Secondary screen cabinet information acquisition identification and check alarm system based on image identification

Abstract

The invention provides a secondary screen cabinet information acquisition identification and check alarm system based on image identification, and the system comprises an image shooting module which is used for shooting a secondary screen cabinet; the image recognition module is used for carrying out image recognition on the shot image and determining target information; the information checking module is used for checking the target information; and the exception alarm module is used for pushing exception results on the human-computer interaction interface and giving an alarm when the exception is checked out. According to the secondary screen cabinet information acquisition identification and check alarm system based on image identification, information acquisition identification and check alarm are automatically carried out on a secondary screen cabinet, the problem of low manual detection efficiency is solved, the comprehensiveness of system work is improved, intelligent fault early warning and information feedback, instant alarm and information pushing are realized, and the working efficiency is improved. The speed of subsequent response to the cabinet fault is improved, and the efficiency and reliability of equipment management are improved to a great extent.

Description

Technical Field

[0001] The present invention belongs to the technical field of computer data processing, and in particular relates to a secondary screen cabinet information collection, identification and verification alarm system based on image recognition. Background Art

[0002] With the rapid development of industrial automation and intelligent manufacturing, the safety and reliability of secondary cabinets, as key power distribution and control equipment in power systems, have become particularly important. Traditional secondary cabinet inspection methods mainly rely on manual inspections and visual observations. Although they can ensure the normal operation of equipment to a certain extent, they have the following defects:

[0003] Low efficiency of manual inspection: Manual inspection is not only time-consuming, but also limited by the professional level and experience of maintenance personnel, which can easily lead to missed inspections or false inspections, and cannot ensure a comprehensive understanding of the equipment status.

[0004] Lack of intelligent processing: Existing technologies are relatively simple in terms of data processing and information verification, and are often unable to automatically identify and analyze the operating status of equipment, making it difficult to achieve intelligent fault warning and information feedback.

[0005] Imperfect alarm mechanism: When abnormal situations occur, traditional methods rely more on manual recording and feedback, which cannot achieve instant alarms and information push, affecting the response speed to potential faults.

[0006] Therefore, a new technical solution is urgently needed to combine image capture, recognition and intelligent verification technologies to achieve automated monitoring and abnormal alarms for secondary screen cabinets, thereby improving the efficiency and reliability of equipment management. Summary of the invention

[0007] In order to solve the above problems, the present invention provides a secondary screen cabinet information collection, identification and verification alarm system based on image recognition, which automatically collects information, identifies and verifies alarms for secondary screen cabinets, overcomes the problem of low efficiency of manual detection, improves the comprehensiveness of system work, realizes intelligent fault warning and information feedback, instant alarm and information push, improves the speed of subsequent response to cabinet failures, and greatly improves the efficiency and reliability of equipment management.

[0008] An embodiment of the present invention provides a secondary screen cabinet information collection, identification and verification alarm system based on image recognition, including:

[0009] An image shooting module is used to shoot the secondary screen cabinet;

[0010] An image recognition module is used to perform image recognition on the captured image and determine target information;

[0011] An information checking module, used for checking target information;

[0012] The abnormal alarm module is used to push the abnormal results on the human-computer interaction interface and issue an alarm when an abnormality is found.

[0013] Optionally, the image capturing module captures the secondary screen cabinet, including:

[0014] The inspection robot takes photos of the secondary screen cabinets;

[0015] and / or,

[0016] Inspection personnel use handheld mobile terminals to take photos of the secondary screen cabinets.

[0017] Optionally, the target information includes at least: the insertion and withdrawal positions of the pressure plate in the secondary panel cabinet and the indication status of the signal light.

[0018] Optionally, the information checking module checks the target information, including:

[0019] For periodic panel cabinet verification work, the target information is verified based on the normal operating status table data of the secondary panel cabinet preset in the database;

[0020] For the secondary panel cabinet verification work before startup after maintenance, select the panel cabinet involved in the work in the status verification system, automatically generate a secondary panel cabinet status inspection ticket, and compare it with the data uploaded by the perception layer.

[0021] Optionally, the secondary screen cabinet information collection, identification and verification alarm system based on image recognition also includes:

[0022] The result push module is used to push normal results on the human-computer interaction interface when there is no abnormality in the verification.

[0023] Optionally, the secondary screen cabinet information collection, identification and verification alarm system based on image recognition also includes:

[0024] Supplementary work modules for:

[0025] Conduct three-dimensional construction of the site where the secondary screen cabinet is located to obtain a three-dimensional model of the site;

[0026] Mapping the information to be mapped into the on-site three-dimensional model to obtain an on-site status model; wherein the information to be mapped includes at least: historically determined target information and historically verified anomalies;

[0027] Identify multiple supplementary information collection requirements in the field status model;

[0028] Based on the needs of various supplementary information collection, a collection baseline is constructed in the on-site status model and multiple path ranges and their corresponding collection postures are planned on the collection baseline;

[0029] Provides close guidance to the target within any range;

[0030] When the photographed target enters the i-th passing range, based on the posture relationship between the collection posture of the i-th passing range and the collection posture of the i±1-th passing range, the guiding collection posture is determined according to the collection posture of the i-th passing range and the collection posture of the i±1-th passing range;

[0031] Guiding the shooting target to shoot in a guided acquisition posture to obtain a new shot image; wherein the shooting target includes at least: an inspection robot and an inspection personnel;

[0032] Perform additional image recognition and information verification alarms on newly captured images.

[0033] Optionally, the supplementary work module identifies multiple supplementary information collection requirements in the field status model, including:

[0034] Extracting features from the on-site state model to obtain a first feature vector;

[0035] Determine a target demand search rule corresponding to the first feature vector from a demand search rule library; wherein the second feature vector corresponding to the target demand search rule matches the first feature vector;

[0036] Based on the target demand search rules, multiple supplementary information collection demands are searched out in the field status model.

[0037] Optionally, the supplementary work module constructs a collection baseline in the field status model based on various supplementary information collection requirements and plans multiple path ranges on the collection baseline and their corresponding collection postures, including:

[0038] Based on various supplementary information collection requirements, multiple areas to be photographed and corresponding information to be photographed are determined;

[0039] Traverse each area to be photographed in turn;

[0040] Each time the traversal is performed, the model area corresponding to the traversed area to be photographed is determined from the field status model, and the information display direction of the information to be photographed corresponding to the traversed area to be photographed is determined from the secondary screen cabinet configuration library;

[0041] Based on the model area and information display direction, the surface to be photographed is set in the on-site status model;

[0042] After traversing each area to be photographed, based on the conditions for clustering the faces to be photographed, multiple face clusters to be photographed and their corresponding minimum enclosing spheres are determined in the scene state model according to the faces to be photographed set during each traversal; wherein the minimum enclosing sphere encloses all faces to be photographed in the same face cluster to be photographed at the minimum;

[0043] Draw a collection baseline in the scene state model; wherein the lengths of multiple perpendicular lines drawn from the first sphere center of each minimum enclosing sphere corresponding to each to-be-photographed face cluster to the collection baseline are all less than or equal to a length threshold;

[0044] Taking the perpendicular points of each perpendicular line as the second sphere center and the preset radius length as the radius, a plurality of spherical areas are respectively made;

[0045] Each spherical area is used as a plurality of path ranges; wherein the path range corresponds to the smallest enclosing sphere corresponding to the first sphere center on the perpendicular line where the second sphere center is located;

[0046] Determine the acquisition posture corresponding to each passing range; wherein the acquisition posture is the posture of completely photographing all the to-be-photographed surfaces in the smallest enclosing sphere corresponding to the passing range within the passing range;

[0047] The clustering conditions of the to-be-photographed faces include:

[0048] The straight-line distance between any two faces to be photographed in the same face cluster to be photographed is less than or equal to the distance threshold;

[0049] and,

[0050] The spatial angle between the center line between any two to-be-photographed surfaces in the same to-be-photographed surface cluster and the aisle line where the to-be-photographed surfaces are located falls within the standard angle range.

[0051] Optionally, the supplementary working module determines the guiding collection posture based on the posture relationship between the collection posture of the ith passing range and the collection posture of the i±1th passing range, according to the collection posture of the ith passing range and the collection posture of the i±1th passing range, including:

[0052] Determine a guided acquisition posture determination rule corresponding to the posture relationship from a guided acquisition posture determination rule library;

[0053] Based on the guiding collection posture determination rule, the guiding collection posture is determined according to the collection posture of the i-th passing range and the collection posture of the i±1-th passing range.

[0054] Compared with the prior art, the beneficial effects are: the secondary screen cabinet information collection, identification and verification alarm system automatically performs information collection, identification and verification alarm on the secondary screen cabinet, overcoming the problem of low efficiency of manual detection, improving the comprehensiveness of system work, and realizing intelligent fault warning and information feedback, instant alarm and information push, which improves the subsequent response speed to cabinet failures and greatly improves the efficiency and reliability of equipment management. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a structural schematic diagram of a secondary screen cabinet information collection, identification and verification alarm system based on image recognition in the present invention. DETAILED DESCRIPTION

[0056] The specific implementation method of the secondary screen cabinet information collection, identification and verification alarm system based on image recognition of the present invention is described in detail below with reference to the accompanying drawings.

[0057] The embodiment of the present invention provides a secondary screen cabinet information collection, identification and verification alarm system based on image recognition, such as Figure 1 As shown, including:

[0058] Image shooting module 1, used for shooting the secondary screen cabinet;

[0059] Image recognition module 2, used to perform image recognition on the captured image and determine target information;

[0060] Information checking module 3, used to check the target information;

[0061] The abnormal alarm module 4 is used to push the abnormal result on the human-computer interaction interface and issue an alarm when an abnormality is found.

[0062] In the above technical scheme, based on the image recognition technology, the image of the secondary screen cabinet is identified to determine the target information; the target information may be the cabinet status information displayed on the display screen on the secondary screen cabinet, etc.; after determining the target information, the target information is checked to determine whether the secondary screen cabinet has an abnormal state. If so, the abnormal result is pushed on the human-computer interaction interface and an alarm is issued.

[0063] The present invention automatically collects, identifies, verifies and alarms information on secondary screen cabinets, overcomes the problem of low efficiency of manual detection, improves the comprehensiveness of system work, realizes intelligent fault warning and information feedback, real-time alarm and information push, improves the speed of subsequent response to cabinet failures, and greatly improves the efficiency and reliability of equipment management.

[0064] In one embodiment, the image capturing module captures the secondary screen cabinet, including:

[0065] The inspection robot takes photos of the secondary screen cabinets;

[0066] and / or,

[0067] Inspection personnel use handheld mobile terminals to take photos of the secondary screen cabinets.

[0068] In the above technical solution, the inspection robot is equipped with a high-resolution camera, which can automatically cruise and take real-time images of the secondary cabinet. At the same time, the inspection personnel use a handheld mobile terminal to take pictures, and the terminal is connected to the central control system in real time. The captured images will be transmitted to the image recognition module to determine the target information.

[0069] In one embodiment, the target information includes at least: the insertion and withdrawal positions of the pressure plates in the secondary panel cabinet and the indication status of the signal lights.

[0070] In the above technical solution, the image recognition module uses an image processing algorithm to analyze the position of the pressure plate to determine whether it is in a delivery state or a retraction state; and recognizes the color and state of the signal light (such as red light, green light, flashing, etc.).

[0071] In one embodiment, the information checking module checks the target information, including:

[0072] For periodic panel cabinet verification work, the target information is verified based on the normal operating status table data of the secondary panel cabinet preset in the database;

[0073] For the secondary panel cabinet verification work before startup after maintenance, select the panel cabinet involved in the work in the status verification system, automatically generate a secondary panel cabinet status inspection ticket, and compare it with the data uploaded by the perception layer.

[0074] In the above technical solution, for periodic screen cabinet verification work, the system automatically extracts the information of relevant secondary screen cabinets based on the preset normal operating status table data. The system compares the target information obtained through real-time monitoring with the standard data in the database to check whether the pressure plate position and signal light status are as expected. Any inconsistency will be marked as abnormal, and a report will be automatically generated for relevant personnel to handle; for example: by comparing the actual position of the pressure plate in the image with the preset standard position, it is determined whether it is in the delivery state or the return state. If it is found that the pressure plate is not delivered or returned as expected, an alarm will be issued.

[0075] For the secondary panel cabinet verification work before startup after maintenance, after the maintenance is completed, the operator selects the panel cabinet that needs to be checked in the status verification system; the system automatically generates a secondary panel cabinet status inspection ticket, listing the items that need to be checked; the standard data in the inspection ticket is compared with the real-time data uploaded by the perception layer to ensure that all indicators meet the normal operating standards; if any problems are found, the system will prompt and record them to ensure the safe startup of the equipment.

[0076] In one embodiment, the secondary screen cabinet information collection, identification and verification alarm system based on image recognition also includes:

[0077] The result push module is used to push normal results on the human-computer interaction interface when there is no abnormality in the verification.

[0078] In one embodiment, the secondary screen cabinet information collection, identification and verification alarm system based on image recognition also includes:

[0079] Supplementary work modules for:

[0080] S01. Perform three-dimensional construction on the site where the secondary screen cabinet is located to obtain a three-dimensional model of the site;

[0081] In step S01, when the site is three-dimensionally constructed, three-dimensional modeling of the site can be performed based on three-dimensional construction technology (such as BIM, GIS, etc.) to obtain a three-dimensional model of the site;

[0082] S02, mapping the information to be mapped into the on-site three-dimensional model to obtain an on-site status model; wherein the information to be mapped includes at least: target information determined in history and anomalies verified in history;

[0083] In step S02, when mapping the information to be mapped, the associated position of the information to be mapped is determined (for example, if the information to be mapped is an abnormal alarm record in the history of a cabinet, then its associated position is the cabinet), and the information to be mapped is set next to the associated position;

[0084] S03, identifying multiple supplementary information collection requirements in the field status model;

[0085] In step S03, the need for collecting supplementary information is the need to collect supplementary information from the secondary cabinet in order to further analyze the alarm anomaly, track the progress of the alarm anomaly processing, etc.;

[0086] S04. Based on various supplementary information collection requirements, a collection baseline is constructed in the on-site status model and multiple path ranges on the collection baseline and their corresponding collection postures are planned;

[0087] In step S04, the collection posture corresponds to the passing range one by one. When the shooting target enters the passing range, it is appropriate to adopt the corresponding collection posture to collect information from the secondary screen cabinet;

[0088] S05, guiding the shooting target to a position close to any passing range;

[0089] In step S05, when performing the nearby guidance, first determine the actual position of the photographed target in the scene, then determine the closest passing range to the actual position, and guide the photographed target to enter the passing range;

[0090] S06. When the photographed target enters the i-th passing range, based on the posture relationship between the collection posture of the i-th passing range and the collection posture of the i±1-th passing range, the guiding collection posture is determined according to the collection posture of the i-th passing range and the collection posture of the i±1-th passing range;

[0091] In step S06, i=1, 2, 3, ..., N; N is the total number of the passing ranges; the posture relationship can be the posture similarity between two acquisition postures, etc.; based on the posture relationship, according to the acquisition posture of the i-th passing range and the acquisition posture of the i±1-th passing range, the guiding acquisition posture is determined; when shooting with the guiding acquisition posture, it is convenient for the shooting target to switch the posture for shooting the adjacent passing range;

[0092] S07, guiding the shooting target to shoot in a guided acquisition posture to obtain a new shot image; wherein the shooting target at least includes: an inspection robot and an inspection personnel;

[0093] S08: Perform additional image recognition and information verification and alarm on the newly captured image.

[0094] In step S08, image recognition is performed again on the newly captured image to determine new target information and then checked again. If an abnormality is found during the check, an alarm is issued.

[0095] Generally, there is often a need for supplementary information collection at the site of the secondary screen cabinet. When it occurs, the staff needs to arrange the supplementary information collection by themselves. However, the supplementary information collection is mostly decentralized, and the staff needs to find the collection locations and arrange them one by one, which is very cumbersome and reduces the efficiency of supplementary information collection. Most of the supplementary information collection needs are urgent, which creates contradictions. The embodiments of the present invention can solve these problems, construct a collection baseline and plan multiple path ranges and their corresponding collection postures on the collection baseline, based on which the shooting target is quickly guided to collect supplementary information, without the need for staff to arrange by themselves, which improves convenience and efficiency of supplementary information collection, and is particularly suitable for supplementary information collection needs.

[0096] In one embodiment, the supplementary work module identifies multiple supplementary information collection requirements in the field status model, including:

[0097] S031, extracting features from the on-site state model to obtain a first feature vector;

[0098] In step S031, when extracting features, the content features in the scene state model are advanced to construct a first feature vector in vector form;

[0099] S032, determining a target demand search rule corresponding to the first feature vector from a demand search rule library; wherein a second feature vector corresponding to the target demand search rule matches the first feature vector;

[0100] In step S032, the target demand search rule corresponds to the second feature vector. When the first feature vector matches the second feature vector, it means that the target demand search rule is suitable for searching for the supplementary information collection requirements for the on-site status model. For example, if the content feature of the second feature vector is that a secondary screen cabinet frequently fails and is in the fault handling stage, the corresponding target demand search rule is to search for the location of the secondary screen cabinet and generate a supplementary information collection requirement for fault handling tracking at the location.

[0101] S033. Based on the target demand search rules, multiple supplementary information collection requirements are searched in the site status model.

[0102] In one embodiment, the supplementary work module constructs a collection baseline in the on-site state model based on various supplementary information collection requirements and plans multiple path ranges on the collection baseline and their corresponding collection postures, including:

[0103] S041. Determine multiple areas to be photographed and corresponding information to be photographed based on various supplementary information collection requirements;

[0104] In step S041, the supplementary information collection requirement reflects which secondary screen cabinets need to be supplemented with information and what information needs to be supplemented, that is, as a plurality of areas to be photographed and corresponding information to be photographed;

[0105] S042, traversing each area to be photographed in sequence;

[0106] S043, each time during traversal, determining the model area corresponding to the traversed area to be photographed from the site status model, and determining the information display direction of the information to be photographed corresponding to the traversed area to be photographed from the secondary screen cabinet configuration library;

[0107] In step S043, the information to be photographed will have an information display direction, which is related to the direction of the display screen of the secondary screen cabinet where it is located;

[0108] S044. Based on the model area and the information display direction, a surface to be photographed is set in the scene state model;

[0109] In step S044, when setting the surface to be photographed, the facing direction of the surface to be photographed is consistent with the information display direction, and the intersection line between the surface to be photographed and the plane of the model area is parallel to at least one side of the plane of the model area; when the shooting target shoots toward the surface to be photographed, the information to be photographed can be photographed at the best angle;

[0110] S045, after traversing each area to be photographed, based on the to-be-photographed face clustering condition, according to the to-be-photographed faces set in each traversal, multiple to-be-photographed face clusters and their corresponding minimum enclosing spheres are determined in the scene state model; wherein the minimum enclosing spheres minimumly enclose all to-be-photographed faces in the same to-be-photographed face cluster;

[0111] S046, drawing a collection baseline in the scene state model; wherein the lengths of multiple perpendicular lines drawn from the first sphere center of each minimum enclosing sphere corresponding to each to-be-photographed face cluster to the collection baseline are all less than or equal to a length threshold;

[0112] In step S046, the length threshold may be 50 centimeters;

[0113] S047, taking the vertical point of each vertical line as the second sphere center and using the preset radius length as the radius to respectively create a plurality of spherical areas;

[0114] S048, taking each spherical area as a plurality of path ranges; wherein the path ranges correspond one to one with the smallest enclosing sphere corresponding to the first sphere center on the perpendicular line where the second sphere center is located;

[0115] S049, determining the collection posture corresponding to each passing range; wherein the collection posture is a posture for completely photographing all the to-be-photographed surfaces in the smallest enclosing sphere corresponding to the passing range within the passing range;

[0116] The clustering conditions of the to-be-photographed faces include:

[0117] The straight-line distance between any two to-be-photographed faces in the same to-be-photographed face cluster is less than or equal to a distance threshold; the distance threshold may be 30 centimeters;

[0118] and,

[0119] The spatial angle between the center line between any two faces to be photographed in the same face cluster to be photographed and the aisle line where the faces to be photographed are located falls into the standard angle range. The standard angle range can be, for example, 30 degrees to 80 degrees; the aisle line where the face to be photographed is located refers to the aisle line in front of the cabinet unit where the secondary screen cabinet corresponding to the face to be photographed is located where the information to be photographed is located (generally, the secondary screen cabinets are arranged in multiple columns, and aisles are set between two adjacent columns of secondary screen cabinets for inspection, and the route of the aisle is the aisle line).

[0120] In the embodiment of the present invention, when constructing a collection baseline in a scene state model and planning multiple path ranges on the collection baseline and their corresponding collection postures, the surface to be photographed is first determined, then the surface cluster to be photographed is determined, and finally the minimum enclosing sphere is introduced. Based on the minimum enclosing sphere, the collection baseline is drawn and multiple path ranges and their corresponding collection postures are planned, which greatly improves the work efficiency of the system and improves the accuracy of the determination of the collection baseline, etc.; secondly, the condition of clustering the surface to be photographed is introduced to improve the rationality of the determination of the surface cluster to be photographed.

[0121] In one embodiment, the supplementary working module determines the guiding collection posture based on the posture relationship between the collection posture of the i-th passing range and the collection posture of the i±1-th passing range, according to the collection posture of the i-th passing range and the collection posture of the i±1-th passing range, including:

[0122] S061, determining a guidance collection posture determination rule corresponding to the posture relationship from a guidance collection posture determination rule library;

[0123] S062. Based on the guiding collection posture determination rule, determine the guiding collection posture according to the collection posture of the i-th passing range and the collection posture of the i±1-th passing range.

[0124] There are guiding collection posture determination rules corresponding to different posture relationships in the guiding collection posture determination rule library. The guiding collection posture determination rules are rules for guiding collection postures represented by posture relationships. For example: if the posture relationship is that the posture similarity between the collection posture of the i-th passing range and the collection posture of the i±1-th passing range is 90%, then the guiding collection posture determination rule is to use the collection posture of the i-th passing range as the guiding collection posture (the two have a high similarity, and the posture switching is more convenient, and the collection posture of the i-th passing range is directly used as the guiding collection posture).

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not intended to limit the present invention. A person skilled in the art should understand that the specific implementation of the present invention can be modified or replaced by equivalents, but these modifications or changes are within the scope of protection of the pending claims.

Claims

1. A secondary screen cabinet information collection, identification and verification alarm system based on image recognition, characterized in that: The secondary screen cabinet information collection, identification and verification alarm system includes: An image shooting module is used to shoot the secondary screen cabinet; An image recognition module is used to perform image recognition on the captured image and determine target information; An information checking module, used for checking target information; The abnormal alarm module is used to push the abnormal results on the human-computer interaction interface and issue an alarm when an abnormality is found.

2. According to the image recognition-based secondary cabinet information collection, identification and verification alarm system of claim 1, it is characterized in that: The image shooting module shoots the secondary screen cabinet, including: The inspection robot takes photos of the secondary screen cabinets; and / or, Inspection personnel use handheld mobile terminals to take photos of the secondary screen cabinets.

3. According to the image recognition-based secondary cabinet information collection, identification and verification alarm system of claim 1, it is characterized in that: The target information includes at least: the insertion and withdrawal positions of the pressure plates in the secondary panel cabinet and the indication status of the signal lights.

4. According to the image recognition-based secondary cabinet information collection, identification and verification alarm system of claim 1, it is characterized in that: The information checking module checks the target information, including: For periodic panel cabinet verification work, the target information is verified based on the normal operating status table data of the secondary panel cabinet preset in the database; For the secondary panel cabinet verification work before startup after maintenance, select the panel cabinet involved in the work in the status verification system, automatically generate a secondary panel cabinet status inspection ticket, and compare it with the data uploaded by the perception layer.

5. According to the image recognition-based secondary cabinet information collection, identification and verification alarm system of claim 1, it is characterized in that: The secondary screen cabinet information collection, identification and verification alarm system also includes: The result push module is used to push normal results on the human-computer interaction interface when there is no abnormality in the verification.

6. According to the image recognition-based secondary cabinet information collection, identification and verification alarm system of claim 1, it is characterized in that: The secondary screen cabinet information collection, identification and verification alarm system also includes: Supplementary work modules for: Conduct three-dimensional construction of the site where the secondary screen cabinet is located to obtain a three-dimensional model of the site; Mapping the information to be mapped into the on-site three-dimensional model to obtain an on-site status model; wherein the information to be mapped includes at least: historically determined target information and historically verified anomalies; Identify multiple supplementary information collection requirements in the field status model; Based on the needs of various supplementary information collection, a collection baseline is constructed in the on-site status model and multiple path ranges and their corresponding collection postures are planned on the collection baseline; Provides close guidance to the target within any range; When the photographed target enters the i-th passing range, based on the posture relationship between the collection posture of the i-th passing range and the collection posture of the i±1-th passing range, the guiding collection posture is determined according to the collection posture of the i-th passing range and the collection posture of the i±1-th passing range; Guiding the shooting target to shoot in a guided acquisition posture to obtain a new shot image; wherein the shooting target includes at least: an inspection robot and an inspection personnel; Perform additional image recognition and information verification alarms on newly captured images.

7. According to the image recognition-based secondary cabinet information collection, identification and verification alarm system of claim 6, it is characterized in that: The supplementary work module identifies multiple supplementary information collection requirements in the field status model, including: Extracting features from the on-site state model to obtain a first feature vector; Determine a target demand search rule corresponding to the first feature vector from a demand search rule library; wherein the second feature vector corresponding to the target demand search rule matches the first feature vector; Based on the target demand search rules, multiple supplementary information collection demands are searched out in the field status model.

8. According to the image recognition-based secondary cabinet information collection, identification and verification alarm system of claim 6, it is characterized in that: The supplementary work module constructs a collection baseline in the field status model based on various supplementary information collection requirements and plans multiple path ranges on the collection baseline and their corresponding collection postures, including: Based on various supplementary information collection requirements, multiple areas to be photographed and corresponding information to be photographed are determined; Traverse each area to be photographed in turn; Each time the traversal is performed, the model area corresponding to the traversed area to be photographed is determined from the field status model, and the information display direction of the information to be photographed corresponding to the traversed area to be photographed is determined from the secondary screen cabinet configuration library; Based on the model area and information display direction, the surface to be photographed is set in the on-site status model; After traversing each area to be photographed, based on the conditions for clustering the faces to be photographed, multiple face clusters to be photographed and their corresponding minimum enclosing spheres are determined in the scene state model according to the faces to be photographed set during each traversal; wherein the minimum enclosing sphere encloses all faces to be photographed in the same face cluster to be photographed at the minimum; Draw a collection baseline in the scene state model; wherein the lengths of multiple perpendicular lines drawn from the first sphere center of each minimum enclosing sphere corresponding to each to-be-photographed face cluster to the collection baseline are all less than or equal to a length threshold; Taking the perpendicular points of each perpendicular line as the second sphere center and the preset radius length as the radius, a plurality of spherical areas are respectively made; Each spherical area is used as a plurality of path ranges; wherein the path range corresponds to the smallest enclosing sphere corresponding to the first sphere center on the perpendicular line where the second sphere center is located; Determine the acquisition posture corresponding to each passing range; wherein the acquisition posture is the posture of completely photographing all the to-be-photographed surfaces in the smallest enclosing sphere corresponding to the passing range within the passing range; The clustering conditions of the to-be-photographed faces include: The straight-line distance between any two faces to be photographed in the same face cluster to be photographed is less than or equal to the distance threshold; and, The spatial angle between the center line between any two to-be-photographed surfaces in the same to-be-photographed surface cluster and the aisle line where the to-be-photographed surfaces are located falls within the standard angle range.

9. According to the image recognition-based secondary cabinet information collection, identification and verification alarm system of claim 6, it is characterized in that: The supplementary working module determines the guiding collection posture based on the posture relationship between the collection posture of the i-th passing range and the collection posture of the i±1-th passing range, according to the collection posture of the i-th passing range and the collection posture of the i±1-th passing range, including: Determine a guided acquisition posture determination rule corresponding to the posture relationship from a guided acquisition posture determination rule library; Based on the guiding collection posture determination rule, the guiding collection posture is determined according to the collection posture of the i-th passing range and the collection posture of the i±1-th passing range.

Images