Image recognition method, device and storage medium for remote inspection of intelligent power distribution room

Abstract

The present invention relates to an image recognition method for remote inspection applications of intelligent power distribution rooms. In view of the content of the inspection report of the intelligent power distribution room, videos of different states of objects and equipment involved in the civil engineering information and equipment information in the power room are shot, and the video is converted into pictures frame by frame using a video playback tool. An image library and a model library are established using an image recognition basic service application, and pictures are uploaded to the image library, and the pictures are annotated. The annotated results are trained as a model using an algorithm service application. In view of the image recognition content of the inspection of the intelligent power distribution room, image recognition function configuration, scene step configuration and scene configuration are performed, and after the scene configuration is completed, it is provided to various image recognition application scenarios. The video preset position of each recognized object is set using an intelligent operation and inspection application and a video service application, and the camera is associated with the preset position for initialization, and an association relationship is established between the preset position and the scene step.

Description

Technical Field

[0001] The present invention relates to the field of power production management, and in particular to an image recognition method, device and storage medium for remote inspection of an intelligent power distribution room. Background Art

[0002] In recent years, power companies have been building distribution networks and laying distribution network equipment on a large scale. At the same time, the amount of equipment operation and maintenance has continued to increase, and the risk of equipment operation has also increased. The workload of inspection team members is serious. With the construction of intelligent distribution rooms, the laying of secondary monitoring equipment can monitor the operation data of distribution network equipment in real time. Considering the future advancement of the system and improving the work efficiency and quality of inspection personnel, it is necessary to conduct in-depth research on the application of artificial intelligence, big data analysis, machine learning and other technologies in key technologies of intelligent inspection, do a good job in the top-level design of the system, support distribution network intelligent inspection technology, and promote the "intelligent" management of distribution networks.

[0003] The image recognition method for remote inspection of intelligent power distribution rooms can solve the problem of lack of inspection personnel, reduce enterprise operating costs, solve equipment problems in a timely manner, and improve equipment self-healing capabilities. Through the application of image recognition algorithms and image recognition technology application research, it is reflected in intelligent applications, effectively reducing the workload of personnel inspections, reducing personnel expenses, achieving high reliability and high economic benefits, enhancing application scenarios, and realizing accurate, safe, and reliable equipment inspection and operation and maintenance modes, improving the level of differentiated operation and maintenance management, and supporting intelligent inspection applications for distribution network production. Summary of the invention

[0004] In order to solve the above technical problems or at least partially solve the above technical problems, the present invention provides an image recognition method, device and storage medium for remote inspection application of an intelligent power distribution room.

[0005] In a first aspect, the present invention provides an image recognition method for remote inspection application of an intelligent power distribution room, comprising:

[0006] Based on the inspection report content of the intelligent distribution room, videos of the objects and equipment in different states involving the civil engineering information and equipment information in the distribution room are shot, and video playback tools are used to convert the videos into pictures frame by frame.

[0007] Furthermore, the image library and model library are established using the basic image recognition service application, the images are uploaded to the image library, the images are annotated, and the algorithm service application is used to perform model training on the annotated results.

[0008] Furthermore, for the image recognition content of the intelligent distribution room inspection, image recognition function configuration, scene step configuration and scene configuration are performed, and after the scene configuration is completed, it is provided to various image recognition application scenarios.

[0009] Furthermore, the video preset positions of each recognized object are set by using the intelligent operation and inspection application and the video service application, the camera is associated with the preset position for initialization, and the preset position is associated with the scene step.

[0010] Furthermore, task information is obtained according to the remote inspection task number, including the district bureau, feeder, power room, equipment ledger information, inspection personnel, etc., and the equipment ledger information is initialized to the inspection report page.

[0011] Furthermore, the intelligent operation and maintenance application is used to obtain real-time equipment monitoring data, and the data and diagnostic analysis results are displayed in the equipment list by associating the monitoring equipment with the primary equipment.

[0012] Furthermore, according to the remote patrol application scenario, the scene steps are preset, the association between the scene steps and the preset positions, as well as the relationship between the preset positions and the cameras are established, the camera preset positions are called in sequence, and the images are captured using the video stream and image recognition is performed through the image recognition basic service application. The recognized images and results are displayed in the civil engineering information and equipment information lists. The recognition results support manual confirmation. If it is still an abnormal state after manual confirmation, the abnormal information will be displayed in the equipment abnormality list.

[0013] Furthermore, after the remote inspection is completed, the inspection results are saved, including inspection work order operation information, equipment monitoring data, image recognition status information, exception information and other data. If there are abnormal records in the equipment abnormality list, a status inspection work order is issued, and the operation and maintenance personnel go to the site to deal with the problem, realizing closed-loop management of work orders, improving differentiated operation and maintenance management capabilities, and supporting intelligent inspection applications for distribution network production.

[0014] In the second aspect, the present application provides a system of an image recognition method for remote inspection applications in intelligent distribution rooms, characterized in that it includes: an image recognition basic service application, a file service application, an algorithm service application, an intelligent operation and inspection application, and a video service application connected through a network.

[0015] In the third aspect, the present application provides a storage medium for an image recognition method for a remote patrol application in an intelligent distribution room. The storage medium for implementing the image recognition method for a remote patrol application in an intelligent distribution room stores at least one instruction, and reads and executes the instruction to implement the image recognition method for a remote patrol application in an intelligent distribution room.

[0016] The above technical solution provided by the embodiment of the present invention has the following advantages compared with the prior art:

[0017] This application targets the content of the inspection report of the intelligent power distribution room, shoots videos of different states of objects and equipment involved in the civil engineering information and equipment information in the power room, and uses video playback tools to convert the video frame by frame into pictures. Use the image recognition basic service application to establish an image library and a model library, upload pictures to the image library, annotate the pictures, and use the algorithm service application to train the model of the annotated results. For the image recognition content of the inspection of the intelligent power distribution room, perform image recognition function configuration, scene step configuration, and scene configuration, and provide various image recognition application scenarios after completing the scene configuration. Use the intelligent operation and inspection application and the video service application to set the video preset position of each recognized object, associate the camera with the preset position for initialization, and establish an association relationship between the preset position and the scene step. Obtain task information according to the remote inspection task number, including district bureau, feeder, power room, equipment ledger information, inspection personnel, etc., and initialize the equipment ledger information to the inspection report page. Use the intelligent operation and inspection application to obtain real-time equipment monitoring data, and display the data and diagnostic analysis results in the equipment list through the association relationship between the monitoring equipment and the primary equipment. According to the remote inspection application scenario, the scene steps are preset, and the relationship between the scene steps and the preset positions, as well as the relationship between the preset positions and the camera, the camera preset positions are called in sequence, and the video stream is used to capture pictures for image recognition through the image recognition basic service application. The recognized pictures and results are displayed in the civil engineering information and equipment information list. The recognition results support manual confirmation. If it is still abnormal after manual confirmation, the abnormal information is displayed in the equipment abnormality list. After the remote inspection is completed, the inspection results are saved, including inspection work order operation information, equipment monitoring data, image recognition status information, abnormal information and other data. If there are abnormal records in the equipment abnormality list, a status inspection work order is issued, and the operation and maintenance personnel go to the site to handle the problem, realizing closed-loop management of work orders and improving differentiated operation and maintenance management capabilities. Research image recognition methods for remote inspection applications in intelligent distribution rooms, realize intelligent inspections, differentiated operation and maintenance management, and support intelligent inspection applications for distribution network production. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0020] Figure 1 The overall design flow chart for developing and executing image recognition for remote inspection application of intelligent power distribution room provided by the embodiment of the present invention;

[0021] Figure 2 An implementation flow chart of formulating and executing image recognition for remote inspection application of intelligent power distribution room provided by an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of an image recognition system for remote inspection of a smart power distribution room provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0025] Example 1

[0026] The embodiment of the present invention provides an image recognition method for a remote inspection application of an intelligent power distribution room, comprising:

[0027] See also Figure 2 As shown in the figure, the process of performing image recognition for the remote inspection application of the intelligent power distribution room includes:

[0028] S10, based on the inspection report content of the intelligent power distribution room, take videos of different states of objects and equipment involved in the civil engineering information and equipment information in the power distribution room, and use video playback tools to convert the videos frame by frame into pictures.

[0029] S20, using the image recognition basic service application to establish an image library and a model library, uploading the image to the file service application image library, annotating the image, and using the algorithm service application to perform model training on the annotated results.

[0030] S30, for the image recognition content of the intelligent distribution room inspection, image recognition function configuration, scene step configuration and scene configuration are performed, and after the scene configuration is completed, it is provided to various image recognition application scenarios.

[0031] S40, using the intelligent operation and inspection application and the video service application to set the video preset position of each recognized object, associate the camera with the preset position for initialization, and establish an association relationship between the preset position and the scene step.

[0032] S50, obtaining task information according to the remote patrol task number.

[0033] S60, using the intelligent operation and inspection application to obtain real-time equipment monitoring data, and displaying the data and diagnostic analysis results in the equipment list by associating the monitoring equipment with the primary equipment.

[0034] S70, according to the remote patrol application scenario, preset the scene steps, establish the association between the scene steps and the preset positions, and the relationship between the preset positions and the camera, and call the camera preset positions in sequence. S80, use the video stream to capture pictures and perform image recognition through the image recognition basic service application, display the recognition pictures and results in the civil engineering information and equipment information list, and support manual confirmation of the recognition results. If it is still abnormal after manual confirmation, the abnormal information will be displayed in the equipment abnormality list.

[0035] S90, after the remote inspection is completed, the inspection result is saved. If there is an abnormal record in the device abnormality list, S100 is executed, otherwise S110 is executed.

[0036] S100, dispatch status inspection work orders, and operation and maintenance personnel go to the site to handle the problems, realizing closed-loop management of work orders.

[0037] S110 generates remote inspection reports through intelligent operation and inspection applications, supports report query and export, and supports intelligent inspection applications for distribution network production.

[0038] Example 2

[0039] See also Figure 3 As shown, an embodiment of the present application provides a system of an image recognition method for a remote inspection application of an intelligent power distribution room, including: an image recognition basic service application, a file service application, an algorithm service application, an intelligent operation and inspection application, and a video service application connected through a network.

[0040] Example 3

[0041] An embodiment of the present application provides a storage medium for an image recognition method for a remote patrol application in an intelligent power distribution room. The storage medium for implementing the image recognition method for a remote patrol application in an intelligent power distribution room stores at least one instruction, and reading and executing the instruction implements the image recognition method for a remote patrol application in an intelligent power distribution room.

[0042] This application targets the content of the inspection report of the intelligent power distribution room, shoots videos of different states of objects and equipment involved in the civil engineering information and equipment information in the power room, and uses video playback tools to convert the video frame by frame into pictures. Use the image recognition basic service application to establish an image library and a model library, upload pictures to the image library, annotate the pictures, and use the algorithm service application to train the model of the annotated results. For the image recognition content of the inspection of the intelligent power distribution room, perform image recognition function configuration, scene step configuration, and scene configuration, and provide various image recognition application scenarios after completing the scene configuration. Use the intelligent operation and inspection application and the video service application to set the video preset position of each recognized object, associate the camera with the preset position for initialization, and establish an association relationship between the preset position and the scene step. Obtain task information according to the remote inspection task number, including district bureau, feeder, power room, equipment ledger information, inspection personnel, etc., and initialize the equipment ledger information to the inspection report page. Use the intelligent operation and inspection application to obtain real-time equipment monitoring data, and display the data and diagnostic analysis results in the equipment list through the association relationship between the monitoring equipment and the primary equipment. According to the remote inspection application scenario, the scene steps are preset, and the relationship between the scene steps and the preset positions, as well as the relationship between the preset positions and the camera, the camera preset positions are called in sequence, and the video stream is used to capture pictures for image recognition through the image recognition basic service application. The recognized pictures and results are displayed in the civil engineering information and equipment information list. The recognition results support manual confirmation. If it is still abnormal after manual confirmation, the abnormal information is displayed in the equipment abnormality list. After the remote inspection is completed, the inspection results are saved, including inspection work order operation information, equipment monitoring data, image recognition status information, abnormal information and other data. If there are abnormal records in the equipment abnormality list, a status inspection work order is issued, and the operation and maintenance personnel go to the site to handle the problem, realizing closed-loop management of work orders and improving differentiated operation and maintenance management capabilities. Research image recognition methods for remote inspection applications in intelligent distribution rooms, realize intelligent inspections, differentiated operation and maintenance management, and support intelligent inspection applications for distribution network production.

[0043] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0044] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0045] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0046] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. An image recognition method for remote inspection of intelligent power distribution room, characterized by: include: According to the inspection report of the intelligent power distribution room, take videos of the objects and equipment in different states involved in the civil engineering information and equipment information in the power distribution room, and use the video playback tool to convert the video frame by frame into pictures; Use the basic image recognition service application to build an image library and a model library, upload images to the image library, annotate the images, and use the algorithm service application to train the model based on the annotated results; For the image recognition content of the intelligent power distribution room inspection, image recognition function configuration, scene step configuration and scene configuration are performed, and after the scene configuration is completed, it is provided to various image recognition application scenarios; Use the intelligent operation and inspection application and the video service application to set the video preset position of each recognized object, associate the camera with the preset position for initialization, and establish an association relationship between the preset position and the scene step; Obtain task information according to the remote inspection task number, including district bureau, feeder, power room, equipment ledger information, and inspection personnel, and initialize the equipment ledger information to the inspection report page; Use intelligent operation and inspection applications to obtain real-time equipment monitoring data, and display the data and diagnostic analysis results in the equipment list by associating the monitoring equipment with the primary equipment; Preset the scene steps according to the remote patrol application scenario, establish the association between the scene steps and the preset positions, as well as the relationship between the preset positions and the camera, call the camera preset positions in sequence, use the video stream to capture pictures, and perform image recognition through the image recognition basic service application, and display the recognition pictures and results in the civil engineering information and equipment information list. The recognition results support manual confirmation. If the abnormal state is still after manual confirmation, the abnormal information will be displayed in the equipment abnormality list; After the remote inspection is completed, the inspection results are saved, including inspection work order operation information, equipment monitoring data, image recognition status information, and exception information. If there are abnormal records in the equipment exception list, a status inspection work order is issued, and the operation and maintenance personnel go to the site to deal with the problem, realizing closed-loop management of work orders, improving differentiated operation and maintenance management capabilities, and supporting intelligent inspection applications for distribution network production.

2. The image recognition method for remote inspection of intelligent power distribution room according to claim 1 is characterized by: The image library nodes and image information of the remote patrol application scenario are established through the image recognition basic service. The image library node information includes the parent node number, node type, node name, and node order. The image information includes the image number, image name, image path, upload date, and the relationship between the image and the image library node.

3. The image recognition method for remote inspection of intelligent power distribution room according to claim 2 is characterized in that: Upload images to the image library node through the image recognition basic service in sequence according to the image type.

4. The image recognition method for remote inspection of intelligent power distribution room according to claim 2 is characterized in that: The model library nodes and model information are established through the image recognition basic service. The model library node information includes the parent node number, node type, node name, and node order. The model information includes the model number, model name, model path, version, number of training sets, number of training times, recognition rate, and the relationship between the model and the model library nodes. The established model is associated with the recognition algorithm.

5. The image recognition method for remote inspection of intelligent power distribution room according to claim 2 is characterized in that: The image is annotated through the image recognition basic service. The annotation information includes the image number, annotation coordinate information, and label name.

6. The image recognition method for remote inspection of intelligent power distribution room according to claim 2 is characterized in that: Through the basic image recognition service, the image annotation association model is completed and the training model association relationship is established.

7. The image recognition method for remote inspection of intelligent power distribution room according to claim 3 is characterized by: Establish the image recognition function configuration of civil engineering information and equipment information according to the recognition content. The image recognition function configuration content includes configuration type, associated number, and function code. The configuration type is to select the type of image recognition, which is divided into node and template. The associated number is the selected node or template number, and the function code is the code of the background service call method.

8. The image recognition method for remote inspection of intelligent power distribution room according to claim 3 is characterized by: According to the remote patrol application, the scene steps are configured. The civil engineering information establishes the lighting switch status, ground pit plate, rat-proof plate, and exhaust fan step scenes. The equipment information establishes the safety, health and environmental status, load switch, grounding knife switch, barometer reading, live display, fault indicator, and three-remote switch step scenes.

9. The image recognition method for remote inspection of intelligent power distribution room according to claim 3 is characterized by: Configure the scene according to the remote patrol scene steps and add the related scene steps to the scene configuration.

10. The image recognition method for remote inspection of intelligent power distribution room according to claim 4 is characterized in that: The intelligent operation and inspection application and the video service application are used to set the preset position information of each camera for identifying objects. The camera preset position information includes the camera number, preset position sequence number, and preset position name.

11. The image recognition method for remote inspection of intelligent power distribution room according to claim 4 is characterized in that: The relationship between the preset position and the scene step is established by using the intelligent operation and inspection application. The association information between the preset position and the scene step includes the preset position number and the scene step number.

12. A system for an image recognition method for remote inspection of an intelligent power distribution room, using the image recognition method for remote inspection of an intelligent power distribution room as claimed in any one of claims 1 to 11, characterized in that: include: Image recognition basic service applications, file service applications, algorithm service applications, intelligent operation and inspection applications, and video service applications connected through the network.

13. A storage medium for an image recognition method for remote inspection of an intelligent power distribution room, characterized in that: At least one instruction is stored thereon, and the instruction is read and executed to implement the image recognition method for remote patrol application of the intelligent power distribution room as described in any one of claims 1-11.

Images