An intelligent cloud box inspection system based on edge computing

Abstract

The present invention provides an intelligent cloud box inspection system based on edge computing, including an inspection planning module, an inspection control module, an image analysis module, and an inspection adjustment module. The inspection planning module is used to plan a basic inspection route. The inspection control module is used to control the shooting of the lens to meet the inspection requirements. The image analysis module is used to analyze and process the images collected during the inspection. The inspection adjustment module adjusts the inspection process based on the analysis results. During the inspection process, this system will make dynamic adjustments based on the basic inspection route and can obtain more detailed inspection information.

Description

Technical Field

[0001] The present invention relates to the field of electric digital data processing, and particularly to an intelligent cloud box inspection system based on edge computing. Background Art

[0002] Inspection systems are widely used in equipment maintenance, fault detection, safety management, etc. Traditional inspection methods usually rely on manual operations. Inspection personnel regularly check and record equipment. Although this method is intuitive, it has problems such as low efficiency, poor accuracy, and missed inspections caused by human factors. Therefore, setting up network cameras near the equipment for inspection can effectively improve the inspection efficiency. At the same time, a corresponding inspection system is needed to control the inspection process to better play the inspection effect.

[0003] The foregoing discussion of the background art is only intended to facilitate the understanding of the present invention. This discussion does not recognize or admit that any of the materials mentioned is part of the common general knowledge.

[0004] Now, many inspection systems have been developed. After a large number of searches and references, it is found that existing inspection systems are like the system disclosed in CN114279572B. These system methods generally include: an infrared thermal imaging camera; the infrared thermal imaging camera is arranged on a terminal platform capable of moving along a preset guide rail; the preset guide rail is arranged around a plurality of electrical cabinets that need to be inspected; a lifting device for lifting the infrared thermal imaging camera to a height corresponding to the infrared transmission window of each electrical cabinet is arranged on the terminal platform; both the terminal platform and the infrared thermal imaging camera are connected to the monitoring cloud platform and work according to the instructions of the monitoring cloud platform. However, this system can only perform inspections according to the preset guide rail during the inspection process and cannot cope with changing situations, and the inspection effect needs to be improved. Summary of the Invention

[0005] The object of the present invention is to propose an intelligent cloud box inspection system based on edge computing for the existing deficiencies.

[0006] The present invention adopts the following technical solutions:

[0007] An intelligent cloud box inspection system based on edge computing includes an inspection planning module, an inspection control module, an image analysis module, and an inspection adjustment module;

[0008] The inspection planning module is used to plan a basic inspection route. The inspection control module is used to control the shooting of the lens to meet the inspection requirements. The image analysis module is used to analyze and process the images collected during the inspection process. The inspection adjustment module adjusts the inspection process based on the analysis results;

[0009] The inspection planning module includes a node marking unit, a node connection unit, a parameter formulation unit, and a resource release unit. The node marking unit is used to mark the shooting nodes including the intelligent cloud box hardware devices. The node connection unit is used to connect the shooting nodes to form a closed route. The parameter formulation unit extracts inspection parameters based on the closed route. The resource release unit is used to release the computing resources used after completing the inspection route;

[0010] The inspection control module includes an inspection route storage unit and an inspection execution unit. The inspection route storage unit is used to save the inspection parameters. The inspection execution unit is used to obtain the inspection parameters and control the lens to execute the inspection process;

[0011] The image analysis module includes an anomaly analysis unit and an anomaly location unit. The anomaly analysis unit is used to analyze the abnormal phenomena in the images. The anomaly location unit is used to locate the abnormal positions;

[0012] The inspection adjustment module includes an ad-hoc inspection planning unit and an insertion interaction unit. The ad-hoc inspection planning unit is used to plan the ad-hoc added inspection parameters. The insertion interaction unit is used to insert the new inspection parameters into the basic inspection route;

[0013] Furthermore, the node connection unit includes a conversion calculation processor, a node sorting processor, and a sorting adjustment processor. The conversion calculation processor is used to calculate the conversion time between two nodes. The node sorting processor is used to sort the shooting nodes in a circular sequence. The sorting adjustment processor is used to adjust the sorting of the shooting nodes in the circular sequence;

[0014] Furthermore, the processing process of the node connection unit for the shooting nodes includes the following steps:

[0015] S1. The node sorting processor sorts the shooting nodes in any order to form a circular sequence;

[0016] S2. The sorting adjustment processor selects two adjacent shooting nodes as target nodes;

[0017] S3. The conversion calculation processor calculates the change value of the inspection cycle after the target nodes are swapped;

[0018] S4. If the change value of the inspection cycle is negative, the sorting adjustment processor swaps the target nodes, clears the cumulative value, and returns to step S2. If the change value of the inspection cycle is positive, it enters step S5;

[0019] S5. Add 1 to the cumulative value. If the cumulative value is n, send the shooting node information in the circular sequence to the parameter formulation unit. If the cumulative value is less than n, return to step S2, where n is the number of shooting nodes;

[0020] Furthermore, the conversion calculation processor calculates the conversion time Tc between two adjacent shooting nodes in the ring sequence according to the following formula:

[0021]

[0022] Among them, (α1, β1) is the state parameter of the previous shooting node in the adjacent shooting nodes, (α2, β2) is the state parameter of the next shooting node in the adjacent shooting nodes, and ω0 is the standard angular velocity;

[0023] The conversion calculation processor calculates the inspection period T of the entire ring sequence according to the following formula:

[0024]

[0025] Where n is the number of shooting nodes, Tc(i, i+1) represents the conversion time between the i-th shooting node and the i+1-th shooting node in the ring sequence, and Tc(n, 1) represents the conversion time between the n-th shooting node and the 1st shooting node in the ring sequence;

[0026] Further, the abnormality locating unit includes a region setting processor and a relative conversion processor, the region setting processor is used to set the position of the abnormal region in the image, and the relative conversion processor is used to convert the abnormal region position information into relative information with respect to the center of the image;

[0027] The abnormal area position information is represented by four rectangular positioning points, and the relative conversion processor calculates the relative information according to the following formula:

[0028]

[0029] Wherein, x, y are the pixel length and width of the captured image, a, b are the pixel length and width of the abnormal area, x1 and x2 are the two horizontal coordinates of the positioning point, y1 and y2 are the two vertical coordinates of the positioning point, η is the magnification factor, and θ is the alignment angle.

[0030] The beneficial effects achieved by the present invention are:

[0031] When planning basic inspection routes, this system can effectively improve inspection efficiency by changing the connection order of shooting nodes and continuously reducing the inspection cycle of basic inspection routes. By discovering abnormal information during the basic inspection process and inserting temporary inspections to obtain detailed information on the abnormalities, the inspection effect can be improved while ensuring inspection efficiency.

[0032] To enable a further understanding of the features and technical content of the present invention, please refer to the following detailed description and drawings of the present invention. However, the provided drawings are only for reference and illustration, and are not intended to limit the present invention. Description of the Drawings

[0033] Figure 1 It is a schematic diagram of the overall structural framework of the present invention;

[0034] Figure 2 It is a schematic diagram of the composition of the patrol inspection planning module of the present invention;

[0035] Figure 3 It is a schematic diagram of the composition of the node marking unit of the present invention;

[0036] Figure 4 It is a schematic diagram of the composition of the node connection unit of the present invention;

[0037] Figure 5 It is a schematic diagram of the composition of the parameter formulation unit of the present invention. Detailed Embodiment

[0038] The following are specific embodiments to illustrate the implementation manners of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of the present invention. Additionally, the drawings of the present invention are only for simple schematic illustration and are not drawn according to actual dimensions, hereby declared. The following embodiments will further detail the related technical content of the present invention, but the disclosed content is not intended to limit the protection scope of the present invention.

[0039] Embodiment 1.

[0040] This embodiment provides an edge computing-based intelligent cloud box patrol inspection system, combined with Figure 1 , including a patrol inspection planning module, a patrol inspection control module, an image analysis module, and a patrol inspection adjustment module;

[0041] The patrol inspection planning module is used to plan a basic patrol inspection route, the patrol inspection control module is used to control the shooting of the lens to meet the patrol inspection requirements, the image analysis module is used to analyze and process the images collected during the patrol inspection, and the patrol inspection adjustment module adjusts the patrol inspection process based on the analysis results;

[0042] The inspection planning module includes a node marking unit, a node connection unit, a parameter formulation unit, and a resource release unit. The node marking unit is used to mark the shooting nodes including the intelligent cloud box hardware devices. The node connection unit is used to connect the shooting nodes to form a closed route. The parameter formulation unit extracts inspection parameters based on the closed route. The resource release unit is used to release the computing resources used after completing the inspection route;

[0043] The inspection control module includes an inspection route storage unit and an inspection execution unit. The inspection route storage unit is used to save the inspection parameters. The inspection execution unit is used to obtain the inspection parameters and control the lens to execute the inspection process;

[0044] The image analysis module includes an anomaly analysis unit and an anomaly location unit. The anomaly analysis unit is used to analyze the abnormal phenomena in the image. The anomaly location unit is used to locate the abnormal positions;

[0045] The inspection adjustment module includes an ad-hoc inspection planning unit and an insertion interaction unit. The ad-hoc inspection planning unit is used to plan the ad-hoc added inspection parameters. The insertion interaction unit is used to insert the new inspection parameters into the basic inspection route;

[0046] The node connection unit includes a conversion calculation processor, a node sorting processor, and a sorting adjustment processor. The conversion calculation processor is used to calculate the conversion time between two nodes. The node sorting processor is used to sort the shooting nodes in a circular sequence. The sorting adjustment processor is used to adjust the sorting of the shooting nodes in the circular sequence;

[0047] The processing process of the node connection unit for the shooting nodes includes the following steps:

[0048] S1. The node sorting processor sorts the shooting nodes in any order to form a circular sequence;

[0049] S2. The sorting adjustment processor selects two adjacent shooting nodes as target nodes;

[0050] S3. The conversion calculation processor calculates the change value of the inspection cycle after the target nodes are swapped;

[0051] S4. If the change value of the inspection cycle is negative, the sorting adjustment processor swaps the target nodes, clears the cumulative value, and returns to step S2. If the change value of the inspection cycle is positive, it enters step S5;

[0052] S5. Add 1 to the cumulative value. If the cumulative value is n, send the shooting node information in the circular sequence to the parameter formulation unit. If the cumulative value is less than n, return to step S2, where n is the number of shooting nodes;

[0053] The conversion calculation processor calculates the conversion time Tc between two adjacent shooting nodes in the ring sequence according to the following formula:

[0054]

[0055] Among them, (α1, β1) is the state parameter of the previous shooting node in the adjacent shooting nodes, (α2, β2) is the state parameter of the next shooting node in the adjacent shooting nodes, and ω0 is the standard angular velocity;

[0056] The conversion calculation processor calculates the inspection period T of the entire ring sequence according to the following formula:

[0057]

[0058] Where n is the number of shooting nodes, Tc(i, i+1) represents the conversion time between the i-th shooting node and the i+1-th shooting node in the ring sequence, and Tc(n, 1) represents the conversion time between the n-th shooting node and the 1st shooting node in the ring sequence;

[0059] The abnormality positioning unit includes a region setting processor and a relative conversion processor, wherein the region setting processor is used to set the position of the abnormal region in the image, and the relative conversion processor is used to convert the abnormal region position information into relative information with respect to the center of the image;

[0060] The abnormal area position information is represented by four rectangular positioning points, and the relative conversion processor calculates the relative information according to the following formula:

[0061]

[0062]

[0063] Wherein, x, y are the pixel length and width of the captured image, a, b are the pixel length and width of the abnormal area, x1 and x2 are the two horizontal coordinates of the positioning point, y1 and y2 are the two vertical coordinates of the positioning point, η is the magnification factor, and θ is the alignment angle.

[0064] Embodiment 2.

[0065] This embodiment includes all the contents of the first embodiment, and provides a smart cloud box inspection system based on edge computing, including an inspection planning module, an inspection control module, an image analysis module, and an inspection adjustment module;

[0066] The inspection route planning module is used to plan a basic inspection route. The inspection control module is used to control the shooting of the camera lens to meet the inspection requirements. The image analysis module is used to analyze and process the images collected during the inspection process. The inspection adjustment module adjusts the inspection process based on the analysis results;

[0067] Combined with Figure 2 , the inspection route planning module includes a node marking unit, a node connection unit, a parameter formulation unit, and a resource release unit. The node marking unit is used to mark the shooting nodes including the intelligent cloud box hardware devices. The node connection unit is used to connect the shooting nodes to form a closed route. The parameter formulation unit extracts inspection parameters based on the closed route. The resource release unit is used to release the computing resources used after completing the inspection route;

[0068] The inspection control module includes an inspection route storage unit and an inspection execution unit. The inspection route storage unit is used to save the inspection parameters. The inspection execution unit is used to obtain the inspection parameters and control the camera lens to execute the inspection process;

[0069] The image analysis module includes an anomaly analysis unit and an anomaly location unit. The anomaly analysis unit is used to analyze the abnormal phenomena in the images. The anomaly location unit is used to locate the abnormal positions;

[0070] The inspection adjustment module includes an ad-hoc inspection planning unit and an insertion interaction unit. The ad-hoc inspection planning unit is used to plan the ad-hoc added inspection parameters. The insertion interaction unit is used to insert the new inspection parameters into the basic inspection route;

[0071] Combined with Figure 3 , the node marking unit includes a device identification processor, a center calibration processor, and a status recording processor. The device identification processor is used to identify the hardware devices in the captured images. The center calibration processor is used to adjust the camera lens to make the hardware device at the center of the image. The status recording processor is used to record the camera lens status parameters as the shooting nodes;

[0072] The status parameters of each shooting node are represented by the horizontal angle α and the pitch angle β;

[0073] Combined with Figure 4 , the node connection unit includes a conversion calculation processor, a node sorting processor, and a sorting adjustment processor. The conversion calculation processor is used to calculate the conversion time between two nodes. The node sorting processor is used to sort the shooting nodes in a circular sequence. The sorting adjustment processor is used to adjust the sorting of the shooting nodes in the circular sequence;

[0074] The conversion calculation processor calculates the conversion time Tc between two adjacent shooting nodes in the circular sequence according to the following formula:

[0075]

[0076] Among them, (α1, β1) are the state parameters of the previous shooting node among adjacent shooting nodes, (α2, β2) are the state parameters of the subsequent shooting node among adjacent shooting nodes, and ω0 is the standard angular velocity;

[0077] The conversion calculation processor calculates the inspection cycle T of the entire circular sequence according to the following formula:

[0078]

[0079] Among them, n is the number of shooting nodes, Tc(i, i + 1) represents the conversion time between the i-th shooting node and the (i + 1)-th shooting node in the circular sequence, and Tc(n, 1) represents the conversion time between the n-th shooting node and the 1st shooting node in the circular sequence;

[0080] The processing process of the node connection unit for the shooting nodes includes the following steps:

[0081] S1. The node sorting processor sorts the shooting nodes in any order to form a circular sequence;

[0082] S2. The sorting adjustment processor selects two adjacent shooting nodes as target nodes;

[0083] S3. The conversion calculation processor calculates the change value of the inspection cycle after the target nodes are swapped;

[0084] S4. If the change value of the inspection cycle is negative, the sorting adjustment processor swaps the target nodes, clears the cumulative value, and returns to step S2. If the change value of the inspection cycle is positive, it enters step S5;

[0085] S5. Add 1 to the cumulative value. If the cumulative value is n, enter step S6, and then send the shooting node information in the circular sequence to the parameter determination unit. If the cumulative value is less than n, return to step S2;

[0086] It should be noted that in step S2, each time the shooting nodes are selected, the serial number progresses by one compared with the previous time. That is, if the 3rd and 4th shooting nodes were used as target nodes in the previous time, then the 4th and 5th shooting nodes will be used as target nodes in the next time;

[0087] Combined with Figure 5The parameter formulation unit includes a rotation parameter processor, a focus parameter processor and a format conversion processor. The rotation parameter processor obtains the rotation parameter based on two adjacent shooting nodes, the focus parameter processor obtains the focal length parameter based on the image information of a single shooting node, and the format conversion processor is used to convert the rotation parameter and the focal length parameter into an execution data packet in an executable data format after matching them;

[0088] The resource release unit includes an event activation processor and a resource release processor, wherein the event activation processor is used to generate an activation signal after waiting for the complete execution data packet to be produced, and the resource release processor is used to release the computing power resources used by the inspection planning module after the activation signal is generated;

[0089] The inspection route storage unit includes a data packet register and a data packet management processor, wherein the data packet register is used to store executable data packets, and the data packet management processor is used to control and manage the process of sending executable data packets to the inspection execution unit;

[0090] The inspection execution unit includes an inspection management processor and a lens control processor, wherein the inspection management processor is used to receive and manage executable data packets, and the shooting execution processor controls the lens to rotate, focus and shoot based on the executable data packets;

[0091] The abnormality analysis unit includes a standard image register and an image comparison processor, wherein the standard image register is used to store the normal hardware device image of each shooting node, and the image comparison processor is used to compare the normal hardware device image with the inspection shooting image to determine whether there is an abnormality;

[0092] The abnormality positioning unit includes a region setting processor and a relative conversion processor, wherein the region setting processor is used to set the position of the abnormal region in the image, and the relative conversion processor is used to convert the abnormal region position information into relative information with respect to the center of the image;

[0093] The abnormal area position information is represented by four rectangular positioning points, and the relative conversion processor calculates the relative information according to the following formula:

[0094]

[0095] Where x, y are the pixel length and width of the captured image, a, b are the pixel length and width of the abnormal area, x1 and x2 are the two horizontal coordinates of the positioning point, y1 and y2 are the two vertical coordinates of the positioning point, η is the magnification factor, and θ is the alignment angle;

[0096] The inspection planning unit includes a steering planning processor and a focal length adjustment processor, wherein the steering planning processor plans temporary steering parameters based on the straightening angle, and the prime focal length adjustment processor plans temporary focusing parameters based on the magnification factor;

[0097] The insertion interaction unit includes a time stop signal processor and a data packet insertion processor, wherein the time stop signal processor is used to send a time stop signal to the inspection execution unit when an abnormality is found, and the data packet insertion processor is used to generate a temporary executable data packet and send it to the inspection execution unit;

[0098] After receiving the time-stop signal, the inspection management processor interrupts the execution process of the next execution data packet and waits for receiving a temporary executable data packet, and the inspection management processor inserts and executes the temporary executable data packet;

[0099] The i appearing in the above text is an ordinal number used to indicate a sequence number.

[0100] The contents disclosed above are only preferred feasible embodiments of the present invention, and do not limit the protection scope of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention specification and drawings are included in the protection scope of the present invention. In addition, the elements therein can be updated as technology develops.

Claims

1. An intelligent cloud box inspection system based on edge computing, characterized in that, It includes an inspection planning module, an inspection control module, an image analysis module, and an inspection adjustment module; The inspection planning module is used to plan a basic inspection route. The inspection control module is used to control the shooting of the lens to meet the inspection requirements. The image analysis module is used to analyze and process the images collected during the inspection. The inspection adjustment module adjusts the inspection process based on the analysis results; The inspection planning module includes a node marking unit, a node connection unit, a parameter formulation unit, and a resource release unit. The node marking unit is used to mark the shooting nodes including the intelligent cloud box hardware devices. The node connection unit is used to connect the shooting nodes to form a closed route. The parameter formulation unit extracts the inspection parameters based on the closed route. The resource release unit is used to release the computing resources used after completing the inspection route; The inspection control module includes an inspection route storage unit and an inspection execution unit. The inspection route storage unit is used to save the inspection parameters. The inspection execution unit is used to obtain the inspection parameters and control the lens to execute the inspection process; The image analysis module includes an abnormality analysis unit and an abnormality positioning unit. The abnormality analysis unit is used to analyze the abnormal phenomena in the images. The abnormality positioning unit is used to locate the abnormal positions; The inspection adjustment module includes an ad-hoc inspection planning unit and an insertion interaction unit. The ad-hoc inspection planning unit is used to plan the ad-hoc added inspection parameters. The insertion interaction unit is used to insert the new inspection parameters into the basic inspection route; The node connection unit includes a conversion calculation processor, a node sorting processor, and a sorting adjustment processor. The conversion calculation processor is used to calculate the conversion time between two nodes. The node sorting processor is used to sort the shooting nodes in a circular sequence. The sorting adjustment processor is used to adjust the sorting of the shooting nodes in the circular sequence; The processing process of the node connection unit for the shooting nodes includes the following steps: S1. The node sorting processor sorts the shooting nodes in any order to form a circular sequence; S2. The sorting adjustment processor selects two adjacent shooting nodes as the target nodes; S3. The conversion calculation processor calculates the change value of the inspection cycle after the target nodes are swapped; S4. If the change value of the inspection cycle is negative, the sorting adjustment processor swaps the target nodes, clears the cumulative value, and returns to step S2. If the change value of the inspection cycle is positive, go to step S5; S5. Add 1 to the cumulative value. If the cumulative value is n, send the shooting node information in the circular sequence to the parameter formulation unit. If the cumulative value is less than n, return to step S2, where n is the number of shooting nodes.

2. The intelligent cloud box inspection system based on edge computing according to claim 1, wherein The conversion calculation processor calculates the conversion time Tc between two adjacent shooting nodes in the circular sequence according to the following formula: ； Among them, is the state parameter of the previous shooting node among adjacent shooting nodes, is the state parameter of the next shooting node among adjacent shooting nodes, is the standard angular velocity; The conversion calculation processor calculates the inspection cycle T of the entire circular sequence according to the following formula: ； Where n is the number of shooting nodes, Tc(i, i + 1) represents the conversion time between the i-th shooting node and the (i + 1)-th shooting node in the circular sequence, and Tc(n, 1) represents the conversion time between the n-th shooting node and the 1st shooting node in the circular sequence.

3. The intelligent cloud box inspection system based on edge computing according to claim 2, characterized in that, The anomaly positioning unit includes a region setting processor and a relative conversion processor. The region setting processor is used to set the position of the anomaly region in the image, and the relative conversion processor is used to convert the anomaly region position information into relative information with respect to the image center; The anomaly region position information is represented by 4 positioning points of a rectangle, and the relative conversion processor calculates the relative information according to the following formula: ； ； Where x and y are the pixel length and width of the captured image, a and b are the pixel length and width of the abnormal area, x1 and x2 are the two horizontal coordinates of the positioning point, and y1 and y2 are the two vertical coordinates of the positioning point. is the magnification factor, To correct the angle.

Images