Machine room inspection method and device based on image recognition technology

Abstract

The invention provides a machine room inspection method and device based on an image recognition technology, and relates to the technical field of operation and maintenance management and fault detection. The method comprises the following steps: moving an image acquisition device to an initial inspection position on a slide rail, and controlling the image acquisition device to perform image acquisition on a panel cabinet panel in a machine room based on a preset rotation angle and a preset focal length value to obtain an acquired image; identifying the acquired image, determining a panel area and a panel label in the acquired image, and establishing a coordinate system in the panel area of the panel; according to the panel cabinet labels, a plurality of inspection points located in the panel cabinet panel area, the inspection sequence of the inspection points and the inspection parameters of each inspection point are obtained from an inspection rule base in a matching mode; based on the routing inspection sequence and the routing inspection parameters, routing inspection is performed on the multiple routing inspection points in sequence, and routing inspection images corresponding to the routing inspection points are obtained; and identifying and analyzing the inspection image to obtain an inspection analysis result corresponding to the inspection point.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of operation and maintenance and fault detection, and in particular to a machine room inspection method based on image recognition technology and a device thereof. BACKGROUND

[0002] As important energy production facilities, the operation stability of the machine room equipment of a hydropower plant directly affects the safety and efficiency of power production. In order to ensure that the equipment is in a normal operating state, the hydropower plant needs to regularly inspect various types of mechanical and electrical equipment. The traditional machine room inspection method of the hydropower plant mostly relies on manual operation. The inspection personnel perform on-site inspection according to work experience and confirm the operating condition of the equipment through visual judgment. However, as the types and quantities of mechanical and electrical equipment gradually increase, the complexity and frequency of the inspection task are continuously increasing, and the pure manual inspection mode has been difficult to meet the high requirements of modern hydropower plants on equipment management efficiency and accuracy. SUMMARY

[0003] The present application aims to at least partially solve one of the technical problems in the related art.

[0004] To this end, one object of the present application is to provide a machine room inspection method based on image recognition technology. The method includes moving an image acquisition device to an initial inspection position on a slide rail, controlling the image acquisition device to acquire images of the panel of a disk cabinet in the machine room based on a preset rotation angle and a preset focal length value, identifying the acquired images to determine a disk cabinet panel region and a disk cabinet label in the acquired images, and establishing a coordinate system in the disk cabinet panel region. According to the disk cabinet label, a plurality of inspection points located in the disk cabinet panel region, an inspection order of the plurality of inspection points, and an inspection parameter of each inspection point are matched from an inspection rule library. The inspection parameter includes an inspection point region determination parameter, an image acquisition device adjustment parameter, and an image acquisition device acquisition parameter. Based on the inspection order and the inspection parameter, the plurality of inspection points are sequentially inspected to obtain an inspection image corresponding to each inspection point. The inspection image is identified and analyzed to obtain an inspection analysis result corresponding to the inspection point.

[0005] A second object of the present application is to provide a machine room inspection device based on image recognition technology.

[0006] A third object of the present application is to provide an electronic device.

[0007] A fourth object of the present application is to provide a non-transitory computer readable storage medium.

[0008] A fifth object of the present application is to provide a computer program product.

[0009] To achieve the above object, the first aspect of the present application proposes a machine room inspection method based on image recognition technology, comprising: moving an image acquisition device to an initial inspection position on a slide rail, and controlling the image acquisition device to acquire images of the cabinet panel in the machine room based on a preset rotation angle and a preset focal length value, to obtain acquisition images; identifying the acquisition images to determine the cabinet panel region and the cabinet label in the acquisition images, and establishing a coordinate system in the cabinet panel region; according to the cabinet label, matching a plurality of inspection points located in the cabinet panel region, an inspection order of the plurality of inspection points, and an inspection parameter of each inspection point from an inspection rule library, the inspection parameter including an inspection point region determination parameter, an image acquisition device adjustment parameter, and an image acquisition device acquisition parameter; based on the inspection order and the inspection parameter, sequentially inspecting the plurality of inspection points to obtain an inspection image corresponding to each inspection point; identifying and analyzing the inspection image to obtain an inspection analysis result corresponding to the inspection point.

[0010] According to one embodiment of the present application, when any one of the plurality of inspection points is inspected, it includes: determining the inspection point region corresponding to the inspection point on the coordinate system according to the inspection point region determination parameter corresponding to the inspection point, wherein the inspection point region determination parameter includes the inspection point coordinates and the coordinate expansion parameter; adjusting the image acquisition device according to the image acquisition device adjustment parameter corresponding to the inspection point, wherein the image acquisition device adjustment parameter includes at least one of the target inspection position, the target rotation angle, and the target focal length value; based on the adjusted image acquisition device, acquiring images of the inspection point region according to the image acquisition device acquisition parameter corresponding to the inspection point to obtain the inspection image of the inspection point, wherein the image acquisition device acquisition parameter includes the acquisition frequency and the acquisition number.

[0011] According to one embodiment of the present application, adjusting the image acquisition device according to the image acquisition device adjustment parameter corresponding to the inspection point includes: obtaining the current inspection position, the current rotation angle, and the current focal length value of the image acquisition device; obtaining the angle difference between the target rotation angle and the current rotation angle; obtaining the focal length difference between the target focal length value and the current focal length value; obtaining the displacement difference between the target inspection position and the current inspection position; adjusting the image acquisition device based on the angle difference, the focal length difference, and the displacement difference.

[0012] According to one embodiment of the present application, in response to the inspection point being a flashing state indicator light, the acquisition frequency corresponding to the inspection point is greater than half of the flashing frequency of the flashing state indicator light and less than the flashing frequency of the flashing state indicator light, and the acquisition number corresponding to the inspection point is greater than or equal to 3.

[0013] According to one embodiment of the present application, the inspection image is subjected to recognition analysis to obtain the inspection analysis result corresponding to the inspection point, including: associating the inspection analysis results corresponding to the inspection points having the master-slave relationship to obtain an inspection analysis result set; and determining the inspection analysis result corresponding to the inspection point having the master-slave relationship according to the inspection analysis result in the inspection analysis result set.

[0014] According to one embodiment of the present application, after the inspection image is subjected to recognition analysis to obtain the inspection analysis result corresponding to the inspection point, the method further includes: determining an abnormal inspection point from the plurality of inspection points according to the inspection analysis result; and generating an alarm information according to the disk cabinet label, the abnormal inspection point and the inspection analysis result of the abnormal inspection point.

[0015] According to one embodiment of the present application, the machine room inspection method based on image recognition technology further includes: identifying the collected image to determine the trapezoidal correction parameter of the collected image; performing trapezoidal-to-rectangular processing on the inspection image based on the trapezoidal correction parameter, and generating an inspection log based on the inspection image after the trapezoidal-to-rectangular processing.

[0016] According to one embodiment of the present application, the slide rail is in S-shaped wiring and is suspendedly installed on the top of the machine room.

[0017] To achieve the above purpose, a second aspect of the present application provides a machine room inspection device based on image recognition technology, including: a collection module configured to move an image collection device to an inspection position on a slide rail, and control the image collection device to collect images of disk cabinet panels in a machine room based on a preset rotation angle and a preset focal length value, to obtain collected images; an identification module configured to identify the collected images, determine disk cabinet panel regions and disk cabinet labels in the collected images, and establish a coordinate system in the disk cabinet panel regions; a matching module configured to match a plurality of inspection points located in the disk cabinet panel regions, an inspection order of the plurality of inspection points, and an inspection parameter of each inspection point from an inspection rule library according to the disk cabinet labels, the inspection parameter including an inspection point region determination parameter, an image collection device adjustment parameter, and an image collection device collection parameter; an inspection module configured to sequentially inspect the plurality of inspection points based on the inspection order and the inspection parameter, to obtain an inspection image corresponding to each inspection point; and an analysis module configured to perform recognition analysis on the inspection image to obtain an inspection analysis result corresponding to the inspection point.

[0018] According to one embodiment of the present application, the inspection module is further configured to: determine a corresponding inspection point area of the inspection point on the coordinate system according to an inspection point area determination parameter corresponding to the inspection point, wherein the inspection point area determination parameter comprises an inspection point coordinate and a coordinate expansion parameter; adjust the image acquisition device according to an image acquisition device adjustment parameter corresponding to the inspection point, wherein the image acquisition device adjustment parameter comprises at least one of a target inspection position, a target rotation angle, and a target focal length value; and acquire an image of the inspection point area according to an image acquisition device acquisition parameter corresponding to the image acquisition device after the adjustment, to obtain an inspection image of the inspection point, wherein the image acquisition device acquisition parameter comprises an acquisition frequency and an acquisition quantity.

[0019] According to one embodiment of the present application, the inspection module is further configured to: acquire a current inspection position, a current rotation angle, and a current focal length value of the image acquisition device; acquire an angle difference between the target rotation angle and the current rotation angle; acquire a focal length difference between the target focal length value and the current focal length value; acquire a displacement difference between the target inspection position and the current inspection position; and adjust the image acquisition device based on the angle difference, the focal length difference, and the displacement difference.

[0020] According to one embodiment of the present application, in response to the inspection point being a flashing state indicator light, the acquisition frequency corresponding to the inspection point is greater than half of a flashing frequency of the flashing state indicator light and less than the flashing frequency of the flashing state indicator light, and the acquisition quantity corresponding to the inspection point is greater than or equal to 3.

[0021] According to one embodiment of the present application, the analysis module is further configured to: associate the inspection analysis results corresponding to the inspection points having a master-slave relationship to obtain an inspection analysis result set; and determine the inspection analysis result corresponding to the inspection point having the master-slave relationship according to the inspection analysis results in the inspection analysis result set.

[0022] According to one embodiment of the present application, the machine room inspection device based on image recognition technology further comprises an alarm module, which is configured to: determine an abnormal inspection point from the plurality of inspection points according to the inspection analysis result; and generate an alarm information according to the disk cabinet label, the abnormal inspection point, and the inspection analysis result of the abnormal inspection point.

[0023] According to one embodiment of the present application, the machine room inspection device based on image recognition technology further comprises a log generation module, which is configured to: identify the acquired image to determine a trapezoidal correction parameter of the acquired image; perform trapezoidal-to-rectangular processing on the inspection image based on the trapezoidal correction parameter, and generate an inspection log based on the inspection image after the trapezoidal-to-rectangular processing.

[0024] According to one embodiment of the present application, the slide rail is in an S-shaped layout and is suspendedly installed on the top of the machine room.

[0025] To achieve the above object, the third aspect of the present application provides an electronic device, comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to implement the image recognition technology-based machine room inspection method according to the first aspect of the present application.

[0026] To achieve the above object, the fourth aspect of the present application provides a non-transitory computer readable storage medium storing computer instructions, wherein the computer instructions are used to implement the image recognition technology-based machine room inspection method according to the first aspect of the present application.

[0027] To achieve the above object, the fifth aspect of the present application provides a computer program product, comprising a computer program, wherein the computer program is executed by a processor to implement the image recognition technology-based machine room inspection method according to the first aspect of the present application.

[0028] The present application at least has the following beneficial effects: the present application can quickly find the problems of the equipment through image recognition and automatic analysis, and timely generate analysis results, which helps to identify and handle equipment failures in advance, reduces equipment downtime and maintenance cost, improves reliability, reduces the dependence on manual operation in the inspection process, thereby reduces the labor cost, and avoids the omissions in the manual inspection process. BRIEF DESCRIPTION OF DRAWINGS

[0029] The above and / or additional aspects and advantages of the present application will become apparent and more readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:

[0030] Figure 1 is a schematic diagram of an exemplary implementation of an image recognition technology-based machine room inspection method according to an embodiment of the present application.

[0031] Figure 2 is a schematic diagram of the installation of a sliding rail according to an embodiment of the present application.

[0032] Figure 3 is a schematic diagram of establishing a coordinate system in the panel area of a disk cabinet according to an embodiment of the present application.

[0033] Figure 4 is a schematic diagram of a corresponding inspection point area of a certain inspection point according to an embodiment of the present application.

[0034] Figure 5 is a schematic diagram of an exemplary implementation of an image recognition technology-based machine room inspection method according to an embodiment of the present application.

[0035] Figure 6 is an exemplary schematic diagram of a machine room inspection device based on image recognition technology according to an embodiment of the present application.

[0036] Figure 7 is a schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0037] Embodiments of the present application will be described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numbers represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by reference to the drawings are exemplary and are intended to explain the present application, and cannot be understood as a limitation of the present application.

[0038] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data for analysis, stored data, displayed data, etc.) and signals involved in the present application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards.

[0039] Figure 1 is an exemplary embodiment of a machine room inspection method based on image recognition technology according to an embodiment of the present application, as shown in Figure 1 , the machine room inspection method based on image recognition technology comprises the following steps:

[0040] S101, moving the image acquisition device to an initial inspection position on the slide rail, and controlling the image acquisition device to acquire images of the disk cabinet panel in the machine room based on a preset rotation angle and a preset focal length value, to obtain the acquisition images.

[0041] In the present application, in order to facilitate image acquisition of the disk cabinet panel, the image acquisition device is installed on the slide rail located at the top of the machine room, Figure 2 is an installation schematic diagram of a slide rail according to an embodiment of the present application, as shown in Figure 2 , the slide rail is S-shaped and is suspendedly installed at the top of the machine room, the image acquisition device can move on the slide rail, and the flat part of the slide rail is parallel to the horizontal direction of each disk cabinet panel, so as to ensure the flatness of the horizontal direction of image acquisition.

[0042] Generally, the machine room can include multiple disk cabinets, and accordingly, there are multiple disk cabinet panels to be acquired images. In the present application, each disk cabinet panel corresponds to an initial inspection position, which is the position for acquiring the acquisition images of the disk cabinet panel, generally the corresponding point of the midpoint of the disk cabinet panel on the slide rail, so as to ensure comprehensive acquisition of the overall image of the disk cabinet panel.

[0043] It is not difficult to understand that the slide rail has two ends. In the present application, the image acquisition device can complete the acquisition and analysis of each rack panel in turn along the slide rail from end 1 to end 2 to complete a single round of inspection. After a single round of inspection, the image acquisition device can stay at end 2 and complete the acquisition and analysis of each rack panel in turn along the slide rail from end 2 to end 1 for the next inspection task, and so on.

[0044] S102, identifying the collected image to determine the rack panel area and the rack label in the collected image, and establishing a coordinate system in the rack panel area.

[0045] The collected image of the rack panel obtained above often contains other areas in addition to the rack panel area, such as the floor area, etc. Therefore, in the present application, the collected image needs to be identified to obtain the rack panel area in the collected image.

[0046] In addition, the collected image also needs to be identified (such as identifying the two-dimensional code on the rack panel for indicating the rack label) to determine the rack label. The rack label generally includes the rack number and the rack position parameter.

[0047] After determining the rack panel area in the collected image, a coordinate system needs to be established in the rack panel area for more accurate subsequent determination of the inspection point area. Figure 3 is a schematic diagram of establishing a coordinate system in the rack panel area shown in the present application, as shown in Figure 3 The rack panel area in the collected image is determined by the three-point positioning method of the upper left corner, the lower left corner, and the upper right corner of the rack panel, and a coordinate system is established in the rack panel area (the origin of the coordinate system is selected in the upper left corner area of the rack panel).

[0048] Among them, on the inspection route, if there is a rack panel that does not need to be inspected (such as a discarded rack), the image acquisition device can directly skip collecting the image of the rack panel, that is, the inspection task of the rack panel does not need to be performed.

[0049] Among them, on the inspection route, if there is a rack panel that does not need to be inspected (such as a discarded rack, which is generally shielded by the maintenance personnel on the rack label area), the image acquisition device can collect the image of the rack panel in the original manner. After identifying the collected image, if the rack label is not identified, it means that the rack panel does not need to be inspected, and the inspection task of the rack panel is directly skipped to perform the inspection of the next rack panel.

[0050] S103, according to the disk cabinet label, a plurality of inspection points located in the disk cabinet panel area, the inspection order of the plurality of inspection points and the inspection parameter of each inspection point are matched from the inspection rule library, the inspection parameter includes the inspection point area determination parameter, the image acquisition device adjustment parameter and the image acquisition device acquisition parameter.

[0051] Wherein, the inspection point refers to the equipment that needs to be checked, such as indicator light, digital display area, etc., which is used to display the running state or related parameters of the equipment.

[0052] Wherein, each disk cabinet corresponds to a unique disk cabinet label, after determining the disk cabinet label, according to the disk cabinet label, a plurality of inspection points located in the disk cabinet panel area, the inspection order of the plurality of inspection points and the inspection parameter of each inspection point are matched from the inspection rule library.

[0053] Wherein, the inspection point area determination parameter is used to determine the area of the inspection image to be collected corresponding to the inspection point. Figure 4 is a schematic diagram of the inspection point area corresponding to a certain inspection point shown in the application, as Figure 4 shown, the rectangular frame in the middle of the image is the inspection point area corresponding to a certain inspection point.

[0054] Wherein, the image acquisition device adjustment parameter is used to adjust the rotation angle, displacement, focal length and other parameters of the image acquisition device, so that the image acquisition device can more completely collect the area image corresponding to the inspection point.

[0055] Wherein, the image acquisition device acquisition parameter is used to determine the acquisition frequency and acquisition quantity of the image acquisition device when collecting the inspection point area image.

[0056] S104, based on the inspection order and the inspection parameter, the plurality of inspection points are inspected in turn to obtain the inspection image corresponding to each inspection point.

[0057] For example, if the disk cabinet panel includes N inspection points, the acquisition area of the inspection point 1 is obtained first, then the image acquisition device adjustment parameter corresponding to the inspection point 1 is obtained to adjust the image acquisition device, then the image acquisition device based on the adjusted image acquisition device acquires the image of the acquisition area of the inspection point 1 according to the image acquisition device acquisition parameter corresponding to the inspection point 1, obtains the inspection image corresponding to the inspection point 1, thereby completing the inspection image acquisition of the inspection point 1, and then starting to perform the inspection image acquisition of the inspection point 2. The method of inspection image acquisition of the inspection point 2 is similar to that of the inspection point 1, which will not be described here. In this way, until the acquisition of the inspection image of each of the N inspection points is completed according to the inspection order.

[0058] S105, the inspection image is identified and analyzed to obtain the inspection analysis result corresponding to the inspection point.

[0059] In some embodiments, artificial intelligence methods (such as anomaly recognition models) are used to compare and analyze inspection images with standard inspection images corresponding to inspection points in the database to obtain the status of equipment corresponding to the inspection points and obtain inspection analysis results corresponding to the inspection points.

[0060] In some embodiments, parameter data or indicator light color data on the inspection image are obtained, and then the inspection image is identified and analyzed according to the normal state parameters of the inspection point to obtain the inspection analysis results corresponding to the inspection point.

[0061] An embodiment of the present application proposes a computer room inspection method based on image recognition technology, which comprises the following steps: moving an image acquisition device to an initial inspection position on a slide rail, and controlling the image acquisition device to acquire an image of a cabinet panel in the computer room based on a preset rotation angle and a preset focal length value, to obtain a captured image; identifying the captured image, determining the cabinet panel area and the cabinet label in the captured image, and establishing a coordinate system in the cabinet panel area; matching multiple inspection points located in the cabinet panel area, an inspection sequence of the multiple inspection points, and inspection parameters of each inspection point from an inspection rule library according to the cabinet label, the inspection parameters including inspection point area determination parameters, image acquisition device adjustment parameters, and image acquisition device acquisition parameters; inspecting multiple inspection points in turn based on the inspection sequence and the inspection parameters, to obtain an inspection image corresponding to each inspection point; identifying and analyzing the inspection image, and obtaining an inspection analysis result corresponding to the inspection point. Through image recognition and automatic analysis, this application can quickly detect equipment problems and generate analysis results in a timely manner, which helps to identify and handle equipment failures in advance, reduce equipment downtime and maintenance costs, improve reliability, and reduce the inspection process's reliance on manual operations, thereby reducing labor costs and avoiding omissions that may occur during manual inspections.

[0062] Figure 5 This is a schematic diagram of an exemplary embodiment of a computer room inspection method based on image recognition technology shown in this application, such as Figure 5 As shown, the computer room inspection method based on image recognition technology includes the following steps:

[0063] S501, moving the image acquisition device to an initial inspection position on the slide rail, and controlling the image acquisition device to acquire images of the cabinet panel in the computer room based on a preset rotation angle and a preset focal length value to obtain acquired images.

[0064] S502 , identifying the collected image, determining the cabinet panel area and cabinet label in the collected image, and establishing a coordinate system in the cabinet panel area.

[0065] S503, according to the disk cabinet label, a plurality of inspection points located in the disk cabinet panel region, a plurality of inspection points inspection order and each inspection point inspection parameter are matched from the inspection rule library, the inspection parameter includes inspection point region determination parameter, image acquisition device adjustment parameter and image acquisition device acquisition parameter.

[0066] For the specific implementation of steps S501-S503, refer to the specific description of the relevant part in the above embodiment, which will not be repeated here.

[0067] S504, according to the inspection order, determine the next inspection point to be inspected.

[0068] For example, if the disk cabinet panel includes N inspection points, the inspection order is inspection point 1~inspection point N, the first inspection point is inspection point 1, and the next inspection point to be inspected after inspection point 1 is inspection point 2, and so on.

[0069] S505, according to the inspection point corresponding to the inspection point region determination parameter to determine the corresponding inspection point region on the coordinate system, wherein the inspection point region determination parameter includes inspection point coordinates and coordinate expansion parameter.

[0070] In some embodiments, the coordinate expansion parameter can be the number of horizontal expansion pixels and the number of vertical expansion pixels. It can be recorded in the form of (a, b) in the inspection parameter, a represents the number of horizontal expansion pixels, and b represents the number of vertical expansion pixels.

[0071] S506, according to the image acquisition device adjustment parameter corresponding to the inspection point to adjust the image acquisition device, wherein the image acquisition device adjustment parameter includes at least one of target inspection position, target rotation angle and target focal length value.

[0072] In some embodiments, if the image acquisition device adjustment parameter includes target inspection position, target rotation angle and target focal length value, the current inspection position, current rotation angle and current focal length value of the image acquisition device (i.e. the state of the image acquisition device after the last inspection point is inspected) are obtained; the angle difference between the target rotation angle and the current rotation angle is obtained; the focal length difference between the target focal length value and the current focal length value is obtained; the displacement difference between the target inspection position and the current inspection position is obtained; the image acquisition device is adjusted based on the angle difference, the focal length difference and the displacement difference.

[0073] In some embodiments, if the image acquisition device adjustment parameter includes target rotation angle and target focal length value, the image acquisition device is adjusted only based on the above obtained angle difference and focal length difference.

[0074] In some embodiments, if the image acquisition device adjustment parameter comprises a target rotation angle, the image acquisition device is adjusted only based on the above-obtained angle difference.

[0075] In some embodiments, if the image acquisition device adjustment parameter comprises a target focal length value, the image acquisition device is adjusted only based on the above-obtained focal length difference.

[0076] In some embodiments, if the image acquisition device adjustment parameter comprises a target patrol position, the image acquisition device is adjusted only based on the above-obtained displacement difference.

[0077] S507, based on the adjusted image acquisition device, image acquisition is performed on the patrol point region according to the image acquisition device acquisition parameter corresponding to the patrol point, to obtain a patrol image of the patrol point, wherein the image acquisition device acquisition parameter comprises an acquisition frequency and an acquisition number.

[0078] It is not difficult to understand that on the disk cabinet panel, there are often indicator lights in a flickering state, such as the Link / Act light of the network interface data transmission indication, which is usually in a flickering state when data flow transmission occurs. The indicator light with the "flickering" feature generally cycles in "bright-off-bright" cycles.

[0079] In this application, in order to accurately collect the patrol image of the indicator light in the flickering state, and avoid collecting only the image of the indicator light in the "bright" state or only the image of the indicator light in the "off" state, if the patrol point is an indicator light in the flickering state, the acquisition frequency corresponding to the patrol point is set to be greater than half of the flickering frequency of the indicator light in the flickering state and less than the flickering frequency of the indicator light in the flickering state, and the acquisition number corresponding to the patrol point is greater than or equal to 3. In this way, a plurality of continuous interval frame images can be taken to determine the "flickering" feature.

[0080] S508, whether all the plurality of patrol points located in the disk cabinet panel region are completely patrolled.

[0081] S509, in response to the plurality of patrol points located in the disk cabinet panel region not being completely patrolled, the above steps S504-S508 are repeatedly executed.

[0082] S510, in response to the plurality of patrol points located in the disk cabinet panel region being completely patrolled, the patrol image is identified and analyzed to obtain a patrol analysis result corresponding to the patrol point.

[0083] In some embodiments, since the device may exhibit a plurality of different operating states under different working conditions, such as natural changes in working conditions, regular rotation, network data throughput fluctuations, and other normal operating modes, it is necessary to identify and handle these complex state combinations. Specifically, the state of the device with a master-slave relationship (for example, two devices, one master and one standby) may have multiple combinations, each of which can be considered "normal" under different operating modes. For example, in the master-slave relationship, both the state of device A as master and device B as standby (A master B standby) and the state of device A as standby and device B as master (A standby B master) are normal operation, while if both devices are standby (A standby B standby), it is obviously not in normal operation state. Therefore, in this application, the database should predefine and store a set of inspection results, where the state of some inspection points may have logical associations reflecting the master-slave mode of the device. These logical associations can be described by conditional expressions, for example, {(A = 1 & B = 0) | (A = 0 & B = 1) & C}, where A and B represent the master-slave state of the device, C represents other related parameters or conditions, and the logical operators & and | represent the combination relationship of the state. These logical rules help understand the possible state changes of different inspection points under the master-slave mode.

[0084] In this application, the inspection analysis results corresponding to the inspection points with a master-slave relationship are associated to obtain a set of inspection analysis results; and the inspection analysis results corresponding to the inspection points with a master-slave relationship are determined according to the inspection analysis results in the set of inspection analysis results.

[0085] For example, if the state combination of devices A and B is (A = 1, B = 0) or (A = 0, B = 1) and other conditions (such as C) are met, the inspection result corresponding to the state combination is considered normal; if both devices are standby (A = 0, B = 0), an alarm or abnormal identification will be triggered to prompt the abnormal operation of the device.

[0086] S511, determining an abnormal inspection point from the plurality of inspection points according to the inspection analysis result.

[0087] If the inspection analysis result of the inspection point is normal inspection, a conclusion of "normal inspection" is given.

[0088] If the inspection analysis result of the inspection point is abnormal inspection, a conclusion of "abnormal inspection" is given.

[0089] S512, generating an alarm information according to the disk cabinet label, the abnormal inspection point and the inspection analysis result of the abnormal inspection point.

[0090] In this application, the alarm information is generated according to the disk cabinet label, the abnormal inspection point and the inspection analysis result of the abnormal inspection point, so that the maintenance personnel can know in time to intervene in maintenance in time.

[0091] The application can quickly find problems of equipment through image recognition and automatic analysis, and generate analysis results in time, which helps to identify and handle equipment faults in advance, reduces equipment downtime and maintenance cost, improves reliability, reduces dependence on manual operation in the inspection process, thereby reducing labor cost and avoiding omissions in the manual inspection process.

[0092] Further, it is not difficult to understand that, since the slide rail is installed on the top of the machine room, the acquisition image acquired by the image acquisition device is acquired based on a top-down perspective, and compared with the perspective of the panel of the disk cabinet, the acquisition image will be trapezoidal distortion, therefore, in the application, it is necessary to identify the acquisition image, determine the trapezoidal correction parameter of the acquisition image, then perform trapezoidal to rectangular processing on the inspection image based on the trapezoidal correction parameter, and generate the inspection log based on the inspection image after the trapezoidal to rectangular processing. The inspection log here is mainly to facilitate the maintenance personnel to carry out review and traceability work. In addition, the image in the inspection log can also be used for retraining of the abnormal identification model corresponding to the inspection image.

[0093] Figure 6 is an exemplary schematic diagram of a machine room inspection device based on image recognition technology shown by the application, as Figure 6 shown, the machine room inspection device based on image recognition technology 600 includes an acquisition module 601, an identification module 602, a matching module 603, an inspection module 604 and an analysis module 605, wherein:

[0094] The acquisition module 601 is configured to move the image acquisition device to the inspection position on the slide rail, and control the image acquisition device to acquire images of the panel of the disk cabinet in the machine room based on the preset rotation angle and the preset focal length value.

[0095] The identification module 602 is configured to identify the acquisition image, determine the disk cabinet panel region and the disk cabinet label in the acquisition image, and establish a coordinate system in the disk cabinet panel region.

[0096] The matching module 603 is configured to match a plurality of inspection points located in the disk cabinet panel region, an inspection order of the plurality of inspection points and an inspection parameter of each inspection point from the inspection rule library according to the disk cabinet label, wherein the inspection parameter includes an inspection point region determination parameter, an image acquisition device adjustment parameter and an image acquisition device acquisition parameter.

[0097] The inspection module 604 is configured to sequentially inspect the plurality of inspection points based on the inspection order and the inspection parameter, and obtain the inspection image corresponding to each inspection point.

[0098] The analysis module 605 is configured to identify and analyze the inspection image, and obtain the inspection analysis result corresponding to the inspection point.

[0099] The device can quickly find problems of the equipment through image recognition and automatic analysis, and timely generate analysis results, which helps to identify and handle equipment failures in advance, reduces equipment downtime and maintenance costs, improves reliability, reduces the dependence on manual operation in the inspection process, thereby reducing labor costs and avoiding omissions in the manual inspection process. The device can quickly find problems of the equipment through image recognition and automatic analysis, and timely generate analysis results, which helps to identify and handle equipment failures in advance, reduces equipment downtime and maintenance costs, improves reliability, reduces the dependence on manual operation in the inspection process, thereby reducing labor costs and avoiding omissions in the manual inspection process.

[0100] Further, the inspection module 604 is further configured to: determine a corresponding inspection point area of the inspection point on the coordinate system according to an inspection point area determination parameter corresponding to the inspection point, wherein the inspection point area determination parameter comprises an inspection point coordinate and a coordinate expansion parameter; adjust the image acquisition device according to an image acquisition device adjustment parameter corresponding to the inspection point, wherein the image acquisition device adjustment parameter comprises at least one of a target inspection position, a target rotation angle, and a target focal length value; and acquire an image of the inspection point area according to an image acquisition device acquisition parameter corresponding to the image acquisition device after adjustment, to obtain an inspection image of the inspection point, wherein the image acquisition device acquisition parameter comprises an acquisition frequency and an acquisition quantity.

[0101] Further, the inspection module 604 is further configured to: acquire a current inspection position, a current rotation angle, and a current focal length value of the image acquisition device; acquire an angle difference between the target rotation angle and the current rotation angle; acquire a focal length difference between the target focal length value and the current focal length value; acquire a displacement difference between the target inspection position and the current inspection position; and adjust the image acquisition device based on the angle difference, the focal length difference, and the displacement difference.

[0102] Further, in response to the inspection point being a flashing state indicator light, the acquisition frequency corresponding to the inspection point is greater than half of the flashing frequency of the flashing state indicator light and less than the flashing frequency of the flashing state indicator light, and the acquisition quantity corresponding to the inspection point is greater than or equal to 3.

[0103] Further, the analysis module 605 is further configured to: associate the inspection analysis results corresponding to the inspection points having a master-slave relationship to obtain an inspection analysis result set; and determine the inspection analysis result corresponding to the inspection point having the master-slave relationship according to the inspection analysis results in the inspection analysis result set.

[0104] Further, the machine room inspection device 600 based on image recognition technology further comprises an alarm module, which is configured to: determine an abnormal inspection point from the plurality of inspection points according to the inspection analysis result; and generate an alarm information according to the disk cabinet label, the abnormal inspection point, and the inspection analysis result of the abnormal inspection point.

[0105] Further, the machine room inspection device 600 based on the image recognition technology further comprises a log generation module, configured to identify the collected image, determine the trapezoidal correction parameter of the collected image, perform trapezoidal-to-rectangular processing on the inspection image based on the trapezoidal correction parameter, and generate an inspection log based on the inspection image after the trapezoidal-to-rectangular processing.

[0106] Further, the slide rail is in S-shaped arrangement and is suspendedly installed on the top of the machine room.

[0107] To achieve the above-mentioned embodiments, the embodiments of the present application further propose an electronic device 700, as shown in the figure, the electronic device 700 comprises a processor 701 and a memory 702 in communication connection with the processor, the memory 702 stores instructions executable by at least one processor, and the instructions are executed by the at least one processor 701 to implement the machine room inspection method based on the image recognition technology as shown in the above-mentioned embodiments. Figure 7

[0108] To achieve the above-mentioned embodiments, the embodiments of the present application further propose a non-transitory computer readable storage medium storing computer instructions, wherein the computer instructions are used to make a computer implement the machine room inspection method based on the image recognition technology as shown in the above-mentioned embodiments.

[0109] To achieve the above-mentioned embodiments, the embodiments of the present application further propose a computer program product comprising a computer program, the computer program implements the machine room inspection method based on the image recognition technology as shown in the above-mentioned embodiments when executed by a processor.

[0110] In the description of the present application, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0111] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0112] ​In the description of the specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are contained in at least one embodiment or example of the present application. In the specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples without contradiction.

[0113] Although the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are exemplary and are not to be construed as limiting the present application, and the person skilled in the art can make changes, modifications, replacements and variations to the above-described embodiments within the scope of the present application.

Claims

1. A computer room inspection method based on image recognition technology, characterized in that: include: Moving the image acquisition device to an initial inspection position on the slide rail, and controlling the image acquisition device to acquire an image of a cabinet panel in the computer room based on a preset rotation angle and a preset focal length value to obtain a captured image; Recognizing the acquired image, determining a cabinet panel area and a cabinet label in the acquired image, and establishing a coordinate system in the cabinet panel area; According to the cabinet label, a plurality of inspection points located within the cabinet panel area, an inspection order of the plurality of inspection points, and inspection parameters of each of the inspection points are matched and obtained from an inspection rule library, wherein the inspection parameters include inspection point area determination parameters, image acquisition device adjustment parameters, and image acquisition device acquisition parameters; Based on the inspection sequence and the inspection parameters, the plurality of inspection points are inspected in sequence to obtain an inspection image corresponding to each inspection point; The inspection image is identified and analyzed to obtain the inspection analysis result corresponding to the inspection point.

2. The method according to claim 1, characterized in that When inspecting any one of the multiple inspection points, the method includes: Determining the inspection point area corresponding to the inspection point on the coordinate system according to inspection point area determination parameters corresponding to the inspection point, wherein the inspection point area determination parameters include inspection point coordinates and coordinate extension parameters; Adjusting the image acquisition device according to the image acquisition device adjustment parameters corresponding to the inspection point, wherein the image acquisition device adjustment parameters include at least one of a target inspection position, a target rotation angle, and a target focal length value; Based on the adjusted image acquisition device, image acquisition is performed on the inspection point area according to acquisition parameters of the image acquisition device corresponding to the inspection point to obtain an inspection image of the inspection point, wherein the acquisition parameters of the image acquisition device include acquisition frequency and acquisition quantity.

3. The method according to claim 2, characterized in that The adjusting the image acquisition device according to the image acquisition device adjustment parameters corresponding to the inspection point includes: Obtaining the current inspection position, current rotation angle and current focal length value of the image acquisition device; Obtaining an angle difference between the target rotation angle and the current rotation angle; Obtaining a focal length difference between the target focal length value and the current focal length value; Obtaining a displacement difference between the target inspection position and the current inspection position; The image acquisition device is adjusted based on the angle difference, the focal length difference, and the displacement difference.

4. The method according to claim 3, characterized in that The method further comprises: In response to the inspection point being a flashing indicator light, the collection frequency corresponding to the inspection point is greater than half of the flashing frequency of the flashing indicator light and less than the flashing frequency of the flashing indicator light, and the collection quantity corresponding to the inspection point is greater than or equal to 3.

5. The method according to any one of claims 1 to 4, characterized in that The identifying and analyzing the inspection image to obtain the inspection analysis result corresponding to the inspection point includes: Correlate the inspection analysis results corresponding to the inspection points in the master-slave relationship to obtain an inspection analysis result set; The inspection analysis results corresponding to the inspection points having the master-slave relationship are determined according to the inspection analysis results in the inspection analysis result set.

6. The method according to claim 5, characterized in that After the inspection image is identified and analyzed to obtain the inspection analysis result corresponding to the inspection point, the method further includes: Determining an abnormal inspection point from the plurality of inspection points according to the inspection analysis result; An alarm message is generated according to the cabinet label, the abnormal inspection point, and the inspection analysis result of the abnormal inspection point.

7. The method according to claim 6, characterized in that The method further comprises: Identifying the acquired image and determining a keystone correction parameter of the acquired image; The inspection image is subjected to a trapezoidal-rectangular transformation process based on the trapezoidal correction parameter, and an inspection log is generated based on the inspection image subjected to the trapezoidal-rectangular transformation process.

8. The method according to claim 1, characterized in that The slide rail is arranged in an S-shape and is suspended on the top of the machine room.

9. A machine room inspection device based on image recognition technology, characterized in that: include: An acquisition module is used to move the image acquisition device to an inspection position on the slide rail, and control the image acquisition device to acquire images of the cabinet panels in the computer room based on a preset rotation angle and a preset focal length value to obtain acquired images; an identification module, configured to identify the acquired image, determine a cabinet panel area and a cabinet label in the acquired image, and establish a coordinate system in the cabinet panel area; a matching module, configured to match, from an inspection rule library, a plurality of inspection points located within the panel area of ​​the cabinet, an inspection sequence of the plurality of inspection points, and inspection parameters for each of the inspection points, based on the cabinet label; the inspection parameters including inspection point area determination parameters, image acquisition device adjustment parameters, and image acquisition device acquisition parameters; An inspection module, configured to inspect the plurality of inspection points in sequence based on the inspection sequence and the inspection parameters, and obtain an inspection image corresponding to each inspection point; The analysis module is used to identify and analyze the inspection image and obtain the inspection analysis results corresponding to the inspection points.

10. An electronic device comprising: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

Images