optimization module being used to optimize preprocessing of the actual view information to obtain preprocessed view information; A processing module, the processing module is used to extract and process the pre-processed view information to obtain actual image features; A comparison module is used to compare the preprocessed view information and the actual image features through AI image recognition technology to obtain a comparison result.

9. A visualization terminal based on AI image recognition, characterized in that: include: A memory, a processor, and a visualization program based on AI image recognition stored in the memory and executable on the processor, wherein the visualization program based on AI image recognition, when executed by the processor, implements the steps of the visualization method based on AI image recognition as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a visualization program based on AI image recognition, and when the visualization program based on AI image recognition is executed by the processor, the steps of the visualization method based on AI image recognition as described in any one of claims 1 to 7 are implemented.

Images