Temperature detection method and device based on multispectral image and storage medium

Abstract

The invention discloses a temperature detection method and device based on a multispectral image and a storage medium, and the method comprises the steps: carrying out the temperature detection of a target region in an obtained multispectral image, and obtaining the temperature information of the target region; performing abnormal temperature judgment on the target area by at least one preset abnormal temperature judgment strategy according to the temperature information to obtain a judgment result corresponding to each preset abnormal temperature judgment strategy; and determining a temperature detection result of the target area according to a judgment result corresponding to each preset abnormal temperature judgment strategy. According to the scheme, temperature detection can be efficiently and accurately carried out.

Description

Technical Field

[0001] This application relates to the field of target detection technology, and in particular to a temperature detection method, device, and storage medium. Background Technology

[0002] With the rapid development of an intelligent society, temperature detection technology has become increasingly important in various fields.

[0003] Taking the temperature detection scenario of power transmission equipment as an example, abnormal temperatures in power transmission equipment can lead to problems such as accelerated line aging, difficulty in heat dissipation, and fire hazards. Traditional temperature detection methods typically use infrared thermal imaging technology or manual temperature measurement. Although these methods can identify obvious overheating anomalies, they are difficult to capture early temperature anomalies in small areas. Furthermore, they rely on manual experience analysis, which is inefficient, increases labor costs, and is prone to misjudgment or missed detection.

[0004] Therefore, there is an urgent need for an efficient and accurate temperature detection method that can quickly and precisely identify abnormal temperatures. Summary of the Invention

[0005] This application provides at least one temperature detection method, apparatus, device, and computer-readable storage medium based on multispectral images.

[0006] The first aspect of this application provides a temperature detection method based on multispectral images, comprising: performing temperature detection processing on a target region in an acquired multispectral image to obtain temperature information of the target region; judging the abnormal temperature of the target region according to the temperature information using at least one preset abnormal temperature judgment strategy to obtain a judgment result corresponding to each preset abnormal temperature judgment strategy; and determining the temperature detection result of the target region according to the judgment result corresponding to each preset abnormal temperature judgment strategy.

[0007] In one embodiment, the step of judging the abnormal temperature of the target area according to the temperature information using at least one preset abnormal temperature judgment strategy to obtain the judgment result corresponding to each preset abnormal temperature judgment strategy includes: judging whether there is an abnormal temperature position in the target area according to the temperature information and a preset temperature threshold, wherein the temperature information of the abnormal temperature position is greater than the preset temperature threshold; in response to the existence of the abnormal temperature position in the target area, performing abnormal marking processing on the abnormal temperature position to obtain the judgment result.

[0008] In one embodiment, the target area is judged to have an abnormal temperature based on the temperature information using at least one preset abnormal temperature judgment strategy, and the judgment result corresponding to each preset abnormal temperature judgment strategy is obtained. This includes: constructing a spatiotemporal temperature field sequence based on the temperature of the target area at different detection times; determining the temperature change rate of the target area based on the spatiotemporal temperature field sequence and the detection time; and determining that the target area is abnormal in response to the temperature change rate being greater than a temperature change threshold, thereby obtaining the judgment result.

[0009] In one embodiment, before determining that the target area is abnormal in response to the temperature change rate being greater than a temperature change threshold and obtaining the determination result, the method further includes: acquiring a data distribution range of the temperature change rate; and analyzing the data distribution range according to a preset sliding window to obtain the temperature change threshold.

[0010] In one embodiment, the step of judging the abnormal temperature of the target area according to the temperature information using at least one preset abnormal temperature judgment strategy to obtain the judgment result corresponding to each preset abnormal temperature judgment strategy includes: determining the abnormal temperature location in the target area according to the temperature information; performing temperature diffusion analysis on the abnormal temperature location to obtain the diffusion analysis result of the adjacent area of ​​the abnormal temperature location; and judging the target area as abnormal in response to the diffusion analysis result indicating that the temperature change rate of the adjacent area is greater than the temperature change threshold, thereby obtaining the judgment result.

[0011] In one embodiment, determining the temperature detection result of the target area based on the judgment results corresponding to each preset abnormal temperature judgment strategy includes: acquiring the judgment results corresponding to each preset abnormal temperature judgment strategy, wherein the judgment results include abnormal results; determining the temperature abnormality level of the target area based on the number of abnormal results; and generating a temperature detection result corresponding to the temperature abnormality level.

[0012] In one embodiment, before performing temperature detection processing on the target region in the acquired multispectral image to obtain the temperature information of the target region, the method further includes: acquiring a multispectral image of an initial region; performing target detection processing on the multispectral image to obtain the target region in the initial region.

[0013] In one embodiment, the temperature information includes an infrared temperature matrix and a spatiotemporal temperature field sequence. The step of performing temperature detection processing on the target region in the acquired multispectral image to obtain the temperature information of the target region includes: extracting the infrared features of the target region in the multispectral image to obtain the infrared temperature matrix; and constructing the spatiotemporal temperature field sequence of the target region based on the infrared temperature matrices of the same target region acquired at different times.

[0014] A second aspect of this application provides a temperature detection device based on a multispectral image, comprising: a temperature detection module for performing temperature detection processing on a target region in an acquired multispectral image to obtain temperature information of the target region; an anomaly judgment module for judging the abnormal temperature of the target region according to the temperature information using at least one preset abnormal temperature judgment strategy to obtain a judgment result corresponding to each preset abnormal temperature judgment strategy; and a result determination module for determining the temperature detection result of the target region according to the judgment result corresponding to each preset abnormal temperature judgment strategy.

[0015] A third aspect of this application provides an electronic device, including a memory and a processor, wherein the processor is used to execute program instructions stored in the memory to implement the above-described temperature detection method based on multispectral images.

[0016] The fourth aspect of this application provides a computer-readable storage medium having program instructions stored thereon, which, when executed by a processor, implement the above-described temperature detection method based on multispectral images.

[0017] The above scheme acquires multispectral images containing the target region using a multispectral camera, thereby determining the temperature information of the target region based on the multispectral data. Then, it determines whether the temperature information of the target region is abnormal according to at least one preset abnormal temperature judgment strategy, obtaining the judgment result corresponding to each preset abnormal temperature judgment strategy. The judgment results of different strategies can be the same or different, and the judgment results can include normal results and / or abnormal results. Therefore, the judgment results corresponding to each preset abnormal temperature judgment strategy can be combined and analyzed to determine the temperature detection result of the target region, achieving efficient and accurate temperature detection.

[0018] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this application. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with this application and, together with the specification, serve to explain the technical solutions of this application.

[0020] Figure 1 This is a schematic flowchart of an exemplary embodiment of the temperature detection method based on multispectral images according to this application;

[0021] Figure 2 This is an exemplary data preprocessing flowchart of the temperature detection method based on multispectral images in this application;

[0022] Figure 3This is an exemplary flowchart of determining temperature information in the temperature detection method based on multispectral images of this application;

[0023] Figure 4 This is an exemplary flowchart for determining the rate of temperature change in the temperature detection method based on multispectral images of this application;

[0024] Figure 5 This is an overall schematic diagram of the temperature detection method based on multispectral images in this application, which determines the temperature detection result according to multiple strategies.

[0025] Figure 6 This is a block diagram illustrating a temperature detection device based on multispectral images, as shown in an exemplary embodiment of this application.

[0026] Figure 7 This is a schematic diagram of the structure of an embodiment of the electronic device of this application;

[0027] Figure 8 This is a schematic diagram of the structure of an embodiment of the computer-readable storage medium of this application. Detailed Implementation

[0028] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0029] In the following description, specific details such as particular system architectures, interfaces, and technologies are presented for illustrative purposes rather than for limiting purposes, in order to provide a thorough understanding of this application.

[0030] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Furthermore, "many" in this document means two or more. Moreover, the term "at least one" in this document means any combination of at least two of any one or more of a plurality of objects. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0031] For ease of understanding, one of the applicable scenarios of this application is illustrated below. The temperature detection method of this application can be applied to a variety of temperature detection scenarios, and is not limited here, such as the temperature detection scenario of power equipment.

[0032] Traditional methods for temperature detection of power transmission equipment typically employ infrared thermal imaging or manual static temperature measurement. While these methods can identify obvious overheating anomalies, they struggle to detect early temperature anomalies in smaller areas (such as contact points and clamps). Traditional methods often rely on threshold methods for judgment and alerts, lacking accuracy in detecting slow-moving thermal problems (such as component aging or increased line contact resistance). Furthermore, they heavily depend on manual experience, leading to inefficiency, increased labor costs, and a high risk of misjudgment or missed detections.

[0033] With the increasing demand for intelligent power equipment, there is an urgent need for more efficient and accurate temperature detection methods to reduce power outages or equipment damage caused by faults, ensure the safe and stable operation of the power system, and improve the level of refined management of power transmission equipment.

[0034] Please see Figure 1 , Figure 1 This is a schematic flowchart of an exemplary embodiment of the temperature detection method based on multispectral images according to this application. Specifically, it may include the following steps:

[0035] Step S110: Perform temperature detection processing on the target area in the acquired multispectral image to obtain the temperature information of the target area.

[0036] Multispectral images refer to images containing light across multiple wavelengths, encompassing a wider spectral range and more spectral bands compared to common visible light images. Temperature information can include various data points reflecting the temperature of the target area, such as the temperature value at each point within the target area, and / or the average temperature of the target area, and / or the average temperature of certain portions of the target area. Furthermore, the temperature values ​​determined from the multispectral data can be arranged according to the temporal sequence of the various multispectral data points, thereby reflecting the temperature changes of the target area over time.

[0037] For example, conventional methods typically employ drone inspections for temperature detection. The method of this application can similarly utilize drones for multispectral image acquisition, or alternatively, a fixed camera can be used for multispectral image acquisition.

[0038] Preferably, a fixed multispectral camera array can be set up in specific application scenarios to continuously detect target areas (such as some micro-local areas) of power transmission equipment. The fixed multispectral camera array provides sustainability in temperature detection, enabling continuous data acquisition around the clock, avoiding the risk of missed detections due to inspection intervals. It is particularly suitable for scenarios requiring real-time tracking of micro-local temperature changes, and the fixed camera does not need to deal with issues such as flight vibrations. It can directly integrate a high-resolution sensor, and combined with spatial calibration from a fixed viewing angle, achieve temperature difference detection with a smaller threshold trigger. The multispectral camera is equipped with a high-resolution multispectral sensor, enabling simultaneous acquisition of multiple bands, and ensuring timestamp alignment of the multispectral data through hardware synchronization signals. The temperature information of the target area can then be determined based on the multispectral data.

[0039] It should be noted that when deploying multispectral cameras in front of power transmission equipment, the camera's field of view must at least include the key heat-generating components of the target equipment (the key heat-generating components can be the target area or a portion of the target area). High-sensitivity multispectral sensor arrays can cover spectral ranges such as visible light and mid-to-long-wave infrared light. Therefore, by acquiring the infrared band, which is closely related to thermal radiation characteristics, minute temperature changes on the equipment surface can be captured. Then, the acquired multispectral data can be preprocessed, and temperature characteristics can be analyzed to obtain the temperature information of the target area.

[0040] For example, you can refer to... Figure 2 As shown, Figure 2 This is an exemplary data preprocessing flowchart of the temperature detection method based on multispectral images according to this application. A synchronization signal is triggered by receiving a software instruction to align the temporal sequence of multispectral data (multispectral images) acquired by the multispectral sensor. Image features (e.g., corner features) of different spectral images are extracted, and spatial registration is performed using these features to ensure pixel-based spatial alignment of multiple spectral images. The error of feature points in overlapping areas of different spectral images can also be calculated. If the error exceeds a preset error threshold, feature points of the spectral images are re-extracted. If the error is less than or equal to the preset error threshold, the information can be integrated and output for subsequent temperature detection.

[0041] Furthermore, one can refer to, for example Figure 3 As shown, Figure 3This is an exemplary flowchart of the temperature detection method based on multispectral images in this application, illustrating the determination of temperature information. During temperature detection, the target region is obtained by segmenting the area containing key components of the device in the spatiotemporally aligned multispectral image using a deep learning model. Then, the temperature information is determined using infrared band features from the multispectral camera (e.g., extracting the infrared temperature matrix of a micro-local area and constructing a spatiotemporal temperature field sequence). The spatiotemporal temperature field sequence refers to a data sequence obtained by arranging temperature values ​​detected at the same location or in the same region in chronological order (e.g., arranging multiple time-determined infrared temperature matrices in chronological order). Each element in the infrared temperature matrix of the target region can represent the temperature value of each point (pixel) in the target region, or each element can represent the temperature values ​​of multiple points (pixels) in the target region (e.g., by referring to the principle of downsampling, the temperature values ​​of multiple points can be represented by the same value (e.g., the average value)). Further details are omitted here.

[0042] Step S120: Based on the temperature information, the target area is judged for abnormal temperature using at least one preset abnormal temperature judgment strategy, and the judgment result corresponding to each preset abnormal temperature judgment strategy is obtained.

[0043] Referring to the preceding steps, temperature information may include, for example, infrared temperature matrices and spatiotemporal temperature field sequences. This application pre-sets multiple abnormal temperature judgment strategies, which can be used to determine whether temperature anomalies exist in a target area based on the acquired temperature information. These strategies can perform single-point temperature judgment, diffuse temperature judgment, and temperature change trend judgment on the target area, respectively, from point to area and combined with the prediction of temperature change trends, to accurately detect whether temperature anomalies exist in the target area.

[0044] In the execution of the temperature detection method of this application, one or more preset abnormal temperature judgment strategies can be selected to judge abnormal temperature, and the judgment result corresponding to the selected preset abnormal temperature judgment strategy can be obtained. No limitation is made here.

[0045] It should also be noted that different judgment strategies operate on different principles; therefore, the judgment results obtained from different judgment strategies may be the same or different, which is not limited here. The final judgment result may include normal results and / or abnormal results.

[0046] Step S130: Determine the temperature detection result of the target area based on the judgment result corresponding to each preset abnormal temperature judgment strategy.

[0047] Based on the steps described above, one or more preset abnormal temperature judgment strategies can be selected in the specific implementation process, and therefore one or more judgment results can be obtained.

[0048] If only one preset abnormal temperature judgment strategy is selected, the judgment result of that strategy will be the temperature detection result of the target area. If multiple preset abnormal temperature judgment strategies are selected, the judgment results of the multiple strategies can be combined for comprehensive analysis to obtain the temperature detection result of the target area.

[0049] For example, in a scenario with multiple preset abnormal temperature judgment strategies, if the number of abnormal results is less than or equal to a preset threshold, a temperature detection result indicating that the target area temperature is normal can be obtained. Conversely, if the number of abnormal results is greater than the preset threshold, a temperature detection result indicating that the target area temperature is abnormal can be obtained.

[0050] Alternatively, a temperature detection result indicating that the temperature of the target area is normal can be obtained when the ratio of abnormal results to all judgment results (abnormality ratio) is less than or equal to a preset ratio threshold. Conversely, a temperature detection result indicating that the temperature of the target area is abnormal can be obtained when the ratio of abnormal results to all judgment results is greater than the preset ratio threshold.

[0051] As another example, temperature detection results can be set to different anomaly levels in stages, based on the number or proportion of abnormal results. A higher anomaly level indicates a more severe anomaly. Different alert methods can be used to indicate anomalies for different anomaly levels; details will not be elaborated here.

[0052] As can be seen, this application acquires multispectral images containing the target region using a multispectral camera, thereby determining the temperature information of the target region based on the multispectral data of the multispectral images. Then, it can determine whether the temperature information of the target region is abnormal according to at least one preset abnormal temperature judgment strategy, obtaining the judgment result corresponding to each preset abnormal temperature judgment strategy. The judgment results of different strategies can be the same or different, and the judgment results can include normal results and / or abnormal results. Therefore, the judgment results corresponding to each preset abnormal temperature judgment strategy can be combined and analyzed to determine the temperature detection result of the target region, achieving efficient and accurate temperature detection.

[0053] Based on the above embodiments, this application embodiment describes the steps of judging the abnormal temperature of a target area according to temperature information using at least one preset abnormal temperature judgment strategy, and obtaining the judgment result corresponding to each preset abnormal temperature judgment strategy. Specifically, the method of this embodiment includes the following steps:

[0054] Based on the temperature information and the preset temperature threshold, it is determined whether there are abnormal temperature locations in the target area. If the temperature information of the abnormal temperature location is greater than the preset temperature threshold, the abnormal temperature location is marked as abnormal in response to the existence of abnormal temperature location in the target area, and the determination result is obtained.

[0055] In conjunction with the foregoing embodiments, this embodiment provides an exemplary description of one of the preset abnormal temperature judgment strategies that can be adopted in this application.

[0056] For ease of illustration, the embodiments of this application use pixels as the basis for single-point temperature judgment. Specifically, each pixel in the target area can be traversed and the temperature value of each point can be obtained (e.g., the temperature value of each element in an infrared temperature matrix can be obtained). Then, the temperature value of the currently traversed point can be compared with a preset temperature threshold. If the temperature value of the currently traversed point is greater than the preset temperature threshold, the location of the currently traversed point can be determined as an abnormal temperature location and anomaly marking processing can be performed. Conversely, if the temperature value of the currently traversed point is less than or equal to the preset temperature threshold, the location of the currently traversed point can be determined as a normal temperature location, and normal marking processing can be performed or no marking processing can be performed, which is not limited here.

[0057] Furthermore, if abnormal temperature locations exist in the target area, a judgment result (abnormal result) can be obtained, outputting relevant data of the abnormal temperature locations (such as the abnormal start timestamp, abnormal duration, and peak occurrence time of the abnormal intensity) and generating a prompt message. Conversely, if no abnormal temperature locations exist in the target area, a judgment result (normal result) can be obtained. The judgment of abnormal results can also be constrained by the number of abnormal temperature locations. For example, if the number of abnormal temperature locations in the target area exceeds a preset threshold, a judgment result (abnormal result) is obtained indicating that the temperature in the target area is abnormal. The number of abnormal temperature locations can also be correlated with the abnormality level, with a positive correlation between the number of abnormal temperature locations and the abnormality level. For example, an abnormality counter can be preset, and the counter can be incremented (e.g., +1) when an abnormal result is obtained, or it can be incremented every time an abnormal temperature location is obtained; there are no limitations here. Subsequently, the value of the abnormality counter can be used as the abnormality level, or the corresponding abnormality level can be determined based on the value of the abnormality counter (e.g., each abnormality level can correspond to a numerical range), meaning the value of the abnormality counter is also positively correlated with the abnormality level.

[0058] It should also be noted that a single power transmission device in this application may include multiple target areas requiring temperature detection. Therefore, multiple target areas can be detected using a single multispectral camera, or multiple multispectral cameras can be used to detect multiple target areas separately (which may include a one-to-one and / or one-to-many relationship), without limitation here. That is, the acquired multispectral image may include multiple target areas.

[0059] Similarly, this application can also compare the number of target areas (abnormal areas) with abnormal temperatures with a preset threshold number of areas. When the number of abnormal areas exceeds the threshold number, a determination result (abnormal result) is obtained that the temperature of the target area is abnormal. Alternatively, the abnormal counter in the aforementioned example can be used to increment the counter (e.g., +1) if at least one abnormal area exists, or to increment it for each existing abnormal area; no limitation is made here.

[0060] In this strategy, where anomalies are accumulated for each detected anomalous region, if multiple anomalous regions are detected, the distances between them can be analyzed (e.g., distances can be determined by the geometric centers of each region). Two anomalous regions with a distance less than a preset threshold are merged into one anomalous region for reporting. That is, if two anomalous regions exist with a distance less than the preset threshold (perhaps the temperature of one anomalous region is affecting the other), these two anomalies can be considered as a single anomalous result. For example, the anomalous counter can be incremented by 1 (instead of 2).

[0061] Based on the above embodiments, this application embodiment describes the steps of judging the abnormal temperature of a target area according to temperature information using at least one preset abnormal temperature judgment strategy, and obtaining the judgment result corresponding to each preset abnormal temperature judgment strategy. Specifically, the method of this embodiment includes the following steps:

[0062] Based on the temperature of the target area at different detection times, a spatiotemporal temperature field sequence is constructed; the temperature change rate of the target area is determined based on the spatiotemporal temperature field sequence and the detection time; in response to the temperature change rate being greater than the temperature change threshold, the target area is determined to be abnormal, and the determination result is obtained.

[0063] In conjunction with the foregoing embodiments, this embodiment primarily focuses on determining anomalies in temperature change trends. Optionally, when determining whether a temperature change trend is abnormal, this application may base its judgment on all points or regions, or it may base its judgment on points or regions with single-point temperature anomalies identified in the foregoing embodiments; no limitation is imposed here.

[0064] For example, a multispectral camera can continuously detect a target area, obtaining multispectral data of the target area at different detection times, and thus obtaining the temperature of the target area at different detection times (e.g., an infrared temperature matrix). By combining the infrared temperature matrices of the same target area at different detection times in a temporal sequence, a spatiotemporal temperature field sequence of the target area can be constructed, which characterizes the sequence of temperatures formed by each point in the target area at different times. The method for obtaining the temperature of the target area may include, but is not limited to, determining it based on the average temperature of each point in the target area.

[0065] By selecting a detection time segment and the corresponding temperature value (such as the initial temperature value and the final temperature value) in the spatiotemporal temperature field sequence of the target area, the rate of temperature change of the target area during that detection time can be determined (for example, based on the ratio of the temperature difference between the initial temperature value and the final temperature value to the corresponding detection time).

[0066] If the temperature change rate of the target area is greater than the temperature change threshold, the target area can be determined to be abnormal, and a judgment result (abnormal result) can be obtained. Conversely, if the temperature change rate of the target area is less than or equal to the temperature change threshold, the target area can be determined to be normal, and a judgment result (normal result) can be obtained. If the strategy of this embodiment is combined with the aforementioned embodiments, this embodiment can also accumulate the abnormality counter when an abnormal result is obtained.

[0067] In another exemplary embodiment, this application can determine the rate of temperature change not only based on the overall average temperature of the target region, but also based on sliding window and difference calculations. For example, after obtaining the spatiotemporal temperature field sequence, the infrared temperature matrix of the same region (which may be the target region or a part of the target region) is subjected to time-dimensional sliding average processing to obtain the temperature of each window at different times, with a preset time window as the interval (the window size is not limited). The instantaneous rate of temperature change can be obtained by calculating the temperature difference (ΔT) between two adjacent windows and dividing it by the time interval (Δt) between the two windows.

[0068] Furthermore, the rate of temperature change within the three windows closest to the current time can be calculated and compared with a pre-set trend strength index. The trend strength index requires that at least two sub-windows have the same slope direction, and the overall trend's coefficient of variation ≤ 0.3 (temperature change threshold) to be considered normal (e.g., the coefficient of variation can be determined by dividing the standard deviation of the slopes of the three windows by the mean of the absolute values ​​of the slopes of the three windows, etc., which is not limited here); otherwise, it is considered abnormal, resulting in an abnormal result.

[0069] Furthermore, based on the window division, a first-order forward difference operation can be performed on the data within the window to quantify the temperature change rate characteristics.

[0070] Based on the above embodiments, this application embodiment describes the steps prior to determining an anomaly in the target area and obtaining the determination result in response to a temperature change rate exceeding a temperature change threshold. Specifically, the method of this embodiment includes the following steps:

[0071] Obtain the data distribution range of the temperature change rate; analyze the data distribution range according to the preset sliding window to obtain the temperature change threshold.

[0072] It should be noted that the temperature change threshold of this application can be preset and fixed, or can be flexibly adjusted (for example, calibrated according to historical data of temperature change rate, dynamic verification and scenario adaptation, etc.), which is not limited here. The judgment process that supports flexible adjustment of the threshold has stronger generality and can make more accurate judgments in combination with the temperature change situation.

[0073] Combined with the foregoing embodiments, after obtaining the temperature change rates at multiple times, the main distribution interval of the temperature change rate can be determined by analyzing the statistics of multiple historical temperature change rates (such as extreme values, means, standard deviations, etc.). Then, the minimum value of the historical temperature change rate can be iteratively updated according to a preset sliding window and a preset step size.

[0074] For example, assume that the total window length is n sliding windows [T1, T2... Tn], calculate the minimum value Tmin1 in the n sliding windows. When the sliding window slides forward by one step, update the minimum value Tmin2 of the total window [T2, T3... Tn+1] to generate a candidate threshold sequence. Analyze the threshold crossing times in combination with its difference array (that is, the number of times the change rate signal Rt = Tt - Tt-1 crosses the threshold. Among them, Rt > Tmin and Rt-1 < Tmin is an upward crossing, and Rt < Tmin and Rt-1 > Tmin is a downward crossing). By screening the crossing times that meet the preset conditions (such as meeting the requirements of false alarm rate and missed alarm rate), the optimal threshold is determined, that is, the available temperature change threshold is obtained. Among them, if the threshold is too high (the number of threshold crossings is small), the missed alarm rate will increase, and if the threshold is too low (the number of threshold crossings is large), the false alarm rate will increase.

[0075] Reference can be made to Figure 4 as shown in Figure 4 is a flowchart of an exemplary determination of temperature change rate in the temperature detection method based on multi-spectral images of this application. Combining the temperature change threshold calibration process of this embodiment with the spatio-temporal temperature field sequence processing process of the foregoing embodiment, the characteristic information of the temperature change rate can be obtained.

[0076] On the basis of the above embodiments, the embodiments of this application illustrate the steps of judging the abnormal temperature of the target area according to at least one preset abnormal temperature judgment strategy based on the temperature information and obtaining the judgment results corresponding to each preset abnormal temperature judgment strategy. Specifically, the method of this embodiment includes the following steps:

[0077] Determine the abnormal temperature position in the target area according to the temperature information; perform temperature diffusion analysis on the abnormal temperature position to obtain the diffusion analysis result of the adjacent area of the abnormal temperature position; in response to the diffusion analysis result indicating that the temperature change rate of the adjacent area is greater than the temperature change threshold, determine that the target area is abnormal and obtain the judgment result.

[0078] In conjunction with the aforementioned embodiments, this embodiment mainly focuses on anomaly detection of temperature diffusion.

[0079] Referring to the method in the foregoing embodiments, the locations of abnormal temperatures in the target area (such as the location of a single-point temperature anomaly and / or the location of an abnormal temperature change trend) can be determined based on the acquired temperature information. Spatial diffusion analysis is then performed on these abnormal temperature locations (temperature diffusion is triggered when the temperature change rate of the adjacent area of ​​an abnormal temperature location exceeds a set threshold), thus determining that the target area is abnormal. Furthermore, it is possible to selectively determine that adjacent areas are abnormal. The range of adjacent areas is not limited in this application; the maximum range can cover the entire image area. For example, when analyzing an abnormal temperature location, its neighborhood can be a 4-neighborhood or an 8-neighborhood, etc., and is not limited here.

[0080] When a temperature diffusion trigger is determined, abnormal area location data can be output (such as the pixel coordinates, physical location label, and spatial topological relationship with adjacent areas). Temperature diffusion parameters of the abnormal area can also be output, such as the diffusion rate quantization value (the number of adjacent areas whose temperature exceeds a threshold) and the diffusion influence radius (i.e., the spatial range of influence of the abnormal area, such as the farthest distance between the abnormal area and its adjacent areas that are determined to be abnormal).

[0081] It should also be noted that this embodiment can also be combined with the method of the foregoing embodiments. After the anomaly is judged and the anomaly result is obtained, the anomaly counter can be incremented.

[0082] Based on the above embodiments, this application embodiment describes the steps for determining the temperature detection result of the target area according to the judgment result corresponding to each preset abnormal temperature judgment strategy. Specifically, the method of this embodiment includes the following steps:

[0083] Obtain the judgment results corresponding to each preset abnormal temperature judgment strategy, including abnormal results; determine the temperature abnormality level of the target area based on the number of abnormal results; generate temperature detection results corresponding to the temperature abnormality level.

[0084] Referring to the foregoing embodiments, this embodiment can execute various judgment strategies provided in the foregoing embodiments to obtain the judgment results of each strategy. The execution order of the judgment strategies is not limited. These judgment results may include abnormal results and / or normal results.

[0085] For example, reference can be made to, such as Figure 5 As shown, Figure 5This is an overall schematic diagram of the temperature detection method based on multispectral images in this application, which determines the temperature detection result according to multiple strategies. For ease of understanding, the execution process of this embodiment is a general description, and the specific implementation steps can be referred to the methods of the foregoing embodiments. For example, after obtaining the preprocessed multispectral time-series data, it can be determined whether there is a single point in the target area whose temperature exceeds the temperature threshold. If so, the abnormal temperature location is determined, an abnormal result is obtained, and the abnormal counter is incremented by 1. Then, a temperature change trend analysis can be performed on the abnormal temperature location to determine whether the temperature change rate is greater than the temperature change threshold. If so, an abnormal result is obtained, and the abnormal counter is incremented by 1. Then, it is also possible to determine whether the temperature change rate of the adjacent area of ​​the abnormal temperature location is greater than the temperature change threshold (the temperature change threshold can be the same as or different from the temperature change threshold in the previous judgment process, which is not limited here). If so, an abnormal result is obtained, and the abnormal counter is incremented by 1 again. Based on these judgment strategies, more judgment strategies can be set as needed according to the actual application scenario, which will not be elaborated here. When the selected judgment strategy is completed, or when certain preset conditions are met during the execution of a judgment strategy, the corresponding temperature anomaly level can be determined based on the number of anomalies in the current anomaly counter, and a temperature detection result corresponding to the temperature anomaly level can be generated.

[0086] For ease of explanation, this embodiment divides temperature anomalies into three levels (low, medium, and high). In practical applications, other levels (such as numerical classifications) can be set as needed, which is not limited here. Different temperature anomaly levels correspond to different temperature detection results, and the measures to deal with different temperature anomaly levels are also different. For example, a low-level temperature anomaly can be addressed by sending an anomaly notification to the multispectral camera backend and displaying specific anomaly information; a medium-level temperature anomaly can be addressed by simultaneously activating a preset self-diagnostic program in the power transmission equipment, which can be configured as needed and will not be elaborated here. The self-diagnostic program can determine the location and cause of the anomaly and attempt self-repair; a high-level temperature anomaly can be addressed by activating an audible and visual alert and simultaneously notifying the relevant maintenance personnel.

[0087] It should also be noted that, to avoid accumulating abnormal results from multiple separate temperature measurements, the abnormality counter can be reset at certain stages. For example, after executing all the judgment strategies selected in a single temperature measurement process, the abnormality counter can be reset to zero.

[0088] Based on the above embodiments, this application embodiment describes the steps prior to performing temperature detection processing on the target region in the acquired multispectral image to obtain the temperature information of the target region. Specifically, the method of this embodiment includes the following steps:

[0089] Acquire a multispectral image of the initial region; perform target detection processing on the multispectral image to obtain the target region in the initial region.

[0090] Referring to the foregoing embodiments, the multispectral camera of this application can acquire images of an initial region, which includes the target region. Subsequently, target detection or segmentation processing can be performed using data such as visible light images to obtain the target region from the initial region.

[0091] Based on the above embodiments, this application embodiment exemplarily describes the steps of performing temperature detection processing on a target region in an acquired multispectral image to obtain temperature information of the target region. The temperature information includes an infrared temperature matrix and a spatiotemporal temperature field sequence. Specifically, the method of this embodiment includes the following steps:

[0092] Infrared features of the target region are extracted from the multispectral image to obtain the infrared temperature matrix; the spatiotemporal temperature field sequence of the target region is constructed based on the infrared temperature matrix of the same target region obtained at different times.

[0093] Referring to the foregoing embodiments, after acquiring a multispectral image, infrared features of the target region can be extracted from the multispectral image to generate an infrared temperature matrix. Then, based on the infrared temperature matrices of the same target region acquired at different times, a spatiotemporal temperature field sequence of the target region can be constructed, which is equivalent to a type of temperature time-series data.

[0094] In summary, the method presented in this application enables early detection and predictive maintenance of micro-local temperature anomalies in power transmission equipment through continuous high-precision detection using a fixed multispectral camera and multispectral time-series fusion analysis. The fixed design, combined with the collaborative calibration of a high-resolution infrared sensor and a visible / infrared module, utilizes continuous time-series data to construct a dynamic temperature model. Deep learning algorithms capture abnormal trends in the rate of temperature change in micro-regions, and a lightweight edge-end model is used to generate real-time alerts and maintenance suggestions. This application overcomes the limitations of intermittent detection in traditional UAV inspections and upgrades fault warnings from "threshold triggering" to "trend prediction" through predictive analysis, significantly improving the proactive operation and maintenance capabilities of the power grid.

[0095] It should be further noted that the execution entity of the temperature detection method based on multispectral images can be a temperature detection device based on multispectral images. For example, the temperature detection method based on multispectral images can be executed by a terminal device, a server, or other processing devices. The terminal device can be a user equipment (UE), computer, mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, in-vehicle device, wearable device, etc. In some possible implementations, the temperature detection method based on multispectral images can be implemented by a processor calling computer-readable instructions stored in memory.

[0096] Figure 6 This is a block diagram illustrating a temperature detection device based on multispectral images, as shown in an exemplary embodiment of this application. Figure 6 As shown, the exemplary temperature detection device 600 based on multispectral images includes: a temperature detection module 610, an anomaly detection module 620, and a result determination module 630. Specifically:

[0097] The temperature detection module 610 is used to perform temperature detection processing on the target area in the acquired multispectral image to obtain the temperature information of the target area.

[0098] The anomaly detection module 620 is used to detect abnormal temperatures in the target area based on temperature information using at least one preset abnormal temperature detection strategy, and to obtain the detection result corresponding to each preset abnormal temperature detection strategy.

[0099] The result determination module 630 is used to determine the temperature detection result of the target area based on the judgment result corresponding to each preset abnormal temperature judgment strategy.

[0100] In this exemplary temperature detection device based on multispectral images, a multispectral image containing the target region is acquired by a multispectral camera, thereby enabling the determination of the temperature information of the target region based on the multispectral data of the multispectral image. Then, whether the temperature information of the target region is abnormal can be determined according to at least one preset abnormal temperature judgment strategy, obtaining the judgment result corresponding to each preset abnormal temperature judgment strategy. The judgment results of different strategies may be the same or different, and the judgment results may include normal results and / or abnormal results. Therefore, the judgment results corresponding to each preset abnormal temperature judgment strategy can be combined and analyzed to determine the temperature detection result of the target region, achieving efficient and accurate temperature detection.

[0101] It should be noted that the apparatus and method provided in the above embodiments belong to the same concept, and the specific ways in which each module and unit performs operations have been described in detail in the method embodiments, and will not be repeated here. In practical applications, the apparatus provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the apparatus can be divided into different functional modules to complete all or part of the functions described above, and this is not a limitation.

[0102] The functions of each module can be found in the embodiments of the temperature detection method based on multispectral images, and will not be repeated here.

[0103] Please see Figure 7 , Figure 7 This is a schematic diagram of an embodiment of the electronic device of this application. The electronic device 100 includes a memory 101 and a processor 102. The processor 102 is used to execute program instructions stored in the memory 101 to implement the steps in any of the above embodiments of the temperature detection method based on multispectral images. In a specific implementation scenario, the electronic device 100 may include, but is not limited to, a microcomputer or a server. In addition, the electronic device 100 may also include mobile devices such as laptops and tablets, which are not limited here.

[0104] Specifically, processor 102 controls itself and memory 101 to implement the steps in any of the above-described embodiments of the temperature detection method based on multispectral images. Processor 102 can also be referred to as a CPU (Central Processing Unit). Processor 102 may be an integrated circuit chip with signal processing capabilities. Processor 102 can also be a general-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor can be a microprocessor or any conventional processor. Furthermore, processor 102 can be implemented using integrated circuit chips.

[0105] In this exemplary electronic device, a multispectral image containing a target region is acquired by a multispectral camera, thereby enabling the determination of temperature information of the target region based on the multispectral data of the multispectral image. Then, whether the temperature information of the target region is abnormal can be determined according to at least one preset abnormal temperature judgment strategy, obtaining the judgment result corresponding to each preset abnormal temperature judgment strategy. The judgment results of different strategies may be the same or different, and the judgment results may include normal results and / or abnormal results. Therefore, the judgment results corresponding to each preset abnormal temperature judgment strategy can be combined and analyzed to determine the temperature detection result of the target region, achieving efficient and accurate temperature detection.

[0106] Please see Figure 8 , Figure 8 This is a schematic diagram of a computer-readable storage medium according to an embodiment of the present application. The computer-readable storage medium 110 stores program instructions 111 that can be executed by a processor. The program instructions 111 are used to implement the steps in any of the above embodiments of the temperature detection method based on multispectral images.

[0107] In this exemplary storage medium, by running program instructions within the storage medium, a multispectral image containing the target region is acquired by a multispectral camera. This allows the temperature information of the target region to be determined based on the multispectral data of the multispectral image. Then, whether the temperature information of the target region is abnormal can be determined according to at least one preset abnormal temperature judgment strategy, yielding the judgment result corresponding to each preset abnormal temperature judgment strategy. The judgment results of different strategies may be the same or different, and the judgment results may include normal results and / or abnormal results. Therefore, the judgment results corresponding to each preset abnormal temperature judgment strategy can be combined and analyzed to determine the temperature detection result of the target region, achieving efficient and accurate temperature detection.

[0108] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0109] The description of the various embodiments above tends to emphasize the differences between the various embodiments. The similarities or similarities between them can be referred to, and for the sake of brevity, they will not be repeated here.

[0110] In the several embodiments provided in this application, it should be understood that the disclosed methods and apparatus can be implemented in other ways. For example, the apparatus implementations described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of devices or units may be electrical, mechanical, or other forms.

[0111] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A temperature detection method based on multispectral images, characterized in that, The method includes: Temperature detection processing is performed on the target region in the acquired multispectral image to obtain the temperature information of the target region; Based on the temperature information, the target area is judged for abnormal temperature using at least one preset abnormal temperature judgment strategy, and the judgment result corresponding to each preset abnormal temperature judgment strategy is obtained. The temperature detection result of the target area is determined based on the judgment result corresponding to each preset abnormal temperature judgment strategy.

2. The method according to claim 1, characterized in that, The step of judging the abnormal temperature of the target area based on the temperature information using at least one preset abnormal temperature judgment strategy, and obtaining the judgment result corresponding to each preset abnormal temperature judgment strategy, includes: Based on the temperature information and a preset temperature threshold, it is determined whether there is an abnormal temperature location in the target area, where the temperature information of the abnormal temperature location is greater than the preset temperature threshold. In response to the presence of the abnormal temperature location in the target area, the abnormal temperature location is marked as abnormal to obtain the determination result.

3. The method according to claim 1, characterized in that, Based on the temperature information, the target area is subjected to anomaly temperature judgment using at least one preset anomaly temperature judgment strategy, and the judgment result corresponding to each preset anomaly temperature judgment strategy is obtained, including: Based on the temperature of the target area at different detection times, a spatiotemporal temperature field sequence is constructed; The temperature change rate of the target region is determined based on the spatiotemporal temperature field sequence and the detection time. In response to the temperature change rate being greater than the temperature change threshold, the target area is determined to be abnormal, and the determination result is obtained.

4. The method according to claim 3, characterized in that, Before determining that the target area is abnormal in response to the temperature change rate being greater than a temperature change threshold, and obtaining the determination result, the method further includes: Obtain the data distribution range of the temperature change rate; The temperature change threshold is obtained by analyzing the data distribution range according to a preset sliding window.

5. The method according to claim 1, characterized in that, The step of judging the abnormal temperature of the target area based on the temperature information using at least one preset abnormal temperature judgment strategy, and obtaining the judgment result corresponding to each preset abnormal temperature judgment strategy, includes: The location of abnormal temperatures in the target area is determined based on the temperature information; Temperature diffusion analysis is performed on the location of the abnormal temperature to obtain the diffusion analysis results of the adjacent region of the location of the abnormal temperature. In response to the diffusion analysis results indicating that the temperature change rate of the adjacent region is greater than the temperature change threshold, the target region is determined to be abnormal, and the determination result is obtained.

6. The method according to claim 1, characterized in that, The step of determining the temperature detection result of the target area based on the judgment results corresponding to each preset abnormal temperature judgment strategy includes: Obtain the judgment results corresponding to each preset abnormal temperature judgment strategy, wherein the judgment results include abnormal results; The temperature anomaly level of the target area is determined based on the number of the abnormal results. Generate temperature detection results corresponding to the stated temperature anomaly level.

7. The method according to claim 1, characterized in that, Before performing temperature detection processing on the target region in the acquired multispectral image to obtain the temperature information of the target region, the method further includes: Acquire a multispectral image of the initial region; The multispectral image is processed for target detection to obtain the target region in the initial region.

8. The method according to claim 7, characterized in that, The temperature information includes an infrared temperature matrix and a spatiotemporal temperature field sequence. The step of performing temperature detection processing on the target region in the acquired multispectral image to obtain the temperature information of the target region includes: Infrared features of the target region are extracted from the multispectral image to obtain the infrared temperature matrix; The spatiotemporal temperature field sequence of the target region is constructed based on the infrared temperature matrix of the same target region obtained at different times.

9. An electronic device, characterized in that, The method includes a memory and a processor, the processor being configured to execute program instructions stored in the memory to implement the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having program instructions stored thereon, characterized in that, When the program instructions are executed by the processor, they implement the method described in any one of claims 1 to 8.

Images