A single board sorting quality monitoring method and system based on intelligent monitoring sensor

Abstract

The present application provides a single board sorting quality monitoring method and system based on an intelligent monitoring sensor, which collects infrared thermal imaging images of the target single board in different partitions; determines the heat loss of the target single board during non-steady-state heat conduction and the thermal conductivity during steady-state heat conduction in each partition; divides each partition abnormally by all heat losses and all thermal conductivities to obtain multiple abnormal thermal conductivity zones of the target single board; determines the thermal conductivity characteristics at the cracks of the target single board, determines the thermal defect index of the target single board according to the thermal gradient distribution of each abnormal thermal conductivity zone and the thermal conductivity characteristics at the cracks, and monitors the target single board with quality defects based on the thermal defect index. The above scheme monitors the target single board with quality defects based on the thermal defect index, and can identify the complex and changeable cracks inside the target single board, thereby improving the quality monitoring accuracy of the target single board.

Description

Technical Field

[0001] The present application relates to the technical field of quality monitoring based on intelligent sensors, and more specifically, to a single board sorting quality monitoring method and system based on intelligent monitoring sensors. Background Art

[0002] Smart sensors are devices that combine sensing technology and information processing technology. They can process information and make decisions based on the perception of external physical quantities, and even communicate with other devices or systems. Compared with traditional sensors, smart sensors can not only collect and convert physical quantities, but also have functions such as self-calibration, self-diagnosis, signal processing and decision analysis. Smart sensors are widely used in industry, medical care, security and other fields. In the wood processing industry, especially in the quality sorting of single boards, smart monitoring sensors can be used to monitor the real-time single board data. Smart monitoring sensors have adaptive data processing and self-calibration functions, and can analyze and process the collected data in real time, realize multi-dimensional quality assessment of single boards, effectively identify defects, and thus improve sorting efficiency and accuracy.

[0003] In the existing single board sorting monitoring field based on intelligent sensors, the intelligent monitoring sensor illuminates the single board through a suitable light source and monitors the single board monitoring image. Then, the image monitored by the intelligent monitoring sensor is analyzed through an image processing algorithm, and edge detection, texture analysis and color recognition methods are applied to monitor the defects of the single board. However, in large-scale production, due to stress concentration in the manufacturing process and the effect of external mechanical stress, cracks are generated inside the target single board, and the morphology of the generated cracks is complex and changeable. Surface detection technology can usually only identify cracks on the surface of the target single board, and cannot identify cracks inside the target single board, which leads to a decrease in the quality monitoring accuracy of the target single board. Therefore, how to identify the complex and changeable cracks inside the target single board, thereby improving the quality monitoring accuracy of the target single board has become a difficult problem faced by the industry. Summary of the invention

[0004] The present application provides a single board sorting quality monitoring method and system based on an intelligent monitoring sensor, which can identify the complex and changeable cracks inside the target single board, thereby improving the quality monitoring accuracy of the target single board.

[0005] In a first aspect, the present application provides a single board sorting quality monitoring method based on an intelligent monitoring sensor, comprising the following steps:

[0006] Thermally stimulate different partitions of the target single board, and collect infrared thermal images of the target single board in different partitions through intelligent monitoring sensors;

[0007] For each partition, the intelligent monitoring sensor performs a local thermal state analysis on the target single board according to the infrared thermal imaging image corresponding to the partition, thereby obtaining the heat loss of the target single board during non-steady-state heat conduction and the thermal conductivity during steady-state heat conduction in each partition, and abnormally divides each partition according to all heat losses and all thermal conductivities to obtain multiple abnormal heat conduction areas of the target single board;

[0008] Acquire thermal conductivity information when cracks are generated in the target single board, extract thermal conductivity characteristics at the cracks of the target single board from the thermal conductivity information, and determine the thermal defect index of the target single board by the intelligent monitoring sensor according to the thermal gradient distribution of each abnormal thermal conductivity area and the thermal conductivity characteristics at the cracks;

[0009] A target single board with quality defects is monitored based on the thermal defect index.

[0010] In some embodiments, the intelligent monitoring sensor performs local thermal state analysis on the target board according to the infrared thermal imaging image corresponding to the partition, specifically including:

[0011] The intelligent monitoring sensor determines the non-steady-state heat conduction and steady-state heat conduction of the target board when conducting heat under the partition according to the infrared thermal imaging image corresponding to the partition;

[0012] The heat loss of the target single board during the non-steady-state heat conduction under the partition is obtained by extracting the non-steady-state heat conduction amount;

[0013] The thermal conductivity of the target single board in the steady-state heat conduction under partitioning is obtained through the steady-state heat conduction extraction.

[0014] In some embodiments, the intelligent monitoring sensor determines the non-steady-state heat conduction and the steady-state heat conduction of the target board when conducting heat under the partition according to the infrared thermal imaging image corresponding to the partition, specifically including:

[0015] Extracting all thermal imaging pixel values ​​of the infrared thermal imaging image according to the intelligent monitoring sensor;

[0016] Determining a local adjacent domain of the infrared thermal imaging image;

[0017] Performing adjacent pixel differences on all thermal imaging pixel values ​​based on the local adjacent domain, thereby obtaining pixel gradient stability corresponding to each thermal imaging pixel value;

[0018] The pixel gradient stability corresponding to each thermal imaging pixel value is compared with the preset pixel gradient stability to obtain the non-steady-state pixel gradient sequence and the steady-state pixel gradient sequence when the target single board conducts heat under the partition;

[0019] Determining the non-steady-state heat conduction amount of the target single board when conducting heat under the partition according to the non-steady-state pixel gradient sequence;

[0020] The steady-state heat conduction of the target board when conducting heat under the partition is determined according to the steady-state pixel gradient sequence.

[0021] In some embodiments, each partition is abnormally divided by all heat losses and all thermal conductivities to obtain multiple abnormal thermal conduction areas of the target board, specifically including:

[0022] Obtain the heat loss of the target board during non-steady-state heat conduction and the thermal conductivity during steady-state heat conduction in each partition;

[0023] Determine the thermal conductivity suspicion of the target board in each partition according to the heat loss of the target board in non-steady-state thermal conduction and the thermal conductivity in steady-state thermal conduction in each partition;

[0024] According to the suspicious degree of thermal conductivity corresponding to each partition, multiple abnormal thermal conductivity areas of the target board are obtained.

[0025] In some embodiments, determining the thermal conductivity suspicion of the target board in each partition according to the heat loss of the target board in non-steady-state thermal conduction and the thermal conductivity in steady-state thermal conduction in each partition specifically includes:

[0026] Obtain the thermal conductivity trust region of the target board in the absence of abnormalities;

[0027] Select a partition as a selected partition, and obtain the heat loss of the target board in the non-steady-state heat conduction and the thermal conductivity in the steady-state heat conduction under the selected partition;

[0028] Compensating the thermal conductivity loss during the steady-state heat conduction according to the heat loss during the non-steady-state heat conduction to obtain a thermal conductivity compensation value;

[0029] Mapping the thermal conductivity compensation value into the thermal conductivity trust region for confidence verification to obtain the thermal conductivity suspicion of the target board in the selected partition;

[0030] Continue to determine the thermal conductivity of the target board in the remaining partitions.

[0031] In some embodiments, extracting the thermal conductivity characteristics of the target single board crack from the thermal conductivity information specifically includes:

[0032] Acquiring a crack thermal image in the thermal conductivity information;

[0033] Extracting all crack pixel points at the target single plate crack from the crack thermal image;

[0034] For each crack pixel, pixel connectivity analysis is performed based on the pixel value corresponding to the crack pixel and the pixel values ​​corresponding to all non-crack pixels, thereby obtaining the pixel association sequence corresponding to each crack pixel;

[0035] The thermal conductivity characteristics of the target veneer cracks are determined according to the pixel association sequence corresponding to each crack pixel point.

[0036] In some embodiments, monitoring a target single board with quality defects based on the thermal defect index specifically includes:

[0037] The thermal defect index is compared with a preset thermal defect index. When the thermal defect index is greater than the preset thermal defect index, it is determined that the target board has quality defects. When the thermal defect index is less than or equal to the preset thermal defect index, it is determined that the target board does not have quality defects.

[0038] In a second aspect, the present application provides a single board sorting quality monitoring based on an intelligent monitoring sensor, including:

[0039] The acquisition module is used to perform thermal excitation on different partitions of the target single board and collect infrared thermal images of the target single board in different partitions through intelligent monitoring sensors;

[0040] A processing module is used for, for each partition, using an intelligent monitoring sensor to perform a local thermal state analysis on a target single board according to an infrared thermal imaging image corresponding to the partition, thereby obtaining heat loss of the target single board during non-steady-state heat conduction and thermal conductivity during steady-state heat conduction in each partition, and performing abnormal division of each partition according to all heat losses and all thermal conductivities to obtain multiple abnormal heat conduction areas of the target single board;

[0041] The processing module is further used to obtain thermal conductivity information when cracks occur in the target single board, extract thermal conductivity characteristics at the cracks of the target single board from the thermal conductivity information, and determine the thermal defect index of the target single board by the intelligent monitoring sensor according to the thermal gradient distribution of each abnormal thermal conductivity area and the thermal conductivity characteristics at the cracks;

[0042] The execution module is used to monitor the target single board with quality defects based on the thermal defect index.

[0043] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned single board sorting quality monitoring method based on intelligent monitoring sensors.

[0044] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned single board sorting quality monitoring method based on an intelligent monitoring sensor is implemented.

[0045] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects:

[0046] In the single board sorting quality monitoring method and system based on intelligent monitoring sensors provided in the present application, first, different partitions of the target single board are thermally excited, and infrared thermal imaging images of the target single board in different partitions are collected by the intelligent monitoring sensor; secondly, for each partition, the intelligent monitoring sensor performs local thermal state analysis on the target single board according to the infrared thermal imaging image corresponding to the partition, and then obtains the heat loss of the target single board in non-steady-state heat conduction and the thermal conductivity in steady-state heat conduction in each partition, and each partition is abnormally divided according to all heat losses and all thermal conductivities to obtain multiple abnormal thermal conduction areas of the target single board; then, the thermal conductivity information when cracks are generated in the target single board is obtained, and the thermal conductivity characteristics at the cracks of the target single board are extracted from the thermal conductivity information, and the thermal defect index of the target single board is determined by the intelligent monitoring sensor according to the thermal gradient distribution of each abnormal thermal conduction area and the thermal conductivity characteristics at the cracks; finally, the target single board with quality defects is monitored based on the thermal defect index.

[0047] It can be seen that the present application can realize the identification of cracks with complex and changeable internal morphology of the target single board, thereby improving the accuracy of quality monitoring of the target single board; first, the target single board is thermally excited by an infrared lamp, and the infrared thermal imaging image of the target single board in different partitions is acquired by the intelligent monitoring sensor, and then the thermal conductivity change of the target single board under thermal excitation is identified, and the internal defects of the target single board are analyzed by the thermal conductivity change, thereby avoiding the problem of being unable to identify the internal cracks of the target single board; secondly, the intelligent monitoring sensor performs local thermal state analysis on the target single board in different partitions based on the infrared thermal imaging image, and obtains the unstable heat conductivity and stable heat conductivity of the target single board when heated in different partitions, so as to identify the thermal conductivity capacity and internal structural integrity of the target single board, thereby avoiding the problem of difficulty in identifying internal defects caused by stress concentration and external mechanical stress during the manufacturing process of the target single board, thereby improving the recognition accuracy of abnormal conditions of the target single board; further The heat loss during non-steady-state heat conduction is extracted through the unstable heat conductivity when the target single board is heated, and the thermal conductivity of the target single board during steady-state heat conduction is extracted through the stable heat conductivity when the target single board is heated, and then the thermal conductivity suspicion of the target single board in different partitions is identified according to the heat loss during non-steady-state heat conduction and the thermal conductivity during steady-state heat conduction, so as to obtain the possible abnormality in the local area of ​​the target single board according to the thermal conductivity suspicion, thereby avoiding the problem of low sorting quality caused by missed detection of detailed parts; then, the intelligent monitoring sensor evaluates the thermal performance and structural integrity of the target single board according to the thermal gradient distribution of the abnormal heat conduction area and the thermal conductivity characteristics of the cracks of the target single board, so as to effectively characterize the possibility of abnormality of the target single board; finally, the target single board with quality defects is monitored based on the thermal performance and structural integrity of the target single board; in summary, the technical solution provided by the present application can realize the identification of cracks with complex and changeable internal morphology of the target single board, thereby improving the quality monitoring accuracy of the target single board. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is an exemplary flow chart of a single board sorting quality monitoring method based on an intelligent monitoring sensor according to some embodiments of the present application;

[0049] Figure 2 is an exemplary flow chart for determining a plurality of abnormal heat conduction areas according to some embodiments of the present application;

[0050] Figure 3 is an exemplary flow chart of determining thermal conductivity characteristics according to some embodiments of the present application;

[0051] Figure 4 is a structural schematic diagram of a single board sorting quality monitoring system based on an intelligent monitoring sensor according to some embodiments of the present application;

[0052] Figure 5 It is a structural schematic diagram of a computer device for implementing a single board sorting quality monitoring method based on an intelligent monitoring sensor as shown in some embodiments of the present application. DETAILED DESCRIPTION

[0053] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0054] refer to Figure 1 , which is an exemplary flow chart of a single board sorting quality monitoring method based on an intelligent monitoring sensor according to some embodiments of the present application. The single board sorting quality monitoring method 100 based on an intelligent monitoring sensor mainly includes the following steps:

[0055] In step 101, different partitions of a target single board are thermally stimulated, and infrared thermal images of the target single board in different partitions are collected by an intelligent monitoring sensor.

[0056] It should be noted that the intelligent monitoring sensor in this application is a device that integrates sensing, data processing and communication functions, and is used to monitor the environment or equipment status in real time and provide analysis and feedback. Compared with traditional sensors, intelligent monitoring sensors can not only accurately collect physical quantities, but also have intelligent functions such as signal processing, self-calibration, self-diagnosis and network connection.

[0057] During specific implementation, when starting the single board sorting quality monitoring, the infrared lamp can be placed at an appropriate distance from the surface of the target single board to ensure that the surface of the target single board is uniformly thermally stimulated, so as to thermally stimulate different partitions of the target single board, and collect infrared thermal images of the target single board in different partitions through the intelligent monitoring sensor. During specific implementation, the intelligent monitoring sensor can also be provided with an infrared thermal imaging sensor unit, which is not specifically limited here.

[0058] It should be noted that, in the present embodiment, different partitions of the target single board are obtained by uniform division, that is, the different partitions are partitions of uniform size. In addition, the infrared thermal imaging image in the present embodiment represents an image of the heat distribution of the target single board in different partitions, and the infrared thermal imaging image contains multiple thermal imaging pixel values, that is, each pixel value is used as a thermal imaging pixel value. In the quality monitoring of the target single board sorting, the target single board can be thermally excited to obtain an infrared thermal imaging image, and then the infrared thermal imaging image can be analyzed to identify defects inside the target single board, such as cracks, voids, delamination, etc. These defects will lead to uneven heat conduction, which is manifested as temperature anomalies in the infrared thermal imaging image.

[0059] It should also be noted that thermal excitation in this application refers to the process of applying heat to a target object to cause temperature changes, thereby utilizing these temperature changes to perform various analyses and tests. In the field of thermal imaging and non-destructive testing, thermal excitation is often used to reveal defects or inhomogeneities inside materials.

[0060] In step 102, for each partition, the intelligent monitoring sensor performs a local thermal state analysis on the target board according to the infrared thermal imaging image corresponding to the partition, and then obtains the heat loss of the target board during non-steady-state heat conduction and the thermal conductivity during steady-state heat conduction in each partition. Each partition is abnormally divided according to all heat losses and all thermal conductivities, and multiple abnormal heat conduction areas of the target board are obtained.

[0061] In some embodiments, the intelligent monitoring sensor performs local thermal state analysis on the target board according to the infrared thermal imaging image corresponding to the partition by specifically adopting the following steps, namely:

[0062] The intelligent monitoring sensor determines the non-steady-state heat conduction and steady-state heat conduction of the target board when conducting heat under the partition according to the infrared thermal imaging image corresponding to the partition;

[0063] The heat loss of the target single board during the non-steady-state heat conduction under the partition is obtained by extracting the non-steady-state heat conduction amount;

[0064] The thermal conductivity of the target single board in the steady-state heat conduction under partitioning is obtained through the steady-state heat conduction extraction.

[0065] Among them, in some embodiments, the intelligent monitoring sensor determines the non-steady-state heat conduction and steady-state heat conduction of the target board when conducting heat under the partition according to the infrared thermal imaging image corresponding to the partition, which can be specifically determined by the following steps, namely:

[0066] Extracting all thermal imaging pixel values ​​of the infrared thermal imaging image according to the intelligent monitoring sensor;

[0067] Determining a local adjacent domain of the infrared thermal imaging image;

[0068] Performing adjacent pixel differences on all thermal imaging pixel values ​​based on the local adjacent domain, thereby obtaining pixel gradient stability corresponding to each thermal imaging pixel value;

[0069] The pixel gradient stability corresponding to each thermal imaging pixel value is compared with the preset pixel gradient stability to obtain the non-steady-state pixel gradient sequence and the steady-state pixel gradient sequence when the target single board conducts heat under the partition;

[0070] Determining the non-steady-state heat conduction amount of the target single board when conducting heat under the partition according to the non-steady-state pixel gradient sequence;

[0071] The steady-state heat conduction of the target board when conducting heat under the partition is determined according to the steady-state pixel gradient sequence.

[0072] In specific implementation, all thermal imaging pixel values ​​of the infrared thermal imaging image can be extracted through the image processing tool OpenCV in the intelligent monitoring sensor, which will not be described in detail here.

[0073] In specific implementation, the local adjacent domain of the infrared thermal imaging image is determined, and the local adjacent domain can be set according to actual needs. For example, the local adjacent domain can be set to four adjacent domains or eight adjacent domains. In this embodiment, the local adjacent domain can be set to eight adjacent domains. In addition, in other embodiments, it can also be set to four adjacent domains, which is not limited here.

[0074] It should be noted that, in this embodiment, the local adjacent domain represents a range formed by a pixel point and its surrounding adjacent pixels.

[0075] In some embodiments, adjacent pixel differences are performed on all thermal imaging pixel values ​​based on the local adjacent domain to obtain the pixel gradient stability corresponding to each thermal imaging pixel value, which can be specifically performed by the following steps, namely:

[0076] Selecting a thermal imaging pixel value from all thermal imaging pixel values ​​as a selected thermal imaging pixel value;

[0077] Taking the selected thermal imaging pixel value as the center, extracting all adjacent thermal imaging pixel values ​​of the selected thermal imaging pixel value in the local adjacent domain;

[0078] Calculating the absolute difference between the selected thermal imaging pixel value and each adjacent thermal imaging pixel value to obtain all thermal imaging pixel differences, and determining the pixel gradient stability corresponding to the selected thermal imaging pixel value according to all thermal imaging pixel differences;

[0079] Continue to determine the pixel gradient stability corresponding to the remaining thermal imaging pixel values.

[0080] In a specific implementation, the pixel gradient stability corresponding to the selected thermal imaging pixel value is determined according to all thermal imaging pixel differences, that is, all thermal imaging pixel differences are averaged and the calculation result is taken inversely to obtain the pixel gradient stability corresponding to the selected thermal imaging pixel value.

[0081] It should be noted that, in other embodiments, other calculation methods may be used to calculate the pixel gradient stability, which is not limited here. In addition, in this embodiment, the pixel gradient stability represents the stability of the pixel gradient within the adjacent range of the current pixel point, that is, the greater the pixel gradient stability, the more stable the pixel gradient within the adjacent range of the current pixel point, and the smaller the pixel gradient stability, the more unstable the pixel gradient within the adjacent range of the current pixel point.

[0082] It should be noted that the preset pixel gradient stability in this embodiment refers to a preset standard pixel gradient stability, which can be set according to actual needs and will not be described in detail here.

[0083] In some embodiments, the pixel gradient stability corresponding to each thermal imaging pixel value is compared with the preset pixel gradient stability to obtain the non-steady-state pixel gradient sequence and the steady-state pixel gradient sequence when the target single board conducts heat under the partition. Specifically, the following steps can be adopted, namely:

[0084] For the pixel gradient stability corresponding to each thermal imaging pixel value, if the pixel gradient stability is less than or equal to the preset pixel gradient stability, the pixel gradient stability is used as the non-steady-state pixel gradient, and then multiple non-steady-state pixel gradients are obtained, and all the non-steady-state pixel gradients are combined to obtain the non-steady-state pixel gradient sequence when the target single board conducts heat under the partition;

[0085] If the pixel gradient stability is greater than the preset pixel gradient stability, the pixel gradient stability is used as the steady-state pixel gradient, and then multiple steady-state pixel gradients are obtained. All steady-state pixel gradients are combined to obtain a steady-state pixel gradient sequence when the target board is conducting heat under partitioning.

[0086] It should be noted that the non-steady-state pixel gradient sequence in the present embodiment includes multiple non-steady-state pixel gradients. The non-steady-state pixel gradient in the present embodiment represents an unstable pixel gradient. In the quality monitoring of the target single board sorting, the quality of the target single board is detected by thermal excitation. When the target single board is abnormal, the collected infrared thermal imaging image will have a certain pixel gradient change. If the pixel gradient changes too much, it means that the possibility of abnormality of the target single board is greater, and vice versa. Therefore, by determining the non-steady-state pixel gradient, the abnormal situation of the target single board in the current area can be effectively identified, so as to make more accurate abnormal alarms.

[0087] It should be noted that the steady-state pixel gradient sequence in the present embodiment includes multiple steady-state pixel gradients. The steady-state pixel gradient in the present embodiment represents a stably changing pixel gradient. In the quality monitoring of the target single board sorting, the quality of the target single board is detected by thermal excitation. When the target single board is abnormal, the collected infrared thermal imaging image will have a certain pixel gradient change. If the pixel gradient changes too much, it means that the possibility of abnormality of the target single board is greater, and vice versa. Therefore, by determining the steady-state pixel gradient, the abnormal situation of the target single board in the current area can be effectively identified, so as to make more accurate abnormal alarms.

[0088] In a specific implementation, the non-steady-state heat conductivity of the target board when conducting heat under partitioning is determined according to the non-steady-state pixel gradient sequence, that is, the mean calculation is performed on the non-steady-state pixel gradient sequence, and the mean calculation result is used as the non-steady-state heat conductivity of the target board when conducting heat under partitioning. In addition, in other embodiments, other calculation methods can also be used to calculate the non-steady-state heat conductivity, which is not limited here.

[0089] It should be noted that, in this embodiment, the non-steady-state heat conductivity represents the unstable heat conductivity of the target board when heated, that is, the larger the non-steady-state heat conductivity is, the larger the unstable heat conductivity of the target board when heated, and the smaller the non-steady-state heat conductivity is, the smaller the unstable heat conductivity of the target board when heated. Since the pixel values ​​in the infrared thermal imaging image can be used to characterize the thermal conductivity of the target board when heated, the non-steady-state heat conductivity can be determined by the non-steady-state pixel gradient, so as to identify the thermal conductivity of the target board in the current area. Usually, when the target board is abnormal, such as an internal crack, the thermal conductivity of the target board at the crack and the non-crack will be different when it is conducting heat. Therefore, by determining the non-steady-state heat conductivity, the abnormality of the target board can be effectively identified.

[0090] In a specific implementation, the steady-state heat conductivity of the target board when conducting heat under partitioning is determined according to the steady-state pixel gradient sequence, that is, the steady-state pixel gradient sequence is averaged, and the average calculation result is used as the steady-state heat conductivity of the target board when conducting heat under partitioning. In addition, in other embodiments, other calculation methods can also be used to calculate the steady-state heat conductivity, which is not limited here.

[0091] It should be noted that, in this embodiment, the steady-state thermal conductivity represents the stable thermal conductivity of the target board when heated, that is, the larger the steady-state thermal conductivity, the larger the stable thermal conductivity of the target board when heated, and the smaller the steady-state thermal conductivity, the smaller the stable thermal conductivity of the target board when heated. Since the pixel values ​​in the infrared thermal imaging image can be used to characterize the thermal conductivity of the target board when heated, the steady-state thermal conductivity can be determined by the steady-state pixel gradient, thereby identifying the thermal conductivity of the target board in the current area. Usually, when the target board is abnormal, such as an internal crack, the thermal conductivity of the target board at the crack and the non-crack will be different when it is conducting heat. Therefore, by determining the steady-state thermal conductivity, the abnormality of the target board can be effectively identified.

[0092] Among them, in some embodiments, the heat loss of the target single board in the non-steady-state heat conduction under the partition can be obtained by extracting the non-steady-state heat conduction from the non-steady-state heat conduction, which can be specifically obtained by the following steps, namely:

[0093] Obtaining the non-steady-state thermal conduction record of the target board under non-steady-state thermal conduction;

[0094] Performing trend fitting on all the non-steady-state thermal conductivity values ​​in the non-steady-state thermal conductivity record to obtain a non-steady-state thermal conductivity trend curve;

[0095] Extract all the non-steady-state thermal conductivity maxima and non-steady-state thermal conductivity minima from the non-steady-state thermal conductivity trend curve;

[0096] Determine the unsteady thermal conductivity trend characteristics of the target board under unsteady thermal conductivity according to all unsteady thermal conductivity minimum values ​​and all unsteady thermal conductivity maximum values;

[0097] The heat loss of the target board during non-steady-state heat conduction under partitioning is determined based on the non-steady-state heat conduction trend characteristics and the non-steady-state heat conduction amount.

[0098] In specific implementation, the non-steady-state thermal conductivity record of the target single board under non-steady-state thermal conductivity can be obtained through the single board quality sorting monitoring database. The non-steady-state thermal conductivity record records the non-steady-state thermal conductivity value of the target single board under historical non-steady-state thermal conductivity. In this embodiment, the non-steady-state thermal conductivity value represents the thermal conductivity of the target single board under non-stable thermal conduction. It should be noted that the single board quality sorting monitoring database in this embodiment is a system for recording and managing various data generated during the single board quality inspection process. The database contains single board inspection information, quality assessment results, defect location and type, temperature data, thermal imaging images and other information, with the purpose of supporting quality monitoring, data analysis and decision making.

[0099] In specific implementation, all the non-steady-state thermal conductivity values ​​in the non-steady-state thermal conductivity record can be trend-fitted by the least square method to obtain a non-steady-state thermal conductivity trend curve, that is, the non-steady-state thermal conductivity value in the non-steady-state thermal conductivity record is used as a vertical coordinate element, and the collection time point of the non-steady-state thermal conductivity value is used as a horizontal coordinate element, so as to perform trend fitting on all the non-steady-state thermal conductivity values ​​in the non-steady-state thermal conductivity record, which will not be described in detail here. In addition, in other embodiments, other fitting methods can also be used for fitting, which is not limited here. It should be noted that the non-steady-state thermal conductivity trend curve in this embodiment represents a thermal conductivity trend change curve of the target single board under non-stable thermal conduction.

[0100] In specific implementation, the first-order derivative method can be used to extract all the non-steady-state thermal conductivity maxima and non-steady-state thermal conductivity minima from the non-steady-state thermal conductivity trend curve, which will not be repeated here. In addition, in other embodiments, other extraction methods can be used to extract all the non-steady-state thermal conductivity maxima and non-steady-state thermal conductivity minima, which are not limited here.

[0101] It should be noted that, in this embodiment, the non-steady-state thermal conductivity maximum value represents the local maximum non-steady-state thermal conductivity value in the non-steady-state thermal conductivity trend curve, and the non-steady-state thermal conductivity minimum value point represents the local minimum non-steady-state thermal conductivity value in the non-steady-state thermal conductivity trend curve.

[0102] In specific implementation, the non-steady-state thermal conductivity trend characteristics of the target board under non-steady-state thermal conduction are determined according to all non-steady-state thermal conductivity minimum values ​​and all non-steady-state thermal conductivity maximum values, that is, the non-steady-state thermal conduction trend characteristics of the target board under non-steady-state thermal conduction are obtained by subtracting the average value of all non-steady-state thermal conductivity minimum values ​​from the average value of all non-steady-state thermal conductivity maximum values. In addition, in other embodiments, other calculation methods can be used to calculate the non-steady-state thermal conduction trend characteristics, which are not limited here.

[0103] It should be noted that the non-steady-state thermal conductivity trend characteristic in this embodiment represents the trend amount of the target single board changing over time under non-steady-state thermal conductivity, that is, the larger the non-steady-state thermal conductivity trend characteristic is, the larger the trend amount of the target single board changing over time under non-steady-state thermal conductivity, and vice versa. Usually, the target single board will have different changing trends under non-steady-state thermal conductivity. Therefore, by determining the non-steady-state thermal conductivity trend characteristic, the thermal conductivity change characteristics of the target single board under non-steady-state thermal conductivity can be effectively identified, and then the non-steady-state thermal conductivity trend characteristic is compared with the current non-steady-state thermal conductivity to analyze the heat loss of the target single board under non-steady-state thermal conductivity.

[0104] In a specific implementation, the heat loss of the target board during non-steady-state heat conduction under partitioning is determined by the non-steady-state heat conduction trend characteristic and the non-steady-state heat conductivity, that is, the absolute difference between the non-steady-state heat conduction trend characteristic and the non-steady-state heat conductivity is calculated to obtain the heat loss of the target board during non-steady-state heat conduction under partitioning. In addition, in other embodiments, other calculation methods may be used to calculate the heat loss, which is not limited here.

[0105] It should be noted that the heat loss under non-steady-state heat conduction in this embodiment represents the heat loss of the target single board under non-steady-state heat conduction, that is, the greater the heat loss, the greater the heat loss of the target single board under non-steady-state heat conduction, and vice versa. Since the target single board will be affected by different factors during heat conduction, the thermal conductivity changes at different times are inconsistent. When defects occur in the target single board, the heat conduction path will be affected, resulting in abnormal heat loss. Therefore, by determining the heat loss under non-steady-state heat conduction, defects in the single board, such as cracks, voids, delamination, etc., can be effectively detected and identified.

[0106] Among them, in some embodiments, the thermal conductivity of the target single board in the steady-state heat conduction under partitioning can be obtained by extracting the steady-state heat conduction value, which can be specifically obtained by the following steps, namely:

[0107] Obtaining steady-state thermal conduction records of the target board under steady-state thermal conduction;

[0108] Extracting an initial steady-state thermal conductivity value and a terminal steady-state thermal conductivity value from the steady-state thermal conductivity record;

[0109] Determine the thermal conductivity change of the target single board under steady-state thermal conductivity according to all initial steady-state thermal conductivity values ​​and final steady-state thermal conductivity values;

[0110] The thermal conductivity of the target single board in the partitioned state during steady-state thermal conduction is determined by the steady-state heat conduction and the thermal conductivity variation.

[0111] In specific implementation, the steady-state thermal conductivity record of the target single board under steady-state thermal conductivity can be obtained through the single board quality sorting monitoring database. The steady-state thermal conductivity record records the steady-state thermal conductivity value of the target single board under historical steady-state thermal conductivity. In this embodiment, the steady-state thermal conductivity value represents the heat conductivity of the target single board under stable thermal conductivity.

[0112] In specific implementation, the thermal conductivity change of the target board under steady-state thermal conduction is determined according to all initial steady-state thermal conductivity values ​​and terminal steady-state thermal conductivity values, that is, the absolute difference between the initial steady-state thermal conductivity value and the terminal steady-state thermal conductivity value can be calculated to obtain the thermal conductivity change of the target board under steady-state thermal conduction. In addition, in other embodiments, other calculation methods can be used to calculate, which is not limited here.

[0113] It should be noted that, in this embodiment, the thermal conductivity variation indicates the degree of thermal conductivity variation of the target board under steady-state thermal conductivity, that is, the greater the thermal conductivity variation, the greater the degree of thermal conductivity variation of the target board under steady-state thermal conductivity, and vice versa.

[0114] In specific implementation, the thermal conductivity of the target board in steady-state thermal conduction under partitioning is determined by the steady-state heat conductivity and the thermal conductivity change, that is, the ratio of the steady-state heat conductivity to the thermal conductivity change is calculated, and the ratio calculation result is used as the thermal conductivity of the target board in steady-state thermal conduction under partitioning. In addition, in other embodiments, other calculation methods can also be used to calculate the thermal conductivity, which is not limited here.

[0115] It should be noted that the thermal conductivity under steady-state thermal conduction in the present application represents the ability of the target single board to conduct heat under steady-state conditions, that is, the greater the thermal conductivity, the greater the ability of the target single board to conduct heat under steady-state conditions, and the smaller the thermal conductivity, the smaller the ability of the target single board to conduct heat under steady-state conditions. By determining the thermal conductivity, the thermal conductivity of the target single board under steady-state conditions can be effectively measured.

[0116] It should be noted that, in this embodiment, the local thermal state analysis represents the process of analyzing the thermal conductivity changes of the target board under the current partition, wherein the intelligent monitoring sensor performs local thermal state analysis on the target board according to the infrared thermal imaging image corresponding to the partition, that is: the intelligent monitoring sensor determines the non-steady-state heat conductivity and steady-state heat conductivity of the target board when conducting heat under the partition according to the infrared thermal imaging image corresponding to the partition; the heat loss of the target board during non-steady-state heat conduction under the partition is extracted from the non-steady-state heat conductivity; the thermal conductivity of the target board during steady-state heat conduction under the partition is extracted from the steady-state heat conductivity, thereby completing the local thermal state analysis.

[0117] In specific implementation, the heat loss of the target board during non-steady-state heat conduction and the thermal conductivity during steady-state heat conduction in each partition are determined by the implementation method of "the intelligent monitoring sensor performs local thermal state analysis on the target board according to the infrared thermal imaging image corresponding to the partition".

[0118] In some embodiments, reference Figure 2 As shown, this figure is an exemplary flow chart for determining abnormal heat conduction areas according to some embodiments of the present application. In this embodiment, each partition is abnormally divided by all heat losses and all thermal conductivities, and multiple abnormal heat conduction areas of the target single board can be obtained by the following steps:

[0119] First, in step 1021, the heat loss of the target board during non-steady-state heat conduction and the thermal conductivity during steady-state heat conduction in each partition are obtained;

[0120] First, in step 1022, the thermal conductivity suspicion of the target board in each partition is determined according to the heat loss of the target board in non-steady-state thermal conduction and the thermal conductivity in steady-state thermal conduction in each partition;

[0121] First, in step 1023, a plurality of abnormal thermal conductivity areas of the target board are obtained according to the thermal conductivity suspicion corresponding to each partition.

[0122] In some embodiments, the following steps may be used to determine the suspicious degree of thermal conductivity of the target board in each partition according to the heat loss of the target board in non-steady-state thermal conduction and the thermal conductivity in steady-state thermal conduction in each partition, namely:

[0123] Obtain the thermal conductivity trust region of the target board in the absence of abnormalities;

[0124] Select a partition as a selected partition, and obtain the heat loss of the target board in the non-steady-state heat conduction and the thermal conductivity in the steady-state heat conduction under the selected partition;

[0125] Compensating the thermal conductivity loss during the steady-state heat conduction according to the heat loss during the non-steady-state heat conduction to obtain a thermal conductivity compensation value;

[0126] Mapping the thermal conductivity compensation value into the thermal conductivity trust region for confidence verification to obtain the thermal conductivity suspicion of the target board in the selected partition;

[0127] Continue to determine the thermal conductivity of the target board in the remaining partitions.

[0128] In specific implementation, the thermal conductivity confidence domain of the target single board in the absence of abnormalities can be obtained through the single board quality sorting monitoring database. The thermal conductivity confidence domain is a pre-set confidence range of thermal conductivity. In this embodiment, the thermal conductivity confidence domain represents the confidence range of the thermal conductivity of the target single board in the absence of abnormalities.

[0129] In specific implementation, the thermal conductivity loss during the steady-state heat conduction is compensated according to the heat loss during the non-steady-state heat conduction to obtain a thermal conductivity compensation value, that is, the heat loss during the non-steady-state heat conduction and the thermal conductivity during the steady-state heat conduction are summed, and the sum is used as the thermal conductivity compensation value. In other embodiments, other calculation methods can also be used to calculate the thermal conductivity compensation value, which is not limited here. It should be noted that the thermal conductivity compensation value in this embodiment represents a value obtained after heat loss compensation is performed on the current thermal conductivity.

[0130] In a specific implementation, the thermal conductivity compensation value is mapped into the thermal conductivity trust domain for confidence verification to obtain the thermal conductivity suspicion of the target board in the selected partition, that is, the thermal conductivity compensation value is mapped into the range of the thermal conductivity trust domain, and the absolute difference between the thermal conductivity compensation value and the minimum thermal conductivity confidence value in the thermal conductivity trust domain is calculated to obtain the thermal conductivity suspicion of the target board in the selected partition. In addition, in other embodiments, other calculation methods may be used to calculate the thermal conductivity suspicion, which is not limited here.

[0131] It should be noted that, in this embodiment, the thermal conductivity suspicion degree indicates the suspicion degree of thermal conductivity anomaly of the target single board in the current partition, that is, the greater the thermal conductivity suspicion degree, the greater the suspicion degree of thermal conductivity anomaly of the target single board in the current partition, indicating that the possibility of defect in the target single board in the current partition is greater; the smaller the thermal conductivity suspicion degree, the smaller the suspicion degree of thermal conductivity anomaly of the target single board in the current partition, indicating that the possibility of defect in the target single board in the current partition is smaller. In the quality sorting of the target single board, an infrared detection method is usually used to perform defect detection on the target single board, so as to identify the tiny defects of the target single board. Therefore, by determining the thermal conductivity suspicion degree, the possibility of defect in the target single board in the current partition can be effectively identified, thereby improving the sorting quality of the target single board.

[0132] In specific implementation, the determination of the suspicious thermal conductivity of the remaining partitions can be carried out by continuing to determine the suspicious thermal conductivity of the target board in the remaining partitions through the above-mentioned determination method of "determining the suspicious thermal conductivity of the target board in the selected partition", which will not be repeated here.

[0133] In some embodiments, the following steps may be used to obtain multiple abnormal thermal conductivity areas of the target board according to the suspicious thermal conductivity corresponding to each partition, namely:

[0134] Obtaining a suspected thermal conductivity threshold during target single board sorting quality monitoring;

[0135] The thermal conductivity suspicious threshold is compared with the thermal conductivity suspicious degree corresponding to each partition, all thermal conductivity suspicious degrees greater than or equal to the thermal conductivity suspicious threshold are extracted, and the partitions corresponding to each extracted thermal conductivity suspicious degree are used as abnormal thermal conductivity areas of the target board. For all thermal conductivity suspicious degrees less than the thermal conductivity suspicious threshold, no processing is performed.

[0136] It should be noted that, in this embodiment, the abnormal thermal conductivity zone indicates a partition where the thermal conductivity of the target single board under thermal excitation is abnormal. In the quality monitoring of the target single board sorting, when the target single board is abnormal, the thermal conductivity of the target single board will also change. For example, when the target single board has a crack, the thermal conductivity at the crack will be different from the thermal conductivity of the normal single board area. Therefore, by identifying and analyzing the abnormal thermal conductivity zone, potential quality problems and defects can be discovered, thereby improving the overall quality and reliability of the product.

[0137] It should also be noted that the abnormal division in the present application means dividing the areas where the thermal conductivity is suspected to exceed the safety range from each partition, wherein each partition is abnormally divided according to all heat losses and all thermal conductivities, that is: obtaining the heat loss of the target board during non-steady-state thermal conduction and the thermal conductivity during steady-state thermal conduction in each partition; determining the thermal conductivity suspicion of the target board in each partition based on the heat loss of the target board during non-steady-state thermal conduction and the thermal conductivity during steady-state thermal conduction in each partition; and obtaining multiple abnormal thermal conduction areas of the target board according to the thermal conductivity suspicion corresponding to each partition, thereby completing the abnormal division of each partition.

[0138] In step 103, the thermal conductivity information of the target single board when cracks are generated is obtained, and the thermal conductivity characteristics of the target single board at the cracks are extracted from the thermal conductivity information. The intelligent monitoring sensor determines the thermal defect index of the target single board according to the thermal gradient distribution of each abnormal thermal conductivity area and the thermal conductivity characteristics at the cracks.

[0139] In a specific implementation, the thermal conductivity information of a target board when cracks are generated is obtained, that is, a target board on which cracks have been generated is obtained, and the thermal conductivity information of the target board when cracks are generated is obtained through an intelligent monitoring sensor. The thermal conductivity information represents the thermal characteristics exhibited by the crack during the heat conduction process when cracks are generated in the target board. These characteristics can provide important information about the existence, location, size and properties of the cracks. The thermal conductivity information includes a crack thermal image collected at the crack when cracks are generated in the target board.

[0140] In some embodiments, reference Figure 3 As shown in FIG. 1 , this figure is an exemplary flow chart of determining thermal conductivity characteristics according to some embodiments of the present application. In this embodiment, the thermal conductivity characteristics of the target single plate cracks can be extracted from the thermal conductivity information by using the following steps:

[0141] First, in step 1031, a crack thermal image in the thermal conductivity information is obtained;

[0142] Next, in step 1032, all crack pixel points at the target single plate crack are extracted from the crack thermal image;

[0143] Then, in step 1033, for each crack pixel point, pixel connectivity analysis is performed according to the pixel value corresponding to the crack pixel point and the pixel values ​​corresponding to all non-crack pixels, thereby obtaining a pixel association sequence corresponding to each crack pixel point;

[0144] Finally, in step 1034, the thermal conductivity characteristics of the cracks in the target single plate are determined according to the pixel association sequence corresponding to each crack pixel point.

[0145] In specific implementation, all crack pixel points at the crack of the target single board can be extracted from the crack thermal image through the Sobel algorithm in image processing, which will not be repeated here. In addition, in other embodiments, other image processing methods can also be used to extract all crack pixel points at the crack of the target single board, which is not limited here. The crack pixel points represent the pixel points at the crack of the target single board.

[0146] In specific implementation, pixel connectivity analysis is performed based on the pixel values ​​corresponding to the crack pixels and the pixel values ​​corresponding to all non-crack pixels, so as to obtain the pixel association sequence corresponding to each crack pixel, that is: for each crack pixel, the pixel value corresponding to the crack pixel is pixel-associated with the pixel values ​​corresponding to all non-crack pixels, so as to obtain the pixel association value corresponding to the non-crack pixel, and the pixel association values ​​corresponding to all non-crack pixels are combined to obtain the pixel association sequence corresponding to each crack pixel. The pixel association can be specifically implemented by calculating the difference method, for example, performing a difference calculation on the pixel value corresponding to the crack pixel and the pixel value corresponding to the non-crack pixel, and using the calculated result as the pixel association value corresponding to the non-crack pixel, so as to associate the pixel value corresponding to the crack pixel with the pixel value corresponding to the non-crack pixel. The details will not be repeated here.

[0147] It should be noted that the pixel connection sequence includes a plurality of pixel connection values, and the pixel connection value represents a connection value between the pixel value of the current pixel point and the pixel values ​​of the surrounding pixels.

[0148] In some embodiments, the following steps may be used to determine the thermal conductivity characteristics of the cracks in the target single plate according to the pixel association sequence corresponding to each crack pixel point, namely:

[0149] Determine the pixel connection extreme value ratio of each pixel connection sequence;

[0150] The average value of the pixel interconnection extreme value ratio of each pixel interconnection sequence is calculated to obtain the thermal conductivity characteristics of the target single plate crack.

[0151] In specific implementation, the pixel interconnection extreme value ratio of each pixel interconnection sequence is determined, that is, for each pixel interconnection sequence, the minimum pixel interconnection value and the maximum pixel interconnection value in the pixel interconnection sequence are calculated to obtain the pixel interconnection extreme value ratio of each pixel interconnection sequence.

[0152] It should be noted that, in this embodiment, the pixel connectivity extreme value ratio represents the minimum pixel connectivity value and the maximum pixel connectivity value, and is used to describe the degree of pixel change in the fracture thermal image.

[0153] It should also be noted that the thermal conductivity characteristics at the cracks in the present application represent the thermal conductivity characteristics and thermal conductivity behavior at the cracks when cracks appear in the target single board. The thermal conductivity characteristics at the cracks in the target single board are usually different from the thermal conductivity characteristics in the normal area. Therefore, by determining the thermal conductivity characteristics at the cracks, the crack detection of the target single board to be inspected can be effectively performed.

[0154] In some embodiments, the smart sensor determines the thermal defect index of the target single board according to the thermal gradient distribution of each abnormal thermal conductivity zone and the thermal conductivity characteristics of the crack by the following steps, namely:

[0155] Acquire infrared thermal imaging images of each abnormal heat conduction area according to the intelligent sensor;

[0156] For each abnormal heat conduction zone, determine the horizontal gradient entropy and the vertical gradient entropy of the infrared thermal imaging image of the abnormal heat conduction zone, determine the thermal gradient distribution of the abnormal heat conduction zone according to the horizontal gradient entropy and the vertical gradient entropy, and then obtain the thermal gradient distribution of each abnormal heat conduction zone;

[0157] Determine the thermal gradient distribution trend of the target single board according to the thermal gradient distribution of each abnormal heat conduction area;

[0158] The thermal defect index of the target single board is determined according to the thermal gradient distribution trend and the thermal conductivity characteristics at the crack.

[0159] In a specific implementation, the infrared thermal imaging image of each abnormal heat conduction area can be obtained according to the data storage library in the intelligent sensor.

[0160] In specific implementation, the horizontal gradient entropy and the vertical gradient entropy of the infrared thermal imaging image of the abnormal heat conduction area can be determined by the following steps, namely:

[0161] Extracting all horizontal gradient values ​​and vertical gradient values ​​of the infrared thermal imaging image;

[0162] The horizontal gradient entropy of the infrared thermal image of the abnormal heat conduction area is determined by all the horizontal gradient values;

[0163] The vertical gradient entropy of the infrared thermal imaging image of the abnormal heat conduction area is determined by all the vertical gradient values.

[0164] In specific implementation, all horizontal gradient values ​​and vertical gradient values ​​of the infrared thermal imaging image can be extracted through the data processing tool Python, which will not be repeated here. In addition, in other implementations, other data processing tools can also be used for extraction, which is not limited here.

[0165] In specific implementation, all horizontal gradient values ​​can be input as input parameters into the entropy function model, and the entropy function model outputs the horizontal gradient entropy of the infrared thermal imaging image of the abnormal heat conduction area. The entropy function model can be processed using an information entropy model. In addition, in other embodiments, other entropy function models can also be used for processing. There is no limitation here. It should be noted that the horizontal gradient entropy in this embodiment represents the degree of chaos in the change of the horizontal gradient value in the infrared thermal imaging image.

[0166] In specific implementation, all vertical gradient values ​​can be input as input parameters into the entropy function model, and the entropy function model outputs the vertical gradient entropy of the infrared thermal imaging image of the abnormal heat conduction area. The entropy function model can be processed using an information entropy model. In addition, in other embodiments, other entropy function models can also be used for processing, which is not limited here. In this embodiment, the vertical gradient entropy represents the degree of chaos in the change of the vertical gradient value in the infrared thermal imaging image.

[0167] In specific implementation, the thermal gradient distribution of the abnormal heat conduction zone is determined according to the horizontal gradient entropy and the vertical gradient entropy, that is, the ratio of the horizontal gradient entropy to the vertical gradient entropy is calculated to obtain the thermal gradient distribution of the abnormal heat conduction zone. In addition, in other embodiments, other calculation methods can be used to calculate the thermal gradient distribution, which is not limited here.

[0168] It should be noted that the thermal gradient distribution in the present application represents the thermal conductivity distribution of the target board in each abnormal thermal conductivity zone, that is, the thermal conductivity distribution of each suspicious thermal conductivity zone is measured by the change in the pixel value of the thermal imaging image. By determining the thermal gradient distribution, the suspicious thermal conductivity of the abnormal thermal conductivity zone can be effectively identified, thereby improving the defect detection accuracy of the target board.

[0169] In specific implementation, the thermal gradient distribution trend of the target single board is determined according to the thermal gradient distribution of each abnormal heat conduction zone, that is, the thermal gradient distribution of each abnormal heat conduction zone is averaged, and the average calculation result is used as the thermal gradient distribution trend of the target single board, and the thermal gradient distribution trend represents the trend change amount of the thermal gradient distribution of the target single board.

[0170] In a specific implementation, the thermal defect index of the target single board is obtained according to the thermal gradient distribution trend and the thermal conductivity characteristics at the crack. The thermal defect index can be calculated by the absolute difference between the thermal gradient distribution trend and the thermal conductivity characteristics at the crack. In addition, in other embodiments, other calculation methods can be used to calculate, which is not limited here.

[0171] It should be noted that the thermal defect index in the present application represents the severity of the abnormal thermal area identified in the thermal imaging analysis of the target single board. The thermal anomaly index is determined based on the thermal gradient distribution and the thermal conductivity characteristics at the cracks to evaluate the structural integrity of the target single board. That is, the larger the thermal defect index, the greater the severity of the abnormal thermal area identified in the thermal imaging analysis of the target single board, indicating that the possibility of the existence of a crack defect in the target single board is greater, that is, the worse the structural integrity of the target single board, the smaller the thermal defect index, the smaller the severity of the abnormal thermal area identified in the thermal imaging analysis of the target single board, indicating that the possibility of the existence of a crack defect in the target single board is smaller, that is, the better the structural integrity of the target single board. Since the thermal conductivity characteristics at the cracks of the target single board are inconsistent with the thermal conductivity characteristics of normal parts, the present application can characterize the severity of the abnormal thermal area identified in the thermal imaging analysis of the target single board by the absolute difference between the thermal gradient distribution trend and the thermal conductivity characteristics at the cracks, thereby improving the defect identification accuracy of the target single board.

[0172] In step 104, target boards with quality defects are detected based on the thermal defect index.

[0173] In some embodiments, the following steps may be specifically adopted to monitor the target single board with quality defects based on the thermal defect index, namely:

[0174] The thermal defect index is compared with a preset thermal defect index. When the thermal defect index is greater than the preset thermal defect index, it is determined that the target board has quality defects. When the thermal defect index is less than or equal to the preset thermal defect index, it is determined that the target board does not have quality defects.

[0175] It should be noted that the preset thermal defect index in this embodiment represents a pre-set value, which can be set according to actual needs and will not be repeated here. The preset thermal defect index is used to identify the degree of cracks in the target board. If it exceeds the threshold, it means that the target board has quality defects.

[0176] In addition, in another aspect of the present application, in some embodiments, the present application provides a single board sorting quality monitoring system based on an intelligent monitoring sensor, referring to Figure 4, which is a structural schematic diagram of a single board sorting quality monitoring system based on an intelligent monitoring sensor according to some embodiments of the present application. The single board sorting quality monitoring system 200 based on an intelligent monitoring sensor includes: a collection module 201, a processing module 202 and an execution module 203, which are described as follows:

[0177] The acquisition module 201 in the present application is mainly used to perform thermal excitation on different partitions of the target single board, and to collect infrared thermal images of the target single board in different partitions through an intelligent monitoring sensor;

[0178] Processing module 202, in the present application, the processing module 202 is mainly used for performing local thermal state analysis of the target single board according to the infrared thermal imaging image corresponding to the partition by the intelligent monitoring sensor for each partition, thereby obtaining the heat loss of the target single board during non-steady-state heat conduction and the thermal conductivity during steady-state heat conduction in each partition, and performing abnormal division of each partition through all heat losses and all thermal conductivities to obtain multiple abnormal heat conduction areas of the target single board;

[0179] The processing module 202 is further used to obtain thermal conductivity information when cracks occur in the target single board, extract thermal conductivity characteristics at the cracks of the target single board from the thermal conductivity information, and determine the thermal defect index of the target single board by the intelligent monitoring sensor according to the thermal gradient distribution of each abnormal thermal conductivity area and the thermal conductivity characteristics at the cracks;

[0180] The execution module 203 in the present application is mainly used to monitor the target single board with quality defects based on the thermal defect index.

[0181] In addition, the present application also provides a computer device, which includes a memory and a processor, the memory stores code, and the processor is configured to obtain the code and execute the above-mentioned single board sorting quality monitoring method based on intelligent monitoring sensors.

[0182] In some embodiments, reference Figure 5 , which is a schematic diagram of the structure of a computer device for implementing a single board sorting quality monitoring method based on an intelligent monitoring sensor according to some embodiments of the present application. The single board sorting quality monitoring method based on an intelligent monitoring sensor in the above embodiment can be Figure 5 The computer device 300 shown in the figure is implemented, and the computer device 300 includes at least one processor 301, a communication bus 302, a memory 303 and at least one communication interface 304.

[0183] The processor 301 can be a general-purpose central processing unit (CPU), or an application specific integrated circuit (ASIC) or one or more for controlling the execution of the single board sorting quality monitoring method based on intelligent monitoring sensors in the present application.

[0184] The communication bus 302 may be used to transmit information between the above-mentioned components.

[0185] The memory 303 may be a read only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read only memory (EEPROM), a compact disc read only memory (CD ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 303 may exist independently and be connected to the processor 301 via the communication bus 302. The memory 303 may also be integrated with the processor 301.

[0186] The memory 303 is used to store the program code for executing the solution of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the program code stored in the memory 303. The program code may include one or more software modules. The determination of the single board sorting quality monitoring method based on the intelligent monitoring sensor in the above embodiment can be implemented by the processor 301 and one or more software modules in the program code in the memory 303.

[0187] The communication interface 304 uses any transceiver or other device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0188] In a specific implementation, as an embodiment, a computer device may include multiple processors, each of which may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (such as computer program instructions).

[0189] The above-mentioned computer device may be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device may be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device or an embedded device. The embodiment of the present application does not limit the type of computer device.

[0190] In addition, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the single board sorting quality monitoring method based on the intelligent monitoring sensor is implemented.

[0191] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0192] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A single board sorting quality monitoring method based on intelligent monitoring sensor, characterized in that: The steps include: Thermally stimulate different partitions of the target single board, and collect infrared thermal images of the target single board in different partitions through intelligent monitoring sensors; For each partition, the intelligent monitoring sensor performs a local thermal state analysis on the target single board according to the infrared thermal imaging image corresponding to the partition, thereby obtaining the heat loss of the target single board during non-steady-state heat conduction and the thermal conductivity during steady-state heat conduction in each partition, and abnormally divides each partition according to all heat losses and all thermal conductivities to obtain multiple abnormal heat conduction areas of the target single board; Acquire thermal conductivity information when cracks are generated in the target single board, extract thermal conductivity characteristics at the cracks of the target single board from the thermal conductivity information, and determine the thermal defect index of the target single board by the intelligent monitoring sensor according to the thermal gradient distribution of each abnormal thermal conductivity area and the thermal conductivity characteristics at the cracks; Monitoring target single boards with quality defects based on the thermal defect index; Among them, each partition is abnormally divided according to all heat losses and all thermal conductivities, and multiple abnormal thermal conduction areas of the target board are obtained, including: Obtain the heat loss of the target board during non-steady-state heat conduction and the thermal conductivity during steady-state heat conduction in each partition; Determine the thermal conductivity suspicion of the target board in each partition according to the heat loss of the target board in non-steady-state thermal conduction and the thermal conductivity in steady-state thermal conduction in each partition; According to the suspicious degree of thermal conductivity corresponding to each partition, multiple abnormal thermal conductivity areas of the target board are obtained; Among them, determining the suspicious degree of thermal conductivity of the target board in each partition according to the heat loss of the target board in non-steady-state thermal conduction and the thermal conductivity in steady-state thermal conduction in each partition specifically includes: Obtain the thermal conductivity trust region of the target board in the absence of abnormalities; Select a partition as a selected partition, and obtain the heat loss of the target board in the non-steady-state heat conduction and the thermal conductivity in the steady-state heat conduction under the selected partition; Compensating the thermal conductivity loss during the steady-state heat conduction according to the heat loss during the non-steady-state heat conduction to obtain a thermal conductivity compensation value; Mapping the thermal conductivity compensation value into the thermal conductivity trust region for confidence verification to obtain the thermal conductivity suspicion of the target board in the selected partition; Continue to determine the thermal conductivity of the target board in the remaining partitions.

2. The method according to claim 1, characterized in that The intelligent monitoring sensor performs local thermal state analysis on the target board according to the infrared thermal imaging image corresponding to the partition, specifically including: The intelligent monitoring sensor determines the non-steady-state heat conduction and steady-state heat conduction of the target board when conducting heat under the partition according to the infrared thermal imaging image corresponding to the partition; The heat loss of the target single board during the non-steady-state heat conduction under the partition is obtained by extracting the non-steady-state heat conduction amount; The thermal conductivity of the target single board in the steady-state heat conduction under partitioning is obtained through the steady-state heat conduction extraction.

3. The method according to claim 2, characterized in that Determining the non-steady-state heat conduction and the steady-state heat conduction of the target board when conducting heat under the partition by the intelligent monitoring sensor according to the infrared thermal imaging image corresponding to the partition specifically includes: Extracting all thermal imaging pixel values ​​of the infrared thermal imaging image according to the intelligent monitoring sensor; Determining a local adjacent domain of the infrared thermal imaging image; Performing adjacent pixel differences on all thermal imaging pixel values ​​based on the local adjacent domain, thereby obtaining pixel gradient stability corresponding to each thermal imaging pixel value; The pixel gradient stability corresponding to each thermal imaging pixel value is compared with the preset pixel gradient stability to obtain the non-steady-state pixel gradient sequence and the steady-state pixel gradient sequence when the target single board conducts heat under the partition; Determining the non-steady-state heat conduction amount of the target single board when conducting heat under the partition according to the non-steady-state pixel gradient sequence; The steady-state heat conduction of the target board when conducting heat under the partition is determined according to the steady-state pixel gradient sequence.

4. The method according to claim 1, characterized in that Extracting the thermal conductivity characteristics of the target single plate crack from the thermal conductivity information specifically includes: Acquiring a crack thermal image in the thermal conductivity information; Extracting all crack pixel points at the target single plate crack from the crack thermal image; For each crack pixel, pixel connectivity analysis is performed based on the pixel value corresponding to the crack pixel and the pixel values ​​corresponding to all non-crack pixels, thereby obtaining the pixel association sequence corresponding to each crack pixel; The thermal conductivity characteristics of the target veneer cracks are determined according to the pixel association sequence corresponding to each crack pixel point.

5. The method according to claim 1, characterized in that Monitoring the target single board with quality defects based on the thermal defect index specifically includes: The thermal defect index is compared with a preset thermal defect index. When the thermal defect index is greater than the preset thermal defect index, it is determined that the target board has quality defects. When the thermal defect index is less than or equal to the preset thermal defect index, it is determined that the target board does not have quality defects.

6. A single board sorting quality monitoring system based on an intelligent monitoring sensor, which uses the method described in any one of claims 1 to 5 to perform single board sorting quality monitoring, characterized in that: The single board sorting quality monitoring system based on intelligent monitoring sensors includes: The acquisition module is used to perform thermal excitation on different partitions of the target single board and collect infrared thermal images of the target single board in different partitions through intelligent monitoring sensors; A processing module is used for, for each partition, using an intelligent monitoring sensor to perform a local thermal state analysis on a target single board according to an infrared thermal imaging image corresponding to the partition, thereby obtaining heat loss of the target single board during non-steady-state heat conduction and thermal conductivity during steady-state heat conduction in each partition, and performing abnormal division of each partition according to all heat losses and all thermal conductivities to obtain multiple abnormal heat conduction areas of the target single board; The processing module is further used to obtain thermal conductivity information when cracks occur in the target single board, extract thermal conductivity characteristics at the cracks of the target single board from the thermal conductivity information, and determine the thermal defect index of the target single board by the intelligent monitoring sensor according to the thermal gradient distribution of each abnormal thermal conductivity area and the thermal conductivity characteristics at the cracks; The execution module is used to monitor the target single board with quality defects based on the thermal defect index.

7. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores codes, and the processor is configured to obtain the codes and execute the single board sorting quality monitoring method based on an intelligent monitoring sensor as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the single board sorting quality monitoring method based on an intelligent monitoring sensor as described in any one of claims 1 to 5 is implemented.

Images