Autonomous object recognition method and device based on edge calculation

By sending data to the management center for processing when an edge computing device fails or performance degradation and generating maintenance scores, equipment management is optimized, and the problem of undetected and managed abnormal equipment in the prior art is solved, and management efficiency and data are improved Processing effect.

CN119027785AActive Publication Date: 2024-11-26SHENZHEN PACA IOT TECH CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411126141.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-11-26
Estimated Expiration
2044-08-16

AI Technical Summary

Technical Problem

Existing object recognition equipment based on edge computing cannot detect and manage when abnormalities occur in edge computing equipment, resulting in a decrease in data processing efficiency and effectiveness. The existing maintenance methods cannot prioritize handling of abnormal equipment, reducing management efficiency.

Method used

When an edge computing device fails or performance deteriorates, the image data is sent to the management center for processing, and the equipment status and identification results generate maintenance scores, priority is given to the management of equipment with poor overall status, and management efficiency is improved.

Benefits of technology

It ensures data processing efficiency and effect, optimizes equipment management processes, improves timely maintenance of abnormal equipment, and improves overall management efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027785A_ABST
    Figure CN119027785A_ABST
Patent Text Reader

Abstract

The invention discloses an autonomous object recognition method and device based on edge computing, and relates to the technical field of object recognition, performance detection is performed on an edge computing device, and when it is detected that the edge computing device has a fault or performance reduction, the state of the edge computing device is recorded firstly; the image data is preprocessed and forwarded to a management and control center based on a wireless network, a maintenance score is generated for each identification device after the state of the edge computing device is combined with an object identification effect, and all the identification devices are sorted according to the maintenance scores to generate a device list; a maintenance order of all the identification devices is selected based on the device list. According to the identification method, when the fault or performance reduction of the edge computing device is analyzed, the data is sent to the management and control center for processing, the data processing efficiency and processing effect are guaranteed, and the management list is generated by regularly combining the state of the edge computing device and the data identification result, so that the identification device with poor overall state can be preferentially managed; the management efficiency is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of object recognition, and in particular to an autonomous object recognition method and device based on edge computing. Background Art

[0002] Edge computing is a technology that transfers computing and data processing capabilities from cloud centers to the edge of the network. It enables data to be processed close to the data source. The purpose of this is to reduce data transmission delays, increase processing speeds, save bandwidth, and enhance data privacy. Autonomous object recognition devices based on edge computing take advantage of this distributed computing to perform real-time object recognition.

[0003] The prior art has the following deficiencies: 1. Identification devices that adopt edge computing usually rely only on edge computing devices for data processing. However, in actual applications, when an edge computing device has an abnormality that causes performance degradation (the edge computing device is not in a faulty state and can still be used), the identification device cannot detect such anomalies, and continuing to use the edge computing device cannot guarantee data processing efficiency and processing effects; 2. There are usually two ways to perform maintenance and management on identification equipment. The first way is that when an identification equipment fails, the identification equipment sends a warning signal to the control center, and the control center sends a maintenance signal to the maintenance personnel. The second way is that the control center pre-sets a maintenance cycle. When the maintenance date arrives, all identification equipment under control is randomly maintained and managed. Both of the above methods cannot give priority to the management of identification equipment with overall abnormalities, which reduces the management efficiency of identification equipment.

[0004] Based on this, the present invention proposes an autonomous object recognition method and device based on edge computing. During operation, the performance of the edge computing device is first analyzed. When the edge computing device fails or the performance degrades, the data is sent to the control center for processing to ensure data processing efficiency and processing effect. The management list is regularly generated based on the edge computing device status and data recognition results, so that the recognition devices with poor overall status can be managed first, thereby improving management efficiency. Summary of the invention

[0005] The purpose of the present invention is to provide an autonomous object recognition method and device based on edge computing to address the deficiencies in the background technology.

[0006] In order to achieve the above object, the present invention provides the following technical solution: an autonomous object recognition method based on edge computing, the recognition method comprising the following steps: When an object appears in the monitoring area of ​​the recognition device, the recognition device automatically collects image data and performs performance testing on the edge computing device; When a failure or performance degradation of the edge computing device is detected, the edge computing device status is recorded, and then the image data is pre-processed and forwarded to the control center based on the wireless network. The control center pre-processes the image data and sends it to the recognition device for object recognition; After the object recognition is completed, the control center evaluates the object recognition effect of the recognition device, combines the edge computing device status with the object recognition effect to generate a maintenance score for each recognition device, sorts all recognition devices according to the maintenance score, generates a device list, and selects the maintenance order of all recognition devices based on the device list.

[0007] In a preferred embodiment, the performance test of the edge computing device includes the following steps: Obtain the power loss coefficient and heat conduction Fourier change coefficient of edge computing equipment; The power loss coefficient and the Fourier change coefficient of heat conduction are calculated comprehensively to obtain the equipment status index; Compare the acquired device status index with a preset first abnormal threshold and a second abnormal threshold, where the first abnormal threshold is used to determine whether the performance of the edge computing device has declined, and the second abnormal threshold is used to determine whether the edge computing device has failed; If the device status index is less than or equal to the first abnormal threshold, it is judged that the performance of the edge computing device has not declined. If the device status index is greater than the first abnormal threshold and the device status index is less than or equal to the second abnormal threshold, it is judged that the performance of the edge computing device has declined. If the device status index is greater than the second abnormal threshold, it is judged that the edge computing device is faulty.

[0008] In a preferred embodiment, the power loss coefficient and the Fourier transform coefficient of heat conduction are comprehensively calculated to obtain the equipment status index, which is expressed as: , where is the device status index, is the power loss coefficient, is the Fourier transform coefficient of heat conduction, , are the proportionality coefficients of the power loss coefficient and the Fourier transform coefficient of heat conduction, respectively, and , Both are greater than 0.

[0009] In a preferred embodiment, the control center evaluates the object recognition effect of the recognition device, including the following steps: The control center obtains the recognition error frequency and recognition time deviation during the object recognition process of the recognition device, normalizes the recognition error frequency and the recognition time deviation, maps the value range of the recognition error frequency and the recognition time deviation to [0,1], obtains the normalized value of the recognition error frequency and the normalized value of the recognition time deviation, and sums the normalized value of the recognition error frequency and the normalized value of the recognition time deviation to obtain the recognition weakening index; The obtained recognition weakening index is compared with the effect threshold. The effect threshold is used to evaluate the object recognition effect of the recognition device. If the recognition weakening index is less than or equal to the effect threshold, the object recognition effect of the recognition device is evaluated to be excellent. If the recognition weakening index is greater than the effect threshold, the object recognition effect of the recognition device is evaluated to be poor.

[0010] In a preferred embodiment, all identified devices are sorted according to the maintenance scores and then a device list is generated, including the following steps: The edge computing device status is combined with the object recognition effect to generate a maintenance score for each recognition device. The calculation expression is: , where Score management, is the device status index, To identify the weakening index, are the weights of the equipment status index and the identification weakening index, respectively, and ; Sort all identified devices from large to small according to the management score, generate a device list, select the maintenance order of all identified devices based on the device list, and send the maintenance order information to relevant maintenance personnel; When there is an identification device in the device list whose management score is greater than the score threshold, the control center generates an emergency maintenance mark for the identification device and sends it to the relevant maintenance personnel.

[0011] In a preferred embodiment, the logic for obtaining the recognition error frequency is: obtaining the number of recognition errors of the recognition device within the object recognition time period; The logic for obtaining the recognition time deviation is: subtracting the standard recognition time from the actual recognition time to obtain the recognition time deviation.

[0012] In a preferred embodiment, the calculation expression of the power loss coefficient is: , where is the power loss coefficient, is the nominal voltage of the device, is the instantaneous voltage, The time frame for monitoring; The calculation expression of the heat conduction Fourier variation coefficient is: , where is in position and time The temperature value will be and time The temperature value of is defined as the Fourier variation coefficient of heat conduction, is in position and time The temperature value, is the time step, which indicates the amount of time advancement at each iteration, is the spatial step length, which indicates the interval length of spatial discretization. is the thermal diffusivity, defined as ,in, is the thermal conductivity, is the density, is the specific heat capacity, and Represents the location and The temperature value at .

[0013] An autonomous object recognition device based on edge computing, comprising a performance detection module, a data forwarding module, and an object recognition module; Performance detection module: When an object appears in the monitoring area of ​​the recognition device, the recognition device automatically collects image data and performs performance detection on the edge computing device; Data forwarding module: When a failure or performance degradation of the edge computing device is detected, the edge computing device status is first recorded, and then the image data is pre-processed and forwarded to the control center based on the wireless network; Object recognition module: used to perform object recognition on preprocessed image data, and the object recognition results are sent to the control center.

[0014] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention automatically collects image data and performs performance detection on edge computing devices. When it is detected that the edge computing device has a fault or performance degradation, the edge computing device status is first recorded, and then the image data pre-processing is forwarded to the control center based on the wireless network. After the object recognition is completed, the control center evaluates the object recognition effect of the recognition device, generates a maintenance score for each recognition device after combining the edge computing device status with the object recognition effect, and sorts all recognition devices according to the maintenance score to generate a device list, and selects the maintenance order of all recognition devices based on the device list. During the operation of this recognition method, the edge computing device is first analyzed for performance. When analyzing the edge computing device failure or performance degradation, the data is sent to the control center for processing to ensure data processing efficiency and processing effect. The management list is generated regularly in combination with the edge computing device status and data recognition results, so that the recognition devices with poor overall status can be managed first, thereby improving management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0016] Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Example 1: Please refer to Figure 1 As shown, the present embodiment provides an autonomous object recognition method based on edge computing, and the recognition method includes the following steps: When an object appears in the monitoring area of ​​the recognition device, the recognition device automatically collects image data and performs performance testing on the edge computing device. When a fault or performance degradation is detected in the edge computing device, the edge computing device status is first recorded, and then the image data pre-processing is forwarded to the control center based on the wireless network. The control center pre-processes the image data and sends it to the recognition device for object recognition. After the object recognition is completed, the control center evaluates the object recognition effect of the recognition device, combines the edge computing device status with the object recognition effect, and generates a maintenance score for each recognition device. All recognition devices are sorted according to the maintenance score to generate a device list, and the maintenance order of all recognition devices is selected based on the device list, and the maintenance order information is sent to relevant maintenance personnel.

[0019] This application automatically collects image data and performs performance testing on edge computing devices. When it is detected that the edge computing device has a fault or performance degradation, the edge computing device status is first recorded, and then the image data pre-processing is forwarded to the control center based on the wireless network. After the object recognition is completed, the control center evaluates the object recognition effect of the recognition device, combines the edge computing device status with the object recognition effect, generates a maintenance score for each recognition device, and sorts all recognition devices according to the maintenance score to generate a device list, and selects the maintenance order of all recognition devices based on the device list. During the operation of this recognition method, the edge computing device is first analyzed for performance. When analyzing the edge computing device failure or performance degradation, the data is sent to the control center for processing to ensure data processing efficiency and processing effect. The management list is generated regularly in combination with the edge computing device status and data recognition results, so that the recognition devices with poor overall status can be managed first, improving management efficiency.

[0020] Embodiment 2: When an object appears in the monitoring area of ​​the recognition device, the recognition device automatically collects image data, including the following steps: 1) Object detection and triggering Area monitoring: The recognition device continuously monitors the specified area and captures environmental information in real time through sensors (such as cameras, infrared sensors, radars, etc.).

[0021] Motion detection: Use background subtraction, optical flow or other motion detection algorithms to identify the movement of objects in the monitoring area. Once an object is detected entering or leaving the monitoring area, image data acquisition is triggered.

[0022] 2) Image Capture Automatic focus and exposure adjustment: Before capturing an image, the recognition device automatically adjusts the camera's focus and exposure parameters to adapt to current lighting conditions, ensuring that clear, well-lit images are captured.

[0023] High-frequency sampling: Adjust the frame rate of image capture based on the speed and importance of the moving object. In the case of high-speed motion, the device may collect image data at a higher frame rate.

[0024] 3) Multi-angle and multi-frame acquisition Multi-angle capture: If the device is equipped with multiple cameras or has mobile capabilities, it can capture objects from different angles to obtain comprehensive information.

[0025] Multi-frame overlay: Continuously capture multiple frames of images in a short period of time, and enhance image clarity and detail through image overlay or stitching technology.

[0026] 4) Image data compression and storage Data compression: compress the collected image data to reduce storage space and transmission bandwidth requirements. Common compression methods include JPEG, PNG or video encoding (such as H.264, H.265).

[0027] Local storage: Store the processed image data in a local storage device. Especially for application scenarios with high real-time requirements, it can be saved locally first for subsequent analysis.

[0028] 5) Data labeling and classification Automatic labeling: Automatically label the captured images according to preset rules or using simple models, such as recording the time, location, environmental conditions, and the category of the detected objects.

[0029] Classification storage: Image data is classified and stored according to labels or categories to facilitate subsequent rapid retrieval and analysis.

[0030] The performance test of edge computing equipment includes the following steps: Obtain the power loss coefficient and heat conduction Fourier change coefficient of edge computing equipment; The power loss coefficient and the Fourier transform coefficient of heat conduction are calculated comprehensively to obtain the equipment status index, which is expressed as follows: , where is the device status index, is the power loss coefficient, is the Fourier transform coefficient of heat conduction, , are the proportionality coefficients of the power loss coefficient and the Fourier transform coefficient of heat conduction, respectively, and , All are greater than 0; The larger the device status index is, the worse the performance of the edge computing device is. The acquired device status index is compared with a preset first abnormal threshold and a second abnormal threshold. The first abnormal threshold is used to determine whether the performance of the edge computing device is degraded, and the second abnormal threshold is used to determine whether the edge computing device is faulty. If the device status index is less than or equal to the first abnormal threshold, it is judged that the performance of the edge computing device has not declined and can continue to be used. If the device status index is greater than the first abnormal threshold and the device status index is less than or equal to the second abnormal threshold, it is judged that the performance of the edge computing device has declined and is not supported. If the device status index is greater than the second abnormal threshold, it is judged that the edge computing device is faulty and requires immediate maintenance.

[0031] The calculation expression of power loss coefficient is: , where is the power loss coefficient, is the nominal voltage of the device, is the instantaneous voltage, The monitoring time range.

[0032] The larger the power loss coefficient, the higher the power consumed by the device at a certain moment. This usually indicates that there may be an abnormality in the device, which may lead to performance degradation or failure. The reasons include the following aspects: 1) Overload operation and component loss When the instantaneous power loss of a device is large, it often means that the device is in an overloaded state. Overload operation causes the internal components of the device to bear higher current and voltage, increasing heat and causing accelerated aging of components. In particular, power modules, processors, and memory, these core components are sensitive to heat. Long-term high power consumption operation may cause: Overheating: When the device generates too much heat, the cooling system may not be able to effectively dissipate the heat, causing the internal temperature to rise, which in turn affects the normal operation of electronic components.

[0033] Insufficient power supply: If the power consumed by the device exceeds the design capacity of the power module, the power supply may not be able to provide sufficiently stable power, causing problems such as device restarts and data loss.

[0034] 2) Current spikes and circuit stress The increase in instantaneous power loss is usually accompanied by a current spike. The current spike will cause greater stress on the circuit and may cause the following problems: Circuit damage: Instantaneous excessive current may damage the device's circuit boards, wires, or other sensitive components, causing partial function failure or complete damage to the device.

[0035] Transient voltage changes: When high-power loads are switched, short-term fluctuations in the power supply voltage may occur. Such transient voltage changes may affect the stability of other circuits or equipment.

[0036] 3) Unstable operating state Large instantaneous power fluctuations are often a sign that the device is performing heavy computing, data processing, or other high-load operations. In this case, the device may: Excessive resource usage: Critical resources such as CPU and memory are heavily occupied, causing the system to respond slowly, or even freeze or restart.

[0037] Increased electromagnetic interference: Increased power may also increase electromagnetic radiation, affecting other electronic devices inside or outside the device, causing signal interference and data errors.

[0038] 4) System load and failure risk When a device is in a high power consumption state for a long time, it means that it is performing a lot of computing tasks or processing a lot of data. This load increases the risk of system errors or crashes, especially when the system's heat dissipation or power supply cannot cope with these loads in time. Possible risks include: Software failure: Excessive load may trigger untested edge conditions, leading to software errors, process deadlocks, and other problems.

[0039] Hardware failure: Persistent high-power states may accelerate hardware aging and failure rates.

[0040] 5) Device abnormality or signs of attack Abnormal increases in instantaneous power may also be an early sign of a device being attacked by a cyber attack or hardware failure. For example: Malware activity: Malware may use a large amount of resources for mining, DDoS attacks, etc., which can cause a significant increase in instantaneous power.

[0041] Premonitions of hardware failure: Some hardware failures may manifest as abnormal power consumption before they occur, such as unstable power supply caused by a power module failure.

[0042] The calculation expression of the Fourier change coefficient of heat conduction is: , where is in position and time The temperature value will be and time The temperature value of is defined as the Fourier variation coefficient of heat conduction, is in position and time The temperature value, is the time step, which indicates the amount of time advancement at each iteration, is the spatial step length, which indicates the interval length of spatial discretization. is the thermal diffusivity, defined as ,in, is the thermal conductivity, is the density, is the specific heat capacity, and Represents the location and The temperature value at .

[0043] The larger the Fourier change coefficient of heat conduction, the more likely it is that the edge computing device has an abnormality that leads to performance degradation or failure, as follows: Overheating: Increased temperatures can cause the device to overheat, which can affect the performance of core components such as the processor and memory. Electronic devices usually operate within a specific temperature range. Exceeding this range can lead to slower computing speeds, increased errors, and even permanent damage to the hardware.

[0044] Increased power consumption: One direct reason for the temperature rise of edge computing devices is the increase in power consumption. When the device is processing complex tasks or due to system abnormalities (such as short circuit, overload, etc.), the power consumption will rise sharply and generate more heat. This will not only cause overheating, but also accelerate the depletion of batteries or power modules and reduce the life of the device.

[0045] Increased burden on the cooling system: When the device temperature is too high, the cooling system (such as fans or heat sinks) will increase its operation to maintain normal temperature. However, if the temperature continues to rise, the cooling system may exceed its design capacity and fail to dissipate heat effectively, causing the internal temperature of the device to rise further.

[0046] Possible signs of hardware failure: A sudden temperature rise may be a sign that some hardware components are about to fail. For example, a power module, circuit board, or some sensor may be abnormally hot due to aging, wear, or failure.

[0047] When a failure or performance degradation of an edge computing device is detected, the edge computing device status is first recorded, and then the image data is pre-processed and forwarded to the control center based on the wireless network, including the following steps: According to the type of wireless network (such as Wi-Fi, 4G / 5G, LoRa, etc.), select the appropriate transmission protocol (such as TCP / IP, UDP, etc.) for data transmission, and send the pre-processed image data to the control center through the wireless network; After receiving the data packet transmitted by the wireless network, the control center pre-processes the image data, including the following: Noise reduction and filtering: Perform noise reduction on images and use filtering technology to remove unnecessary information in the image to improve image quality.

[0048] Image cropping and scaling: Crop or scale the image according to the requirements of the recognition device to ensure that the image meets the input requirements of the recognition device.

[0049] Color space conversion: Converting an image from one color space to another (such as from RGB to grayscale or HSV) so that the recognition algorithm can process the data more efficiently.

[0050] Feature enhancement: Enhance the feature information in the image through edge detection, histogram equalization and other technologies to improve the recognition accuracy of the recognition equipment.

[0051] Compression: compress image data to reduce the bandwidth required for data transmission.

[0052] The control center pre-processes the image data and sends it to the recognition device for object recognition, which includes the following steps: Image loading: The recognition device loads the received image data into memory, ready for further processing.

[0053] Image standardization: Standardize the image, such as adjusting the brightness and contrast, so that it meets the input standards of the recognition model.

[0054] Data enhancement: Perform data enhancement before recognition, such as flipping, rotating, scaling, etc., to increase the robustness of recognition.

[0055] Model loading: The recognition device loads a pre-trained object recognition model (such as a convolutional neural network model) and prepares to process the input image data.

[0056] Object recognition: Image data is input into the recognition model, and the model extracts and classifies the image features to identify the objects in the image.

[0057] Result generation: The recognition model outputs the recognition result, including the category, location, confidence level and other information of the object.

[0058] After the object recognition is completed, the control center evaluates the object recognition effect of the recognition device, including the following steps: The control center obtains the recognition error frequency and recognition time deviation during the object recognition process of the recognition device, normalizes the recognition error frequency and the recognition time deviation, maps the value range of the recognition error frequency and the recognition time deviation to [0,1], obtains the normalized value of the recognition error frequency and the normalized value of the recognition time deviation, and sums the normalized value of the recognition error frequency and the normalized value of the recognition time deviation to obtain the recognition weakening index. The larger the recognition weakening index, the worse the recognition effect of the recognition device. The obtained recognition weakening index is compared with the effect threshold. The effect threshold is used to evaluate the object recognition effect of the recognition device. If the recognition weakening index is less than or equal to the effect threshold, the object recognition effect of the recognition device is evaluated to be excellent. If the recognition weakening index is greater than the effect threshold, the object recognition effect of the recognition device is evaluated to be poor.

[0059] The logic for obtaining the recognition error frequency is as follows: obtain the number of recognition errors of the recognition device within the object recognition time period, and divide the number of recognition errors by the recognition time to obtain the recognition error frequency. The larger the recognition error frequency, the worse the recognition effect of the recognition device.

[0060] The logic for obtaining the recognition time deviation is: subtract the standard recognition time from the actual recognition time to obtain the recognition time deviation. The larger the recognition time deviation is, the longer it takes for the recognition device to recognize the object and the worse the recognition effect of the recognition device is.

[0061] After combining the edge computing device status with the object recognition effect, a maintenance score is generated for each recognition device, and all recognition devices are sorted according to the maintenance score to generate a device list. The maintenance order of all recognition devices is selected based on the device list, and the maintenance order information is sent to relevant maintenance personnel, including the following steps: The edge computing device status is combined with the object recognition effect to generate a maintenance score for each recognition device. The calculation expression is: , where Score management, is the device status index, To identify the weakening index, are the weights of the equipment status index and the identification weakening index, respectively, and .

[0062] All identification devices are sorted from large to small according to the management score to generate a device list. The higher the identification device is ranked in the device list, the worse the overall status of the identification device is and the more priority maintenance is needed. The maintenance order of all identification devices is selected based on the device list, and the maintenance order information is sent to relevant maintenance personnel; When there is an identification device in the device list whose management score is greater than the score threshold, the control center generates an emergency maintenance mark for the identification device and sends it to the maintenance personnel, indicating that the identification device needs to be maintained immediately.

[0063] Embodiment 3: The autonomous object recognition device based on edge computing described in this embodiment includes a performance detection module, a data forwarding module, and an object recognition module; Performance detection module: When an object appears in the monitoring area of ​​the recognition device, the recognition device automatically collects image data and performs performance detection on the edge computing device. The performance detection results are sent to the data forwarding module; Data forwarding module: When a failure or performance degradation of the edge computing device is detected, the edge computing device status is first recorded, and then the image data is pre-processed and forwarded to the control center based on the wireless network. The pre-processed image data is sent to the object recognition module; Object recognition module: used to perform object recognition on preprocessed image data, and the object recognition results are sent to the control center.

[0064] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0065] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0066] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0067] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0068] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. The following steps are included: When an object appears in the monitoring area of ​​the recognition device, the recognition device automatically collects image data and performs performance testing on the edge computing device; When a failure or performance degradation of the edge computing device is detected, the edge computing device status is recorded, and then the image data is pre-processed and forwarded to the control center based on the wireless network. The control center pre-processes the image data and sends it to the recognition device for object recognition; After the object recognition is completed, the control center evaluates the object recognition effect of the recognition device, combines the edge computing device status with the object recognition effect to generate a maintenance score for each recognition device, sorts all recognition devices according to the maintenance score, generates a device list, and selects the maintenance order of all recognition devices based on the device list.

2. The autonomous object recognition method based on edge computing according to claim 1, characterized in that: The performance test of edge computing equipment includes the following steps: Obtain the power loss coefficient and heat conduction Fourier change coefficient of edge computing equipment; The power loss coefficient and the Fourier change coefficient of heat conduction are calculated comprehensively to obtain the equipment status index; Compare the acquired device status index with a preset first abnormal threshold and a second abnormal threshold, where the first abnormal threshold is used to determine whether the performance of the edge computing device has declined, and the second abnormal threshold is used to determine whether the edge computing device has failed; If the device status index is less than or equal to the first abnormal threshold, it is judged that the performance of the edge computing device has not declined. If the device status index is greater than the first abnormal threshold and the device status index is less than or equal to the second abnormal threshold, it is judged that the performance of the edge computing device has declined. If the device status index is greater than the second abnormal threshold, it is judged that the edge computing device is faulty.

3. The autonomous object recognition method based on edge computing according to claim 2, characterized in that: The power loss coefficient and the Fourier transform coefficient of heat conduction are calculated comprehensively to obtain the equipment status index, which is expressed as follows: , In the formula, is the device status index, is the power loss coefficient, is the Fourier transform coefficient of heat conduction, , Both are greater than 0.

4. The autonomous object recognition method based on edge computing according to claim 3 is characterized in that: The control center evaluates the object recognition effect of the recognition device, including the following steps: The control center obtains the recognition error frequency and recognition time deviation during the object recognition process of the recognition device, normalizes the recognition error frequency and the recognition time deviation, maps the value range of the recognition error frequency and the recognition time deviation to [0,1], obtains the normalized value of the recognition error frequency and the normalized value of the recognition time deviation, and sums the normalized value of the recognition error frequency and the normalized value of the recognition time deviation to obtain the recognition weakening index; The obtained recognition weakening index is compared with the effect threshold. The effect threshold is used to evaluate the object recognition effect of the recognition device. If the recognition weakening index is less than or equal to the effect threshold, the object recognition effect of the recognition device is evaluated to be excellent. If the recognition weakening index is greater than the effect threshold, the object recognition effect of the recognition device is evaluated to be poor.

5. The autonomous object recognition method based on edge computing according to claim 4, characterized in that: Generating a device list after sorting all identified devices according to maintenance scores includes the following steps: The edge computing device status is combined with the object recognition effect to generate a maintenance score for each recognition device. The calculation expression is: , In the formula, Score management, is the device status index, To identify the weakening index, , are the weights of the equipment status index and the identification weakening index, respectively, and ; Sort all identified devices from large to small according to the management score, generate a device list, select the maintenance order of all identified devices based on the device list, and send the maintenance order information to relevant maintenance personnel; When there is an identification device in the device list whose management score is greater than the score threshold, the control center generates an emergency maintenance mark for the identification device and sends it to the relevant maintenance personnel.

6. The autonomous object recognition method based on edge computing according to claim 4, characterized in that: The logic for obtaining the recognition error frequency is as follows: obtaining the number of recognition errors of the recognition device within the object recognition time period; The logic for obtaining the recognition time deviation is: subtracting the standard recognition time from the actual recognition time to obtain the recognition time deviation.

7. The autonomous object recognition method based on edge computing according to claim 3, characterized in that: The calculation expression of the power loss coefficient is: In the formula, is the power loss coefficient, is the nominal voltage of the device, is the instantaneous voltage, The time frame for monitoring; The calculation expression of the heat conduction Fourier variation coefficient is: , In the formula, is in position and time The temperature value at position and time The temperature value of is defined as the Fourier variation coefficient of heat conduction, is in position and time The temperature value, is the time step, which indicates the amount of time advancement at each iteration, is the spatial step length, which indicates the interval length of spatial discretization. is the thermal diffusivity, defined as ,in, is the thermal conductivity, is the density, is the specific heat capacity, and Represents the location and The temperature value at .

8. An autonomous object recognition device based on edge computing, used to implement the recognition method according to any one of claims 1 to 7, characterized in that: Including performance detection module, data forwarding module, object recognition module; Performance detection module: When an object appears in the monitoring area of ​​the recognition device, the recognition device automatically collects image data and performs performance detection on the edge computing device; Data forwarding module: When a failure or performance degradation of the edge computing device is detected, the edge computing device status is first recorded, and then the image data is pre-processed and forwarded to the control center based on the wireless network; Object recognition module: used to perform object recognition on preprocessed image data, and the object recognition results are sent to the control center.

Citation Information

Patent Citations

  • Intelligent monitoring and operation maintenance system and method

    CN110209104A

  • Power transmission apparatus, power reception apparatus, control method therefor, and program

    CN114788131A

  • Image recognition method and system based on computer network and storage medium

    CN117523299A

  • Embedded intelligent building monitoring system and monitoring method based on edge computing

    CN118011922A

  • Intelligent equipment monitoring and maintenance system

    CN118134458A