A method, device, and program for remote temperature detection of power equipment.
By performing multiple denoising processes on infrared images of power equipment, including initial denoising, wavelet transform, and wavelet coefficient adjustment, the problem of infrared image noise affecting detection accuracy is solved, and high accuracy in temperature detection of power equipment is achieved.
Patent Information
- Application Number
- CN202410555617.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-07
- Publication Date
- 2026-11-10
AI Technical Summary
In existing temperature detection of power equipment, infrared images are significantly affected by noise, resulting in low detection accuracy.
A multi-stage denoising process is employed, including initial denoising, wavelet transform, zeroing of noise wavelet coefficients, and adjustment of effective signal wavelet coefficients. Through multiple denoising processes, the optimal denoised infrared image is obtained to improve the accuracy of temperature detection.
It effectively removes infrared image noise, improves the accuracy of temperature detection in power equipment, and is suitable for large-scale application and promotion in the field of power equipment monitoring.
Smart Images

Figure CN123065610A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power equipment monitoring technology, specifically relating to a remote temperature detection method, device, and program product for power equipment. Background Technology
[0002] With the increasing electricity consumption in residential and industrial settings, the load on power equipment is constantly increasing. Power equipment carries a high load of power transmission, and during the power transmission process, it is not only directly affected by electricity and heat, but its performance is also gradually weakened due to factors such as increased usage time or adverse environments. Therefore, high load inevitably leads to heat generation in power equipment, resulting in an increasing number of power equipment failures caused by abnormal temperatures.
[0003] Timely detection of abnormal temperatures in power equipment and the implementation of corresponding control measures play a significant role in ensuring the safe and stable operation of power equipment. Currently, with the rapid development of infrared imaging technology, related technologies are increasingly being applied to the detection of power equipment, solving numerous thermal failure problems. Infrared thermal imaging technology detects thermal radiation from objects with temperatures above absolute zero. It features non-contact operation, high sensitivity, immunity to electric field interference, speed, and accuracy. Applying infrared temperature measurement technology to the online monitoring and inspection of power distribution equipment can predict and detect equipment problems in a timely manner, effectively enhancing equipment reliability and reducing equipment accidents. It has become one of the important tools for ensuring the reliable operation of power systems.
[0004] However, infrared images are characterized by high noise and low contrast due to the influence of ambient noise and instrument noise during the generation process. Existing temperature monitoring processes for power equipment only use simple denoising methods (such as mean filtering) to denoise the infrared images, which has limited denoising effect. Therefore, the detection results are affected by noise, resulting in low detection accuracy. Based on this, how to provide a remote temperature detection method for power equipment that can effectively remove the influence of noise and thus improve the accuracy of temperature detection has become an urgent problem to be solved. Summary of the Invention
[0005] The purpose of this invention is to provide a remote temperature detection method, device, and program product for power equipment, in order to solve the problem of low detection accuracy caused by ineffective noise reduction in the existing technology.
[0006] To achieve the above objectives, the present invention adopts the following technical solution:
[0007] Firstly, a method for remote temperature detection of power equipment is provided, including:
[0008] Acquire an infrared image of a specified part of a power equipment, and perform initial denoising processing on the infrared image of the specified part to obtain an initial denoised infrared image;
[0009] The initial denoised infrared image is subjected to wavelet transform processing to obtain the wavelet transformed image signal corresponding to the initial denoised infrared image, and the wavelet coefficient set is formed by using the wavelet coefficients corresponding to each transform scale in the wavelet transformed image signal.
[0010] Wavelet coefficients corresponding to image noise are selected from the set of wavelet coefficients and used as the first wavelet coefficients. The remaining wavelet coefficients in the set of wavelet coefficients are used as the second wavelet coefficients.
[0011] The first wavelet coefficients in the wavelet-transformed image signal are set to zero, and the second wavelet coefficients in the wavelet-transformed image signal are adjusted to the optimal image reconstruction coefficients, so that a denoised wavelet-transformed signal is obtained after the second wavelet coefficients are adjusted.
[0012] The denoised wavelet transform signal is subjected to wavelet reconstruction processing to obtain the optimal denoised infrared image;
[0013] Based on the optimal denoised infrared image, the temperature detection result of a specified part of the power equipment is obtained.
[0014] Based on the above disclosure, after acquiring an infrared image of a designated part of the power equipment, this invention employs a multi-stage denoising process. First, the infrared image undergoes initial denoising to obtain an initial denoised infrared image. Then, wavelet transform processing is performed on the initial denoised infrared image, and the wavelet coefficients obtained from the transform are used for subsequent denoising. Specifically, this invention first filters out wavelet coefficients belonging to image noise and sets them to zero, thus removing the image signal that belongs to noise obtained from the decomposition. Simultaneously, after the aforementioned two denoising processes, the remaining image signal still contains a small amount of noise. Therefore, based on the aforementioned two denoising processes, this invention performs a third denoising process, adjusting the remaining wavelet coefficients (i.e., the wavelet coefficients of the usable signal) to the optimal image reconstruction coefficients (i.e., restoring them to the true, clean wavelet signal). Next, image reconstruction is performed on the wavelet transform signal after the aforementioned coefficient adjustment to obtain the optimal denoised infrared image. Finally, based on the aforementioned optimal denoised infrared image, the temperature detection result of the power equipment can be obtained.
[0015] Through the above design, this invention sequentially performs initial denoising processing on the infrared image of the power equipment, zeroing the wavelet coefficients of the noise signal, and adjusting the wavelet coefficients of the effective signal. Thus, after the aforementioned multiple denoising processes, the noise signal in the original image can be effectively removed, resulting in an optimal denoised infrared image. Then, based on this optimal denoised infrared image, the temperature detection result of the power equipment can be obtained. Based on this, this invention can effectively reduce the impact of noise on temperature detection, thereby improving the accuracy of temperature detection. Therefore, it is very suitable for large-scale application and promotion in the field of power equipment monitoring.
[0016] In one possible design, the infrared image of the specified region is subjected to initial denoising processing to obtain an initial denoised infrared image, including:
[0017] The infrared image is subjected to grayscale transformation to obtain a grayscale image;
[0018] For any pixel in the grayscale image, obtain the first neighborhood region centered on the pixel.
[0019] The first neighborhood region is divided into multiple sub-neighborhood regions, wherein each sub-neighborhood region includes any of the aforementioned pixels;
[0020] Determine the grayscale contrast threshold for each sub-neighborhood region and select the smallest grayscale contrast threshold;
[0021] Calculate the average gray value of all pixels in the sub-neighborhood region corresponding to the minimum gray-level contrast threshold;
[0022] The grayscale value of any pixel is updated to the grayscale mean, and after the grayscale values of all pixels in the grayscale image have been updated, the initial denoised infrared image is obtained.
[0023] In one possible design, the grayscale contrast thresholds for each sub-neighborhood region are determined, including:
[0024] For any sub-neighborhood region, the grayscale contrast threshold of the sub-neighborhood region is calculated using the following formula (1);
[0025]
[0026] In the above formula (1), R represents the grayscale contrast threshold of any sub-neighborhood region, x i This represents the grayscale value of the i-th pixel in any of the sub-neighborhood regions. denoted as the average grayscale value of all pixels in any sub-neighborhood region, m represents the total number of pixels in any sub-neighborhood region, and n represents the contrast coefficient.
[0027] In one possible design, wavelet coefficients corresponding to image noise are selected from the set of wavelet coefficients as the first wavelet coefficients, including:
[0028] For the k-th wavelet coefficient in the wavelet coefficient set, relevant wavelet coefficients of the k-th wavelet coefficient are selected from the wavelet coefficient set. The wavelet transform direction of the relevant wavelet coefficients is the same as that of the k-th wavelet coefficient. The transform scale of the relevant wavelet coefficients is j+1, and j is the transform scale of the k-th wavelet coefficient.
[0029] Based on the k-th wavelet coefficient and the related wavelet coefficient of the k-th wavelet coefficient, the matching value of the wavelet coefficient belonging to the image noise of the k-th wavelet coefficient is calculated;
[0030] Determine whether the matching value is less than the matching threshold corresponding to the j-th transformation scale;
[0031] If so, then the kth wavelet coefficient is taken as a first wavelet coefficient;
[0032] Increment k by 1, and re-select the relevant wavelet coefficients of the k-th wavelet coefficient from the wavelet coefficient set until k equals K. Then, select all wavelet coefficients belonging to image noise from the wavelet coefficient set. The initial value of k is 1, and K is the total number of wavelet coefficients in the wavelet coefficient set.
[0033] In one possible design, the second wavelet coefficients in the wavelet-transformed image signal are adjusted to the optimal image reconstruction coefficients, including:
[0034] For any second wavelet coefficient in the wavelet transform image signal, obtain multiple adjacent wavelet coefficients of the second wavelet coefficient, wherein the multiple adjacent wavelet coefficients are the second wavelet coefficients of each pixel point located on the diagonal within a second neighborhood region centered on the pixel point corresponding to the second wavelet coefficient.
[0035] Using multiple adjacent wavelet coefficients, the noise variance of any second wavelet coefficient is calculated, and based on the noise variance and the multiple adjacent wavelet coefficients, the local variance of any second wavelet coefficient is calculated.
[0036] Based on the local variance and noise variance of any of the second wavelet coefficients, the second wavelet coefficients are adjusted to the corresponding optimal image reconstruction coefficients.
[0037] In one possible design, the noise variance of any second wavelet coefficient is calculated using multiple adjacent wavelet coefficients, including:
[0038] The noise variance of any second wavelet coefficient is calculated according to the following formula (2);
[0039]
[0040] In the above formula (2), Z represents the noise variance of any of the second wavelet coefficients, Z represents the wavelet sequence composed of the multiple adjacent wavelet coefficients, and median represents the median function;
[0041] Accordingly, based on the noise variance and the plurality of adjacent wavelet coefficients, the local variance of any second wavelet coefficient is calculated, which includes:
[0042] The local variance of any second wavelet coefficient is calculated according to the following formula (3);
[0043]
[0044] In the above formula (3), Let z represent the local variance of any of the second wavelet coefficients. a Let A represent the a-th adjacent wavelet coefficient among the plurality of adjacent wavelet coefficients, and let A represent the total number of adjacent wavelet coefficients.
[0045] In one possible design, based on the optimal denoised infrared image, the temperature detection result of a specified location on the power equipment is obtained, including:
[0046] The optimal denoised infrared image is subjected to grayscale mapping processing to obtain a grayscale mapped infrared image;
[0047] The first temperature value corresponding to the maximum gray value and the second temperature value corresponding to the minimum gray value are determined from the gray-scale mapped infrared image.
[0048] Based on the first temperature value, the second temperature value, and the grayscale value of each pixel in the grayscale-mapped infrared image, the temperature value corresponding to each pixel in the grayscale-mapped infrared image is determined.
[0049] Based on the temperature value corresponding to each pixel, the temperature detection result of a specified part of the power equipment is obtained.
[0050] Secondly, a remote temperature detection device for power equipment is provided, comprising:
[0051] The first denoising unit is used to acquire an infrared image of a specified part of the power equipment and perform initial denoising processing on the infrared image of the specified part to obtain an initial denoised infrared image.
[0052] The wavelet transform unit is used to perform wavelet transform processing on the initial denoised infrared image to obtain the wavelet transformed image signal corresponding to the initial denoised infrared image, and to form a wavelet coefficient set using the wavelet coefficients corresponding to each transform scale in the wavelet transformed image signal.
[0053] The second denoising unit is used to select wavelet coefficients corresponding to image noise from the wavelet coefficient set as first wavelet coefficients, and use the remaining wavelet coefficients in the wavelet coefficient set as second wavelet coefficients.
[0054] The second denoising unit is also used to set the first wavelet coefficients in the wavelet-transformed image signal to zero, and to adjust the second wavelet coefficients in the wavelet-transformed image signal to the optimal image reconstruction coefficients, so as to obtain a denoised wavelet-transformed signal after adjusting the second wavelet coefficients.
[0055] An image reconstruction unit is used to perform wavelet reconstruction processing on the denoised wavelet transform signal to obtain the optimal denoised infrared image.
[0056] The temperature detection unit is used to obtain the temperature detection result of a specified part of the power equipment based on the optimal denoised infrared image.
[0057] Thirdly, another remote temperature detection device for power equipment is provided. Taking the device as an electronic device as an example, it includes a memory, a processor, and a transceiver that are connected in sequence. The memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the remote temperature detection method for the power equipment as described in the first aspect or any possible design of the first aspect.
[0058] Fourthly, a storage medium is provided, on which instructions are stored, which, when executed on a computer, perform the remote temperature detection method for the power equipment as described in the first aspect or any possible design of the first aspect.
[0059] Fifthly, a computer program product containing instructions is provided, which, when executed on a computer, cause the computer to perform the remote temperature detection method for the power equipment as described in the first aspect or any possible design of the first aspect.
[0060] Beneficial effects:
[0061] (1) The present invention sequentially performs initial denoising processing on the infrared image of the power equipment, zeroing the wavelet coefficients of the noise signal, and adjusting the wavelet coefficients of the effective signal. Thus, after the aforementioned multiple denoising processes, the noise signal in the original image can be effectively removed, and the optimal denoised infrared image can be obtained. Then, based on the optimal denoised infrared image, the temperature detection result of the power equipment can be obtained. Based on this, the present invention can effectively reduce the influence of noise on temperature detection, thereby improving the accuracy of temperature detection. Therefore, it is very suitable for large-scale application and promotion in the field of power equipment monitoring. Attached Figure Description
[0062] Figure 1 This is a flowchart illustrating the steps of a remote temperature detection method for power equipment provided in an embodiment of the present invention.
[0063] Figure 2 This is a schematic diagram of the structure of remote temperature detection for power equipment provided in an embodiment of the present invention;
[0064] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation
[0065] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in conjunction with the accompanying drawings and descriptions of the embodiments or the prior art. Obviously, the following description of the structure of the accompanying drawings is only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. It should be noted that the description of these embodiments is for the purpose of helping to understand the present invention, but does not constitute a limitation of the present invention.
[0066] It should be understood that although the terms first, second, etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit, without departing from the scope of the exemplary embodiments of the invention.
[0067] It should be understood that the term "and / or" that may appear in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" that may appear in this document describes another relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " that may appear in this document generally indicates that the related objects before and after it are in an "or" relationship.
[0068] Example:
[0069] See Figure 1 As shown, the remote temperature detection method for power equipment provided in this embodiment first uses multiple neighborhood regions of each pixel in the infrared image to smooth the grayscale values of each pixel, thereby achieving initial image denoising. Then, wavelet transform is performed on the initial denoised infrared image, and the obtained wavelet coefficients are used for subsequent denoising processing. Specifically, the wavelet coefficients belonging to image noise are first set to zero to remove image noise signals. Then, the remaining wavelet coefficients are restored to the wavelet coefficients of the true signal. Based on this, noise contained in the remaining effective signal can be removed. Next... By performing wavelet reconstruction on the wavelet transform signal after coefficient adjustment, the optimal denoised infrared image can be obtained. Finally, based on the optimal denoised infrared image, the temperature detection result of the power equipment can be obtained. Thus, this method performs multiple denoising processes on the image before temperature detection, which can effectively reduce the impact of noise on temperature detection and improve the accuracy of temperature detection. Therefore, it is very suitable for large-scale application and promotion in the field of power equipment monitoring. For example, this method can be run on the power equipment monitoring end. Optionally, the power equipment monitoring end can be, but is not limited to, a personal computer (PC), tablet computer, or smartphone. It is understood that the aforementioned execution subject does not constitute a limitation on the embodiments of this application. Accordingly, the operation steps of this method can be, but are not limited to, the steps S1 to S6 below.
[0070] S1. Acquire an infrared image of a designated part of the power equipment, and perform initial denoising processing on the infrared image of the designated part to obtain an initial denoised infrared image; In this embodiment, the designated part of the power equipment may be, but is not limited to, an insulating part or an interface connection part (such as a hidden connection part of the power equipment). Of course, the designated part to be monitored can be specifically set according to actual use, and is not limited to the above example; At the same time, an infrared camera can be used to collect the infrared image of the aforementioned power equipment.
[0071] After obtaining the infrared image of a specified part of the power equipment, it can be subjected to initial denoising processing. In this embodiment, a multi-neighbor grayscale processing method is used to perform initial denoising processing on the infrared image. The process can be, but is not limited to, the steps S11 to S16 below.
[0072] S11. Perform grayscale transformation processing on the infrared image to obtain a grayscale image; in specific applications, after transforming the infrared image into a grayscale image, multiple neighborhoods corresponding to each pixel in the grayscale image can be used to perform grayscale smoothing processing on each pixel; wherein, taking any pixel in the grayscale image as an example, the grayscale smoothing process can be specifically described as follows, but is not limited to the steps S12 to S16 below.
[0073] S12. For any pixel in the grayscale image, obtain a first neighborhood region centered on the pixel. In specific implementations, for example, but not limited to, a 3×3 neighborhood centered on the pixel can be used as the first neighborhood region. After obtaining the first neighborhood region, it can be divided into multiple sub-neighborhood regions so that grayscale smoothing can be performed on the pixel based on the grayscale values of the pixels in each sub-neighborhood region. The neighborhood division process can be, but is not limited to, the steps shown in S13 below.
[0074] S13. Divide the first neighborhood region into multiple sub-neighborhood regions, where each sub-neighborhood region includes any of the aforementioned pixel points. In a specific implementation, based on the aforementioned 3×3 neighborhood, it is further elaborated that the pixel points (arranged in rows) of the first neighborhood region are as follows: x1, x2, x3, x4, x0, x5, x6, x7, and x8, where any of the aforementioned pixel points is x0. Then, the first neighborhood region can be divided into 9 sub-neighborhood regions, where a sub-neighborhood containing four pixel points is a sub-neighborhood region consisting of four pixels (e.g., x1, x2, x4, and x0). The sub-neighborhoods are defined as follows: a sub-neighborhood consisting of x2, x3, x0, and x5; a sub-neighborhood consisting of x4, x0, x6, and x7; and a sub-neighborhood consisting of x0, x5, x7, and x8. There are four sub-neighborhoods containing six pixels (such as a sub-neighborhood consisting of x1, x2, x3, x4, x0, and x5), and a sub-neighborhood containing nine pixels. After the sub-neighborhoods are divided, grayscale smoothing can be performed on any pixel based on the divided sub-neighborhoods. The grayscale smoothing process can be, but is not limited to, the steps S14 to S16 below.
[0075] S14. Determine the grayscale contrast threshold of each sub-neighborhood region and select the smallest grayscale contrast threshold; in specific implementation, take any sub-neighborhood region as an example to illustrate the calculation process of the grayscale contrast threshold; for any sub-neighborhood region, the following formula (1) can be used to calculate the grayscale contrast threshold of any sub-neighborhood region.
[0076]
[0077] In the above formula (1), R represents the grayscale contrast threshold of any sub-neighborhood region, x i This represents the grayscale value of the i-th pixel in any of the sub-neighborhood regions. represents the average grayscale value of all pixels in any sub-neighborhood region, m represents the total number of pixels in any sub-neighborhood region, and n represents the contrast coefficient; in this embodiment, n is the total number of pixels in any sub-neighborhood region. For example, when any sub-neighborhood region contains 4 pixels, n is 4, and when it contains 6 pixels, n is 6.
[0078] Thus, the grayscale contrast threshold of each sub-neighborhood region can be calculated using the aforementioned formula (1). Then, the smallest grayscale contrast threshold can be selected from multiple grayscale contrast thresholds. Finally, the average grayscale value of all pixels in the sub-neighborhood region corresponding to the smallest grayscale contrast threshold can be used as the smoothing value of any pixel. The aforementioned process can be, but is not limited to, the steps S15 and S16 below.
[0079] S15. Calculate the average gray value of all pixels in the sub-neighborhood region corresponding to the minimum gray-level contrast threshold.
[0080] S16. Update the gray value of any pixel to the gray average value, and after updating the gray values of all pixels in the grayscale image, obtain the initial denoised infrared image.
[0081] Based on the aforementioned steps S11 to S16, grayscale smoothing can be performed on each pixel in the grayscale image, thereby filtering out noise points in the infrared image and preserving image detail information to obtain an initial denoised infrared image. After the initial denoising of the infrared image is completed, subsequent denoising processing is required. In this embodiment, wavelet transform is used to perform secondary and tertiary denoising on the initial denoised infrared image. The aforementioned secondary and tertiary denoising processes can be, but are not limited to, as shown in steps S2 to S5 below.
[0082] S2. Perform wavelet transform processing on the initial denoised infrared image to obtain the wavelet transformed image signal corresponding to the initial denoised infrared image, and use the wavelet coefficients corresponding to each transform scale in the wavelet transformed image signal to form a wavelet coefficient set. In specific applications, for example, but not limited to, using the sym4 wavelet function to perform wavelet transform on the initial denoised infrared image, and the transform scale can be set to 3; in this way, the decomposed wavelet transformed signal contains the wavelet coefficients of each pixel in each initial denoised infrared image at three scales and in different decomposition directions; then, the wavelet coefficients at each level (i.e., different decomposition scales) in the wavelet transformed signal can be used to form a wavelet coefficient set, so that secondary and tertiary denoising processing can be performed based on the wavelet coefficient set.
[0083] Optionally, in this embodiment, wavelet coefficients belonging to image noise are first selected from the wavelet coefficient set and then set to zero; at this time, secondary denoising can be completed; then, the remaining wavelet coefficients are adjusted so as to further remove image noise on the basis of secondary denoising; wherein, the aforementioned process can be, but is not limited to, as shown in steps S3 to S5 below.
[0084] S3. Select wavelet coefficients corresponding to image noise from the set of wavelet coefficients as the first wavelet coefficients, and use the remaining wavelet coefficients in the set as the second wavelet coefficients. In specific applications, after wavelet transform, the wavelet coefficients of effective signals have strong correlation between different scales, while the wavelet coefficients of noise signals have weak correlation between different scales. Therefore, based on this principle, the wavelet coefficients can be divided. The division process can be, but is not limited to, the steps S31 to S35 below.
[0085] S31. For the k-th wavelet coefficient in the wavelet coefficient set, relevant wavelet coefficients of the k-th wavelet coefficient are selected from the wavelet coefficient set, wherein the wavelet transform direction of the relevant wavelet coefficients is the same as that of the k-th wavelet coefficient, and the transform scale of the relevant wavelet coefficients is j+1, where j is the transform scale of the k-th wavelet coefficient; in specific applications, it is assumed that the k-th wavelet coefficient is represented as: (Where bd represents the wavelet transform direction, which can be horizontal, vertical, or diagonal; j represents the decomposition scale; and f(i, j) represents the pixel at position (i, j)). Therefore, the relevant wavelet coefficients of the k-th wavelet coefficient are: It also represents the wavelet coefficients of the pixel at position (i,j) in the same transformation direction and at the (j+1)th scale.
[0086] After determining the relevant wavelet coefficients of the k-th wavelet coefficient, the matching value of the wavelet coefficient belonging to the image noise can be calculated based on this, so that the k-th wavelet coefficient can be classified according to the matching value. The calculation process of the matching value can be, but is not limited to, the following step S32.
[0087] S32. Based on the k-th wavelet coefficient and its corresponding wavelet coefficient, calculate the matching value of the wavelet coefficient belonging to image noise. In specific implementation, multiply the k-th wavelet coefficient with its corresponding wavelet coefficient to obtain the matching value of the wavelet coefficient belonging to image noise. After obtaining the matching value, the k-th wavelet coefficient can be classified based on the matching value. The process can be, but is not limited to, the following step S33.
[0088] S33. Determine whether the matching value is less than the matching threshold corresponding to the j-th transform scale; In this embodiment, the matching threshold corresponding to each transform scale can be preset in the power equipment monitoring terminal and called when needed; If the matching value is greater than or equal to the corresponding matching threshold, it is determined to be a wavelet coefficient of a valid signal, that is, as the second wavelet coefficient; otherwise, it can be determined to be a wavelet coefficient corresponding to image noise, that is, the following step S34 is executed.
[0089] S34. If so, then the kth wavelet coefficient is taken as a first wavelet coefficient.
[0090] Based on the aforementioned steps S31 to S34, the division of the k-th wavelet coefficient can be completed. Then, the same principle can be used to classify the next wavelet coefficient. The cyclic classification process is shown in step S35 below.
[0091] S35. Increment k by 1, and re-select the relevant wavelet coefficients of the kth wavelet coefficient from the wavelet coefficient set until k equals K. Then, select all wavelet coefficients belonging to image noise from the wavelet coefficient set, where the initial value of k is 1 and K is the total number of wavelet coefficients in the wavelet coefficient set.
[0092] In addition, in this embodiment, when the transform scale of the kth wavelet coefficient is the maximum transform scale, the value of the kth wavelet coefficient can be directly used as the matching value of the wavelet coefficient corresponding to the image noise; then, step S33 can be executed directly.
[0093] Thus, through the aforementioned steps S31 to S35, the wavelet coefficients can be classified to obtain the wavelet coefficients belonging to image noise (i.e., the first wavelet coefficients) and the wavelet coefficients belonging to the effective image signal (i.e., the second wavelet coefficients). Then, based on this, subsequent denoising processing can be performed, and the process can be, but is not limited to, the steps shown in step S4 below.
[0094] S4. The first wavelet coefficients in the wavelet-transformed image signal are zeroed out, and the second wavelet coefficients in the wavelet-transformed image signal are adjusted to the optimal image reconstruction coefficients, so that a denoised wavelet-transformed signal is obtained after the second wavelet coefficients are adjusted. In this embodiment, setting each first wavelet coefficient to zero is equivalent to changing the signal part corresponding to each first wavelet coefficient to 0. Therefore, this operation is essentially treating it as a noise signal and removing it. At this time, the second denoising can be completed. At the same time, since the image signal obtained after the second denoising still contains a small amount of noise, this embodiment restores each second wavelet coefficient to the wavelet coefficients of the real signal (i.e., adjusts them to the optimal image reconstruction coefficients) to achieve the third denoising, thereby removing noise in the image to the greatest extent.
[0095] In practical applications, the above-mentioned coefficient adjustment process is illustrated below using any second wavelet coefficient as an example. It can be, but is not limited to, the steps S41 to S43 shown below.
[0096] S41. For any second wavelet coefficient in the wavelet transform image signal, obtain multiple adjacent wavelet coefficients of the second wavelet coefficient, wherein the multiple adjacent wavelet coefficients are the second wavelet coefficients of each pixel on the diagonal within a second neighborhood region centered on the pixel corresponding to the second wavelet coefficient; in specific applications, for example, the second neighborhood region can also be a 3×3 region, thus, it is equivalent to taking the second wavelet coefficients of each pixel on the diagonal within the 3×3 region as the adjacent wavelet coefficients corresponding to the aforementioned second wavelet coefficient.
[0097] After determining multiple adjacent wavelet coefficients of any of the aforementioned second wavelet coefficients, the adjustment of any of the second wavelet coefficients can be performed based on the aforementioned multiple adjacent wavelet coefficients. The adjustment process can be, but is not limited to, as shown in steps S42 and S43 below.
[0098] S42. Using multiple adjacent wavelet coefficients, calculate the noise variance of any second wavelet coefficient, and based on the noise variance and the multiple adjacent wavelet coefficients, calculate the local variance of any second wavelet coefficient; in specific applications, for example, but not limited to, the noise variance of any second wavelet coefficient can be calculated according to the following formula (2).
[0099]
[0100] In the above formula (2), Let Z represent the noise variance of any of the second wavelet coefficients, Z represent the wavelet sequence composed of the multiple adjacent wavelet coefficients, and median represent the median function.
[0101] After calculating the noise variance of any of the aforementioned wavelet coefficients based on the aforementioned formula (2), the local variance can be calculated based on the noise variance; for example, but not limited to, the local variance of any of the second wavelet coefficients can be calculated according to the following formula (3).
[0102]
[0103] In the above formula (3), Let z represent the local variance of any of the second wavelet coefficients. a Let A represent the a-th adjacent wavelet coefficient among the plurality of adjacent wavelet coefficients, and let A represent the total number of adjacent wavelet coefficients.
[0104] Thus, based on the aforementioned formulas (2) and (3), the noise variance and local variance of any of the aforementioned second wavelet coefficients can be calculated; then, based on these two, the aforementioned second wavelet coefficients can be adjusted, and the process can be, but is not limited to, as shown in step S43 below.
[0105] S43. Based on the local variance and noise variance of any second wavelet coefficient, adjust any second wavelet coefficient to the corresponding optimal image reconstruction coefficient; in specific implementation, for example, but not limited to, the following formula (4) can be used to adjust any second wavelet coefficient to the corresponding optimal image reconstruction coefficient.
[0106]
[0107] In the above formula (4), s1 represents any second wavelet coefficient, s1′ represents the optimal image reconstruction coefficient corresponding to any second wavelet coefficient, and s2 represents the relevant wavelet coefficient of any second wavelet coefficient; in this embodiment, the determination process of the relevant wavelet coefficient of any second wavelet coefficient can be referred to the aforementioned step S31, and the process will not be repeated here; at the same time, if in the formula If it is less than 0, then the Then the value is 0.
[0108] Therefore, by using the aforementioned formula (4), the adjustment of any of the aforementioned second wavelet coefficients can be completed, restoring them to the wavelet coefficients corresponding to the real signal. Thus, after adjusting all the second wavelet coefficients in the wavelet transform image signal based on the aforementioned steps S41 to S43, the denoised wavelet transform signal can be obtained. Based on this, by processing the wavelet coefficients twice, noise in the image can be removed to the greatest extent, thereby improving the image quality.
[0109] After completing the aforementioned denoising process, wavelet reconstruction can be performed on the denoised wavelet transform signal to obtain the optimal denoised infrared image; wherein, the wavelet reconstruction process is as shown in step S5 below.
[0110] S5. Perform wavelet reconstruction processing on the denoised wavelet transform signal to obtain the optimal denoised infrared image; in this embodiment, wavelet reconstruction is a commonly used technique for image restoration, and its principle will not be elaborated here; after obtaining the optimal denoised infrared image, the temperature detection result of a specified part of the power equipment can be obtained based on the optimal denoised infrared image; wherein, the temperature detection process can be, but is not limited to, as shown in step S6 below.
[0111] S6. Based on the optimal denoised infrared image, obtain the temperature detection result of a specified part of the power equipment; in specific implementation, for example, but not limited to, the following steps S61 to S64 can be used to obtain the aforementioned temperature detection result.
[0112] S61. Perform grayscale mapping processing on the optimal denoised infrared image to obtain a grayscale mapped infrared image; in specific implementation, for any pixel in the optimal denoised infrared image, the following formula (5) can be used to obtain the mapped grayscale value of any pixel.
[0113]
[0114] In the above formula (5), L represents the gray value of any pixel in the optimal denoised infrared image, and L′ represents the mapped gray value of any pixel in the optimal denoised infrared image. max L represents the maximum grayscale value in the optimally denoised infrared image. min This represents the minimum grayscale value in the optimal denoised infrared image.
[0115] Thus, by using the aforementioned formula (5), the grayscale value of each pixel in the optimal denoised infrared image can be mapped to obtain a grayscale mapped infrared image. Then, based on this grayscale mapped infrared image, the temperature detection result of a specified part on the power equipment can be obtained. The detection process is as shown in steps S62 to S64 below.
[0116] S62. Determine the first temperature value corresponding to the maximum gray value and the second temperature value corresponding to the minimum gray value from the gray-scale mapped infrared image.
[0117] S63. Based on the first temperature value, the second temperature value, and the gray value of each pixel in the gray-scale mapped infrared image, determine the temperature value corresponding to each pixel in the gray-scale mapped infrared image; in this embodiment, for example, but not limited to, the following formula (6) can be used to calculate the temperature value corresponding to each pixel in the gray-scale mapped infrared image.
[0118]
[0119] In the above formula (6), T u This represents the temperature value corresponding to the u-th pixel in the grayscale-mapped infrared image, where T1 represents the first temperature value, T2 represents the second temperature value, and L... u ′ represents the gray value of the u-th pixel in the grayscale-mapped infrared image, where u = 1, 2, 3, ..., U, and U is the total number of pixels in the grayscale-mapped infrared image.
[0120] Therefore, by using the aforementioned formula (6), the temperature value corresponding to each pixel in the grayscale mapping infrared image can be calculated, and the temperature detection result of a specified part on the power equipment can be obtained. The process is shown in step S64 below.
[0121] S64. Based on the temperature value corresponding to each pixel, the temperature detection result of the specified part of the power equipment is obtained; in this embodiment, the temperature distribution map of the specified part of the power equipment can be generated based on the temperature value corresponding to each pixel in the grayscale mapping infrared image, so as to visualize the temperature distribution map as the temperature detection result.
[0122] Through the aforementioned steps S61 to S64, the temperature detection results of a specified part of the power equipment can be obtained based on the optimal denoised infrared image. Then, based on this, the equipment temperature warning can be carried out, thereby ensuring the stable operation of the power equipment.
[0123] Thus, through the detailed explanation of the remote temperature detection method for power equipment in the aforementioned steps S1 to S6, this invention performs multiple noise reduction processes on the image before temperature detection, which can effectively reduce the impact of noise on temperature detection, thereby improving the accuracy of temperature detection. Therefore, it is very suitable for large-scale application and promotion in the field of power equipment monitoring.
[0124] like Figure 2 As shown, the second aspect of this embodiment provides a hardware device for implementing the remote temperature detection method for power equipment described in the first aspect of the embodiment, comprising:
[0125] The first denoising unit is used to acquire an infrared image of a specified part of the power equipment and perform initial denoising processing on the infrared image of the specified part to obtain an initial denoised infrared image.
[0126] The wavelet transform unit is used to perform wavelet transform processing on the initial denoised infrared image to obtain the wavelet transformed image signal corresponding to the initial denoised infrared image, and to form a wavelet coefficient set using the wavelet coefficients corresponding to each transform scale in the wavelet transformed image signal.
[0127] The second denoising unit is used to select wavelet coefficients corresponding to image noise from the wavelet coefficient set as first wavelet coefficients, and use the remaining wavelet coefficients in the wavelet coefficient set as second wavelet coefficients.
[0128] The second denoising unit is further configured to zero out the first wavelet coefficients in the wavelet-transformed image signal and adjust the second wavelet coefficients in the wavelet-transformed image signal to the optimal image reconstruction coefficients, so as to obtain a denoised wavelet-transformed signal after adjusting the second wavelet coefficients.
[0129] The image reconstruction unit is used to perform wavelet reconstruction processing on the denoised wavelet transform signal to obtain the optimal denoised infrared image.
[0130] The temperature detection unit is used to obtain the temperature detection result of a specified part of the power equipment based on the optimal denoised infrared image.
[0131] The working process, working details and technical effects of the device provided in this embodiment can be found in the first aspect of the embodiment, and will not be repeated here.
[0132] like Figure 3 As shown, the third aspect of this embodiment provides another remote temperature detection device for power equipment. Taking the device as an electronic device as an example, it includes: a memory, a processor, and a transceiver that are connected in sequence. The memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the remote temperature detection method for power equipment as described in the first aspect of the embodiment.
[0133] For specific examples, the memory may include, but is not limited to, random access memory (RAM), read-only memory (ROM), flash memory, first-in-first-out (FIFO) memory, and / or first-in-last-out (FILO) memory, etc.; specifically, the processor may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor may be implemented using at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), PLA (Programmable Logic Array). The processor may also include a main processor and a coprocessor. The main processor is used to process data in the wake-up state, also known as the CPU (Central Processing Unit); the coprocessor is a low-power processor used to process data in the standby state.
[0134] In some embodiments, the processor may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. For example, the processor may not be limited to microprocessors of the STM32F105 series, reduced instruction set computer (RISC) microprocessors, x86 architecture processors, or processors with integrated embedded neural network processing units (NPUs). The transceiver may be, but is not limited to, a Wi-Fi transceiver, a Bluetooth transceiver, a General Packet Radio Service (GPRS) transceiver, a ZigBee (a low-power LAN protocol based on the IEEE 802.15.4 standard) transceiver, a 3G transceiver, a 4G transceiver, and / or a 5G transceiver. Furthermore, the device may also include, but is not limited to, a power module, a display screen, and other necessary components.
[0135] The working process, working details and technical effects of the electronic device provided in this embodiment can be found in the first aspect of the embodiment, and will not be repeated here.
[0136] The fourth aspect of this embodiment provides a storage medium that stores instructions for a remote temperature detection method for power equipment as described in the first aspect of the embodiment. That is, the storage medium stores instructions that, when executed on a computer, perform the remote temperature detection method for power equipment as described in the first aspect of the embodiment.
[0137] The storage medium refers to a carrier for storing data, which may include, but is not limited to, floppy disks, optical disks, hard disks, flash memory, USB flash drives, and / or memory sticks. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
[0138] The working process, working details and technical effects of the storage medium provided in this embodiment can be found in the first aspect of the embodiment, and will not be repeated here.
[0139] The fifth aspect of this embodiment provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform the remote temperature detection method for power equipment as described in the first aspect of this embodiment, wherein the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.
[0140] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for remote temperature detection of power equipment, characterized in that, include: Acquire an infrared image of a specified part of a power equipment, and perform initial denoising processing on the infrared image of the specified part to obtain an initial denoised infrared image; The initial denoised infrared image is subjected to wavelet transform processing to obtain the wavelet transformed image signal corresponding to the initial denoised infrared image, and the wavelet coefficient set is formed by using the wavelet coefficients corresponding to each transform scale in the wavelet transformed image signal. Wavelet coefficients corresponding to image noise are selected from the set of wavelet coefficients and used as the first wavelet coefficients. The remaining wavelet coefficients in the set of wavelet coefficients are used as the second wavelet coefficients. The first wavelet coefficients in the wavelet-transformed image signal are set to zero, and the second wavelet coefficients in the wavelet-transformed image signal are adjusted to the optimal image reconstruction coefficients, so that a denoised wavelet-transformed signal is obtained after the second wavelet coefficients are adjusted. The denoised wavelet transform signal is subjected to wavelet reconstruction processing to obtain the optimal denoised infrared image; Based on the optimal denoised infrared image, the temperature detection result of a specified part of the power equipment is obtained.
2. The method according to claim 1, characterized in that, The infrared image of the specified region is subjected to initial denoising processing to obtain an initial denoised infrared image, including: The infrared image is subjected to grayscale transformation to obtain a grayscale image; For any pixel in the grayscale image, obtain the first neighborhood region centered on the pixel. The first neighborhood region is divided into multiple sub-neighborhood regions, wherein each sub-neighborhood region includes any of the aforementioned pixels; Determine the grayscale contrast threshold for each sub-neighborhood region and select the smallest grayscale contrast threshold; Calculate the average gray value of all pixels in the sub-neighborhood region corresponding to the minimum gray-level contrast threshold; The grayscale value of any pixel is updated to the grayscale mean, and after the grayscale values of all pixels in the grayscale image have been updated, the initial denoised infrared image is obtained.
3. The method according to claim 2, characterized in that, Determine the grayscale contrast threshold for each sub-neighborhood region, including: For any sub-neighborhood region, the grayscale contrast threshold of the sub-neighborhood region is calculated using the following formula (1); (1) In the above formula (1), This represents the grayscale contrast threshold of any of the sub-neighborhood regions. This represents the grayscale value of the i-th pixel in any of the sub-neighborhood regions. This represents the average grayscale value of all pixels in any given sub-neighborhood region. This represents the total number of pixels in any of the sub-neighborhood regions. This represents the comparison coefficient.
4. The method according to claim 1, characterized in that, The wavelet coefficients corresponding to image noise are selected from the set of wavelet coefficients and used as the first wavelet coefficients, including: For the k-th wavelet coefficient in the wavelet coefficient set, relevant wavelet coefficients of the k-th wavelet coefficient are selected from the wavelet coefficient set. The wavelet transform direction of the relevant wavelet coefficients is the same as that of the k-th wavelet coefficient. The transform scale of the relevant wavelet coefficients is j+1, and j is the transform scale of the k-th wavelet coefficient. Based on the k-th wavelet coefficient and the related wavelet coefficient of the k-th wavelet coefficient, the matching value of the wavelet coefficient belonging to the image noise of the k-th wavelet coefficient is calculated; Determine whether the matching value is less than the matching threshold corresponding to the j-th transformation scale; If so, then the kth wavelet coefficient is taken as a first wavelet coefficient; Increment k by 1, and re-select the relevant wavelet coefficients of the k-th wavelet coefficient from the wavelet coefficient set until k equals K. Then, select all wavelet coefficients belonging to image noise from the wavelet coefficient set. The initial value of k is 1, and K is the total number of wavelet coefficients in the wavelet coefficient set.
5. The method according to claim 1, characterized in that, Adjusting the second wavelet coefficients in the wavelet-transformed image signal to the optimal image reconstruction coefficients includes: For any second wavelet coefficient in the wavelet transform image signal, obtain multiple adjacent wavelet coefficients of the second wavelet coefficient, wherein the multiple adjacent wavelet coefficients are the second wavelet coefficients of each pixel point located on the diagonal within a second neighborhood region centered on the pixel point corresponding to the second wavelet coefficient. Using multiple adjacent wavelet coefficients, the noise variance of any second wavelet coefficient is calculated, and based on the noise variance and the multiple adjacent wavelet coefficients, the local variance of any second wavelet coefficient is calculated. Based on the local variance and noise variance of any of the second wavelet coefficients, the second wavelet coefficients are adjusted to the corresponding optimal image reconstruction coefficients.
6. The method according to claim 5, characterized in that, The noise variance of any second wavelet coefficient is calculated using multiple adjacent wavelet coefficients, including: The noise variance of any second wavelet coefficient is calculated according to the following formula (2); (2) In the above formula (2), This represents the noise variance of any of the second wavelet coefficients. This represents the wavelet sequence composed of the multiple adjacent wavelet coefficients. Represent the median function; Accordingly, based on the noise variance and the plurality of adjacent wavelet coefficients, the local variance of any second wavelet coefficient is calculated, which includes: The local variance of any second wavelet coefficient is calculated according to the following formula (3); (3) In the above formula (3), This represents the local variance of any of the second wavelet coefficients. This represents the a-th adjacent wavelet coefficient among the plurality of adjacent wavelet coefficients. This represents the total number of adjacent wavelet coefficients.
7. The method according to claim 1, characterized in that, Based on the optimal denoised infrared image, the temperature detection result of a specified location on the power equipment is obtained, including: The optimal denoised infrared image is subjected to grayscale mapping processing to obtain a grayscale mapped infrared image; The first temperature value corresponding to the maximum gray value and the second temperature value corresponding to the minimum gray value are determined from the gray-scale mapped infrared image. Based on the first temperature value, the second temperature value, and the grayscale value of each pixel in the grayscale-mapped infrared image, the temperature value corresponding to each pixel in the grayscale-mapped infrared image is determined. Based on the temperature value corresponding to each pixel, the temperature detection result of a specified part of the power equipment is obtained.
8. A remote temperature detection device for power equipment, characterized in that, include: The first denoising unit is used to acquire an infrared image of a specified part of the power equipment and perform initial denoising processing on the infrared image of the specified part to obtain an initial denoised infrared image. The wavelet transform unit is used to perform wavelet transform processing on the initial denoised infrared image to obtain the wavelet transformed image signal corresponding to the initial denoised infrared image, and to form a wavelet coefficient set using the wavelet coefficients corresponding to each transform scale in the wavelet transformed image signal. The second denoising unit is used to select wavelet coefficients corresponding to image noise from the wavelet coefficient set as first wavelet coefficients, and use the remaining wavelet coefficients in the wavelet coefficient set as second wavelet coefficients. The second denoising unit is also used to set the first wavelet coefficients in the wavelet-transformed image signal to zero, and to adjust the second wavelet coefficients in the wavelet-transformed image signal to the optimal image reconstruction coefficients, so as to obtain a denoised wavelet-transformed signal after adjusting the second wavelet coefficients. An image reconstruction unit is used to perform wavelet reconstruction processing on the denoised wavelet transform signal to obtain the optimal denoised infrared image. The temperature detection unit is used to obtain the temperature detection result of a specified part of the power equipment based on the optimal denoised infrared image.
9. A remote temperature detection device for power equipment, characterized in that, include: A memory, a processor, and a transceiver are sequentially connected in communication, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the remote temperature detection method for power equipment as described in any one of claims 1 to 7.
10. A computer program product containing instructions, characterized in that, When the instruction is executed on the computer, it causes the computer to perform the remote temperature detection method for power equipment as described in any one of claims 1 to 7.
