A data processing method and system for infrared thermal imaging

Abstract

The present invention discloses a data processing method and system for infrared thermal imaging. The method includes obtaining infrared thermal imaging information; performing a noise reduction operation on the infrared thermal imaging information to obtain denoised wavelet coefficient values; constructing a threshold function based on the denoised wavelet coefficient values to obtain a local weight threshold function and a high threshold function; performing an image reconstruction operation based on the denoised wavelet coefficient values, the local weight threshold function, and the high threshold function to obtain reconstructed image data; performing a first image enhancement operation on the reconstructed image data to obtain a first enhanced image; performing a second image enhancement operation on the first enhanced image to obtain a second enhanced image; and performing an image output operation on the second enhanced image to obtain an output image. This method can achieve maintaining image details and smoothing the noise.

Description

Technical Field

[0001] The present invention relates to the field of infrared thermal imaging technology, and in particular, to a data processing method and system for infrared thermal imaging. Background Art

[0002] At present, the application of infrared thermal imaging technology in the field of industrial detection is constantly deepening. Its unique advantages make it an important tool for industrial safety monitoring and equipment maintenance. By capturing the infrared radiation on the surface of an object, this technology can display the temperature distribution in real time and accurately, thus helping engineers to timely discover potential faults and safety hazards. In the power industry, infrared thermal imaging is widely used to detect temperature anomalies of equipment such as high-voltage cables, transformers, and switchgear, and can perform non-destructive testing without power interruption, effectively preventing electrical fires and equipment overheating. In the production of electronic components, infrared thermal imaging can quickly identify problems such as poor soldering or inconsistent materials, improving product quality and production efficiency. The key to the application of thermal infrared imaging lies in how to extract more valuable information from thermal infrared images. Among them, image enhancement technology is a basic image processing technology, which can improve the edge and detail features of thermal infrared images, facilitating more detailed observation by analysts.

[0003] In an existing technology, there are two methods: local enhancement and global enhancement. The global enhancement method, histogram equalization, enhances the contrast by adjusting the gray-scale distribution of the entire image, but this method is prone to blurring of image details. The local enhancement method, on the other hand, can better highlight the detail features of the image. Common local enhancement techniques include enhancement based on local standard deviation, gradient enhancement, Retinex transform, and local Wiener filtering, etc. These methods can effectively suppress noise and improve image contrast by processing in the local area of the image, but there are also some limitations. When processing noise, they will further blur the image details. In addition, the noise problem of infrared images is also an important factor affecting their application effect. The noise mainly comes from electronic devices, signal processing systems, and the external environment. To suppress noise, methods such as low-pass filters and median filters are used, but these methods will reduce the detail information of the image while denoising.

[0004] In summary, there is a problem in the existing technology that it is impossible to achieve smooth processing of noise while maintaining image details. Summary of the Invention

[0005] The present invention provides a data processing method and system for infrared thermal imaging to achieve smooth processing of noise while maintaining image details.

[0006] In a first aspect, to solve the above technical problems, the present invention provides a data processing method for infrared thermal imaging, including:

[0007] Obtain infrared thermal imaging information;

[0008] Perform a denoising operation based on the infrared thermal imaging information to obtain denoised wavelet coefficient values;

[0009] Perform a threshold function construction operation based on the denoised wavelet coefficient values to obtain a local weight threshold function and a high threshold function;

[0010] Perform an image reconstruction operation based on the denoised wavelet coefficient values, the local weight threshold function, and the high threshold function to obtain reconstructed image data;

[0011] Perform a first image enhancement operation based on the reconstructed image data to obtain a first enhanced image;

[0012] Perform a second image enhancement operation based on the first enhanced image to obtain a second enhanced image;

[0013] Perform an image output operation based on the second enhanced image to obtain an output image.

[0014] Preferably, in an alternative embodiment, performing a denoising operation based on the infrared thermal imaging information to obtain denoised wavelet coefficient values includes:

[0015] Perform a wavelet decomposition operation based on the infrared thermal imaging information to obtain low-frequency coefficients and first high-frequency coefficients;

[0016] Perform a high-frequency coefficient threshold processing operation based on the first high-frequency coefficients to obtain second high-frequency coefficients;

[0017] Perform a denoised wavelet coefficient value generation operation based on the second high-frequency coefficients and the low-frequency coefficients to obtain denoised wavelet coefficient values.

[0018] Preferably, in an alternative embodiment, performing a threshold function construction operation based on the denoised wavelet coefficient values to obtain a local weight threshold function and a high threshold function includes:

[0019] Perform a first local energy model construction operation based on the denoised wavelet coefficient values to obtain a first local energy model;

[0020] Perform a high threshold function construction operation based on the denoised wavelet coefficient values and the first local energy model to obtain a high threshold function;

[0021] Perform a local weight threshold function construction operation based on the denoised wavelet coefficient values to obtain a local weight threshold function.

[0022] Preferably, in an alternative embodiment, the calculation formula for the first local energy model construction operation is as follows:

[0023]

[0024] The calculation formula for the high - threshold function construction operation is as follows:

[0025]

[0026] The calculation formula for the local - weight threshold function construction operation is as follows:

[0027]

[0028] Wherein, represents the first local energy model, represents the weight of the first local energy model corresponding to the pixel, represents the Gaussian distribution function of the first local energy model, represents the mean value in the Gaussian distribution function of the first local energy model, represents the standard deviation in the Gaussian distribution function of the first local energy model, represents the denoised wavelet coefficient value; represents the high - threshold function, represents the first amplitude parameter, represents the second amplitude parameter, represents the segmentation threshold; represents the local - weight threshold function, represents the pixel value, represents the pixel value, represents the regularization parameter, represents the weight parameter of the th pixel, represents the smoothing coefficient, represents the denoised wavelet coefficient value corresponding to the th pixel,

[0029] Preferably, in an alternative embodiment, according to the denoised wavelet coefficient value, the local - weight threshold function, and the high - threshold function, an image reconstruction operation is performed to obtain reconstructed image data, including:

[0030] Performing a local - weight threshold processing operation according to the denoised wavelet coefficient value and the local - weight threshold function to obtain local - weight wavelet coefficient values;

[0031] Performing a high - threshold processing operation according to the denoised wavelet coefficient value and the high - threshold function to obtain high - threshold wavelet coefficient values;

[0032] Perform a first data fusion operation based on the local weighted wavelet coefficient values and the high threshold wavelet coefficient values to obtain fused wavelet coefficient values;

[0033] Perform a wavelet reconstruction operation based on the fused wavelet coefficient values to obtain reconstructed image data.

[0034] Preferably, in an alternative embodiment, perform a first image enhancement operation based on the reconstructed image data to obtain a first enhanced image, including:

[0035] Perform an operation to obtain Gaussian distribution parameters based on the reconstructed image data to obtain Gaussian distribution parameters;

[0036] Perform a second local energy model construction operation based on the Gaussian distribution parameters to obtain a second local energy model;

[0037] Perform a curve feature extraction operation based on the second local energy model to obtain curve features;

[0038] Perform a first image enhancement operation based on the curve features to obtain a first enhanced image;

[0039] The calculation formula for the second local energy model construction operation is as follows:

[0040]

[0041] Wherein, represents the second local energy model, represents the weight of the second local energy model corresponding to the th pixel, represents the Gaussian distribution function of the second local energy model, represents the mean value in the Gaussian distribution function of the second local energy model, represents the standard deviation in the Gaussian distribution function of the second local energy model, represents the fused wavelet coefficient values.

[0042] Preferably, in an alternative embodiment, perform a second image enhancement operation based on the first enhanced image to obtain a second enhanced image, including:

[0043] Perform a Laplacian operator processing operation based on the first enhanced image to obtain a Laplacian image;

[0044] Perform an exponential transformation operation based on the Laplacian image to obtain an exponential transformation image;

[0045] Perform a Gaussian distribution feature extraction operation based on the first enhanced image to obtain Gaussian distribution features;

[0046] Perform a second data fusion operation based on the Gaussian distribution characteristics and the exponential transformation image to obtain a second enhanced image.

[0047] In a second aspect, the present invention provides a data processing system for infrared thermal imaging, comprising:

[0048] A data acquisition module for acquiring infrared thermal imaging information;

[0049] A noise reduction module for performing a noise reduction operation based on the infrared thermal imaging information to obtain denoised wavelet coefficient values;

[0050] A threshold function construction module for performing a threshold function construction operation based on the denoised wavelet coefficient values to obtain a local weight threshold function and a high threshold function;

[0051] An image reconstruction module for performing an image reconstruction operation based on the denoised wavelet coefficient values, the local weight threshold function, and the high threshold function to obtain reconstructed image data;

[0052] A first image enhancement module for performing a first image enhancement operation based on the reconstructed image data to obtain a first enhanced image;

[0053] A second image enhancement module for performing a second image enhancement operation based on the first enhanced image to obtain a second enhanced image;

[0054] An image output module for performing an image output operation based on the second enhanced image to obtain an output image.

[0055] In a third aspect, the present invention further provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a data processing method for infrared thermal imaging described in any one of the above.

[0056] In a fourth aspect, the present invention further provides a computer-readable storage medium. The computer-readable storage medium includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute a data processing method for infrared thermal imaging described in any one of the above.

[0057] Compared with the prior art, the present invention has the following beneficial effects: By providing a data processing method and system for infrared thermal imaging, it realizes the smoothing processing of noise while maintaining image details. First, the present invention obtains infrared thermal imaging information and performs a noise reduction operation to obtain denoised wavelet coefficient values, effectively reducing the noise components in the image. Then, local weight threshold functions and high threshold functions are constructed using the denoised wavelet coefficient values. The construction of these two functions enables more precise processing of the detail information and overall structure of the image during the image reconstruction process. The reconstructed image data obtained through the image reconstruction operation further reduces the influence of noise while retaining the main features of the image. Then, through two image enhancement operations, local detail enhancement and overall contrast enhancement are respectively performed on the reconstructed image, making the detail information of the image more prominent and improving the overall visual effect at the same time. Finally, the output image realizes the smoothing processing of noise while maintaining image details.

[0058] In summary, the present invention provides a data processing method and system for infrared thermal imaging, aiming to solve the problem in the prior art that it is difficult to effectively suppress noise while enhancing the detail information of the image. The method first obtains infrared thermal imaging information and performs a noise reduction operation on it to obtain denoised wavelet coefficient values. This step reduces the noise components in the image through wavelet decomposition and high-frequency coefficient threshold processing operations. Then, local weight threshold functions and high threshold functions are constructed according to the denoised wavelet coefficient values. The introduction of these two functions enables more precise processing of the detail information and overall structure of the image in the subsequent image reconstruction process, avoiding the loss of detail information and the excessive introduction of noise. The reconstructed image data obtained through the image reconstruction operation further reduces the influence of noise while retaining the main features of the image. Subsequently, two image enhancement operations are performed. The first enhancement operation performs local detail enhancement on the reconstructed image through steps such as constructing a second local energy model and curve feature extraction, making the detail information of the image more prominent; the second enhancement operation performs overall contrast enhancement on the first enhanced image through steps such as Laplacian operator processing and exponential transformation, further improving the overall visual effect of the image. Finally, the output image can provide high-quality image data support for applications in fields such as industrial detection, improving the accuracy and reliability of image analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a schematic flowchart of the data processing method for infrared thermal imaging provided by the first embodiment of the present invention;

[0060] Figure 2 is a schematic structural diagram of the data processing system for infrared thermal imaging provided by the second embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present invention.

[0062] Referring to Figure 1 , the first embodiment of the present invention provides a data processing method for infrared thermal imaging, including the following steps:

[0063] S11 Obtain infrared thermal imaging information;

[0064] S12 Perform a noise reduction operation according to the infrared thermal imaging information to obtain a denoised wavelet coefficient value;

[0065] S13 Perform a threshold function construction operation according to the denoised wavelet coefficient value to obtain a local weight threshold function and a high threshold function;

[0066] S14 Perform an image reconstruction operation according to the denoised wavelet coefficient value, the local weight threshold function, and the high threshold function to obtain reconstructed image data;

[0067] S15 Perform a first image enhancement operation according to the reconstructed image data to obtain a first enhanced image;

[0068] S16 Perform a second image enhancement operation according to the first enhanced image to obtain a second enhanced image;

[0069] S17 Perform an image output operation according to the second enhanced image to obtain an output image.

[0070] It should be noted that, first of all, by obtaining infrared thermal imaging information, it provides the raw data basis for subsequent processing. Then, noise reduction operations are carried out. By using technical means such as wavelet decomposition and high-frequency coefficient threshold processing, the noise components in the image are effectively removed, and the denoised wavelet coefficient values are obtained, thus laying the foundation for the clarity and accuracy of the image. Then, based on the denoised wavelet coefficient values, a local weight threshold function and a high threshold function are constructed. The precise construction of these two functions enables more refined processing of the detailed information and overall structure of the image during the image reconstruction process, avoiding the loss of details and the excessive introduction of noise. The reconstructed image data obtained through the image reconstruction operation, while retaining the main features of the image, further reduces the influence of noise and provides high-quality input for subsequent image enhancement. Subsequently, two image enhancement operations are carried out. The first enhancement operation enhances the local details of the reconstructed image through steps such as constructing a second local energy model and curve feature extraction, making the detailed information of the image more prominent; the second enhancement operation enhances the overall contrast of the first enhanced image through steps such as Laplacian operator processing and exponential transformation, further improving the overall visual effect of the image and making both the details and the overall structure of the image improved. Finally, the output image obtained through the image output operation can provide high-quality image data support for applications in fields such as industrial detection, improving the accuracy and reliability of image analysis.

[0071] In step S11, infrared thermal imaging information is obtained.

[0072] It should be noted that in step S11, obtaining infrared thermal imaging information is the starting step of the entire data processing method, and its purpose is to provide the raw data basis for subsequent operations such as noise reduction, threshold function construction, image reconstruction, and image enhancement. Infrared thermal imaging information refers to the object surface temperature distribution data captured by an infrared thermal imaging device. These data are presented in the form of an image and can reflect the temperature differences in different regions of the object surface. The infrared thermal imaging device includes an infrared detector and a signal processing system. The infrared detector is responsible for receiving the infrared energy radiated from the object surface and converting it into an electrical signal, and the signal processing system processes and converts the electrical signal to finally generate an infrared thermal imaging image. During the process of obtaining infrared thermal imaging information, it is necessary to ensure the normal working state of the device and appropriate environmental conditions to ensure the accuracy and reliability of the obtained image data. In industrial detection scenarios, an infrared thermal imaging device is used to scan power equipment, mechanical devices, etc. to obtain infrared thermal imaging information on their surface temperature distribution, providing a basis for subsequent fault diagnosis and performance evaluation. In addition, when obtaining infrared thermal imaging information, other sensor data such as visible light images and vibration signals can be combined for multi-modal data fusion to obtain more comprehensive detection results. Through these methods, high-quality infrared thermal imaging information can be ensured, laying the foundation for subsequent data processing and analysis.

[0073] In step S12, a denoising operation is performed based on the infrared thermal imaging information to obtain denoised wavelet coefficient values, including:

[0074] Performing a wavelet decomposition operation on the infrared thermal imaging information to obtain low-frequency coefficients and first high-frequency coefficients;

[0075] Performing a high-frequency coefficient threshold processing operation on the first high-frequency coefficients to obtain second high-frequency coefficients;

[0076] Performing a denoised wavelet coefficient value generation operation based on the second high-frequency coefficients and the low-frequency coefficients to obtain denoised wavelet coefficient values.

[0077] It should be noted that in step S12, the denoising operation can accurately identify and remove the noise components in the infrared thermal imaging information while maximizing the retention of the detailed features of the image. First, the infrared thermal imaging information is decomposed into low-frequency coefficients and high-frequency coefficients by using wavelet decomposition technology. Wavelet decomposition is a multi-scale analysis method that can decompose the image signal into wavelet coefficients at different scales. Among them, the low-frequency coefficients mainly contain the overall structure information of the image, while the high-frequency coefficients contain the detailed information such as edges and textures of the image as well as noise. Through this decomposition, different frequency components of the image can be separated, providing a basis for subsequent noise suppression and detail retention. Then, a high-frequency coefficient threshold processing operation is performed on the first high-frequency coefficients. A reasonable threshold needs to be set for this operation, which is determined according to the statistical characteristics and empirical knowledge of the image. Exemplarily, the high-frequency coefficient threshold is set to 0.85. Of course, according to different actual applications and user requirements, the high-frequency coefficient threshold can be set to 0.6, etc. The present invention does not limit this. The high-frequency coefficients below the threshold are considered noise components and are set to zero, while the high-frequency coefficients above the threshold are considered important detailed information of the image and are retained. In this way, the noise components in the image can be effectively removed while retaining the important detailed information. Finally, the processed second high-frequency coefficients are combined with the low-frequency coefficients to perform a denoised wavelet coefficient value generation operation. This process recombines the denoised high-frequency coefficients and low-frequency coefficients through a reconstruction algorithm to generate denoised wavelet coefficient values. The denoised wavelet coefficient values play a key role in subsequent image reconstruction and enhancement processing. In industrial equipment detection, through the denoising operation, image blurring and noise interference caused by factors such as environmental interference, equipment noise, or sensor errors can be removed, making key information such as the thermal abnormal areas and temperature distributions of the equipment more clearly visible, thereby improving the accuracy and reliability of fault diagnosis.

[0078] In step S13, based on the denoised wavelet coefficient values, a threshold function construction operation is performed to obtain a local weight threshold function and a high threshold function, including:

[0079] Perform the first local energy model construction operation based on the denoised wavelet coefficient values to obtain the first local energy model;

[0080] Perform the high threshold function construction operation based on the denoised wavelet coefficient values and the first local energy model to obtain the high threshold function;

[0081] Perform the local weighted threshold function construction operation based on the denoised wavelet coefficient values to obtain the local weighted threshold function.

[0082] The calculation formula for the first local energy model construction operation is as follows:

[0083]

[0084] The calculation formula for the high threshold function construction operation is as follows:

[0085]

[0086] The calculation formula for the local weighted threshold function construction operation is as follows:

[0087]

[0088] Among them, represents the first local energy model, represents the weight of the first local energy model corresponding to the pixel, represents the Gaussian distribution function of the first local energy model, represents the mean value in the Gaussian distribution function of the first local energy model, represents the denoised wavelet coefficient value; represents the high threshold function, represents the first amplitude parameter, represents the second amplitude parameter, represents the segmentation threshold; represents the local weighted threshold function, represents pixel values, represents pixel values, represents the regularization parameter, represents the weight parameter of the pixel, represents the smoothing coefficient, represents the denoised wavelet coefficient value corresponding to the pixel,

[0089] It should be noted that in step S13, the threshold function construction operation can provide accurate parameter guidance for image reconstruction to achieve fine processing of image details and overall structure. This operation is achieved by constructing a local weight threshold function and a high threshold function. First, a first local energy model construction operation is performed according to the denoised wavelet coefficient values. The local energy model is a mathematical model that describes the energy distribution in a local area of an image and can reflect the intensity change of the image in that area. For example, if the denoised wavelet coefficient values are concentrated in a certain local area and the pixel weights in this area are relatively large, then by calculating , a relatively high local energy value can be obtained, where is a Gaussian distribution function with as the mean and as the standard deviation. This means that this area has relatively important structural features in the image. Then, a high threshold function construction operation is performed using the first local energy model and the denoised wavelet coefficient values. The high threshold function compares the value of the first local energy model with a threshold by setting a segmentation threshold . Exemplarily, is set to 0.7, the first amplitude parameter is set to 1.2, the second amplitude parameter is set to 0.8. When is greater than 0.7, the first amplitude parameter =1.2, otherwise the second amplitude parameter =0.8. Of course, according to different actual applications and user requirements, can be set to 0.6, the first amplitude parameter can be set to 1.4, the second amplitude parameter can be set to 0.5, etc. The present invention does not limit this. For areas with relatively high local energy, that is, greater than the set segmentation threshold, the high threshold function will give a larger enhancement amplitude, while for areas with lower energy, a moderate suppression will be performed. Finally, a local weight threshold function construction operation is performed according to the denoised wavelet coefficient values. This function can assign different weights according to the importance of pixel points in the image, so as to protect detailed information and suppress noise during the image reconstruction process. When processing the infrared thermal imaging image of industrial equipment, by constructing a suitable threshold function, the overheated area of the equipment can be highlighted, and at the same time, the surrounding normal temperature area can be smoothed, enabling technicians to perform more accurate fault diagnosis and analysis.

[0090] In step S14, according to the denoised wavelet coefficient values, the local weight threshold function, and the high threshold function, an image reconstruction operation is performed to obtain reconstructed image data, including:

[0091] Perform a local weight threshold processing operation based on the denoised wavelet coefficient value and the local weight threshold function to obtain a local weight wavelet coefficient value;

[0092] Perform a high threshold processing operation based on the denoised wavelet coefficient value and the high threshold function to obtain a high threshold wavelet coefficient value;

[0093] Perform a first data fusion operation based on the local weight wavelet coefficient value and the high threshold wavelet coefficient value to obtain a fused wavelet coefficient value;

[0094] Perform a wavelet reconstruction operation based on the fused wavelet coefficient value to obtain reconstructed image data.

[0095] It should be noted that in step S14, the image reconstruction operation can integrate the information of the denoised wavelet coefficient value, the local weight threshold function, and the high threshold function to accurately restore the details and structure of the image. First, perform a local weight threshold processing operation using the denoised wavelet coefficient value and the local weight threshold function. During this process, each wavelet coefficient value is adjusted according to its corresponding local weight threshold function, which takes into account the importance of the pixel point in the image. By calculating the square of the difference between the pixel value and other pixel values in the neighborhood, and combining the weight parameter, the regularization parameter, and the smoothing coefficient, an appropriate weight is assigned to each pixel point. For the edge pixels in the image, their weights are set higher to retain more detail information. Then, perform a high threshold processing operation based on the denoised wavelet coefficient value and the high threshold function. The high threshold function enhances or suppresses the wavelet coefficient value according to the comparison result between the value of the first local energy model and the segmentation threshold. For regions with high local energy, such as significant features or target regions in the image, the high threshold function gives a larger enhancement amplitude to make them more prominent in the reconstructed image. Then, perform a first data fusion operation on the local weight wavelet coefficient value and the high threshold wavelet coefficient value to obtain a fused wavelet coefficient value. This fusion process combines the results of the local weight threshold processing and the high threshold processing, enabling better balance and coordination between the detail information and the overall structure information of the image. Finally, perform a wavelet reconstruction operation based on the fused wavelet coefficient value to obtain reconstructed image data. Wavelet reconstruction is an inverse operation that recombines the processed wavelet coefficient values to restore the original structure and details of the image. In the processed fused wavelet coefficient values, details such as edges and textures are enhanced, while noise and unimportant details are suppressed, thus obtaining a reconstructed image with high clarity and accurate structure. Through this series of image reconstruction operations, the quality of the infrared thermal imaging image can be effectively improved.

[0096] In step S15, perform a first image enhancement operation based on the reconstructed image data to obtain a first enhanced image, including:

[0097] Perform the operation of obtaining Gaussian distribution parameters based on the reconstructed image data to obtain Gaussian distribution parameters;

[0098] Perform the operation of constructing a second local energy model based on the Gaussian distribution parameters to obtain a second local energy model;

[0099] Perform the operation of extracting curve features based on the second local energy model to obtain curve features;

[0100] Perform the first image enhancement operation based on the curve features to obtain a first enhanced image;

[0101] The calculation formula for the operation of constructing the second local energy model is as follows:

[0102]

[0103] Wherein, represents the second local energy model, represents the weight of the second local energy model corresponding to the th pixel, represents the Gaussian distribution function of the second local energy model, represents the mean value in the Gaussian distribution function of the second local energy model, represents the standard deviation in the Gaussian distribution function of the second local energy model, represents the fused wavelet coefficient value.

[0104] It should be noted that in step S15, image processing technology is used to enhance the local details and overall quality of the reconstructed image data. First, Gaussian distribution parameters are extracted from the reconstructed image data, which include the mean value and the standard deviation. These parameters are crucial for describing the distribution characteristics of the image data in different regions and help understand the brightness distribution of the image. The mean value represents the average brightness of the image region, while the standard deviation describes the degree of dispersion of the brightness values. Next, the second local energy model is constructed using these Gaussian distribution parameters. This model highlights the local variations in the image by calculating the energy values of each pixel point and provides a basis for subsequent feature extraction and enhancement. Specifically, the formula is used to construct the model, where is the weight corresponding to the th pixel, is based on as the mean value, A Gaussian distribution function with standard deviation. This model can capture subtle changes in the image and provide information for image enhancement. After constructing the second local energy model, a curve feature extraction operation is performed. The purpose of this step is to extract key information such as edges and textures from the image. Curve feature extraction can be achieved through gradient operators, Laplacian operators, or model-based fitting techniques, which are not limited in the embodiments of the present invention. These techniques can help identify important structures in the image and provide precise guidance for image enhancement. Finally, the first image enhancement operation is performed in combination with the extracted curve features. This operation enhances the local contrast and sharpness of the image, making the detailed information of the image more prominent. In industrial inspection, the enhanced image can clearly show the thermal anomaly areas of the equipment and provide visual support for fault diagnosis. It can improve the usability and analysis accuracy of infrared thermal imaging images, making the final output image more in line with the requirements of practical applications.

[0105] In step S16, according to the first enhanced image, a second image enhancement operation is performed to obtain a second enhanced image, including:

[0106] According to the first enhanced image, a Laplacian operator processing operation is performed to obtain a Laplacian image;

[0107] According to the Laplacian image, an exponential transformation operation is performed to obtain an exponential transformation image;

[0108] According to the first enhanced image, a Gaussian distribution feature extraction operation is performed to obtain Gaussian distribution features;

[0109] According to the Gaussian distribution features and the exponential transformation image, a second data fusion operation is performed to obtain a second enhanced image.

[0110] It should be noted that in step S16, first, the first enhanced image is processed using the Laplacian operator to obtain a Laplacian image. The Laplacian operator is a second-order derivative operator that can highlight the fast-changing regions in the image, such as edges and textures, thereby enhancing the details of these regions. By calculating the Laplacian image, the high-frequency information in the image can be obtained, and this information corresponds to the significant features in the image. Next, an exponential transformation operation is performed on the Laplacian image to obtain an exponential transformation image. The exponential transformation is a non-linear mapping that can enhance the dark regions in the image while suppressing the bright regions, so that the contrast of the image can be improved, especially in the low gray-level regions of the image. The specific formula for the exponential transformation can be expressed as , where represents the pixel value of the original image, represents the coefficient of the exponential transformation, Denote the transformed pixel values. After obtaining the exponential transformation image, extract the Gaussian distribution features from the first enhanced image again. This step is similar to the Gaussian distribution parameter extraction in step S15, but this time it focuses on the image data after the first enhancement. The extracted Gaussian distribution features, including the mean and standard deviation, can further describe the distribution characteristics of the image data. Finally, perform a second data fusion operation on the Gaussian distribution features and the exponential transformation image to obtain the second enhanced image. The data fusion operation combines the statistical information of the Gaussian distribution features and the contrast enhancement effect of the exponential transformation image, so that the second enhanced image improves the overall contrast while retaining the image details. This fusion operation can be achieved through weighted average, superposition, or other more complex fusion algorithms, aiming to integrate the advantages of the two types of image data to obtain the final high-quality image.

[0111] It is worth noting that in the infrared thermal imaging image of industrial equipment, although the details are improved after the first enhancement operation, the overall contrast is still insufficient. Through the processing in step S16, first use the Laplace operator to extract the edge and texture information of the image, then enhance the dark area of the image through exponential transformation, and finally perform data fusion in combination with the Gaussian distribution features. The processed second enhanced image not only retains the detail information of the first enhanced image, but also improves the overall contrast, making the thermal anomaly area of the equipment more obvious.

[0112] In step S17, perform an image output operation according to the second enhanced image to obtain an output image.

[0113] It should be noted that in step S17, the final image output operation is performed to convert the processed second enhanced image into the final output image for further analysis or direct application to actual industrial detection tasks. The image output operation ensures the availability and practicality of the image data. It includes steps such as format conversion, image compression, and color adjustment to adapt to different display devices or storage requirements. In this step, first, the color and contrast of the second enhanced image are adjusted to the optimal state to clearly display all important details and features. This involves color correction of the image to ensure that the temperature information in the image is accurately reflected. In addition, the image needs to be compressed to reduce the occupancy of storage space or speed up the image transmission, while ensuring that the image quality is not greatly affected. For example, if the processed image is an infrared thermal imaging image of an electrical device, after the second enhancement operation, the thermal abnormal areas of the device can be clearly displayed. In step S17, the color of this image is adjusted to a range suitable for human eye observation, and pseudo-color is used to represent different temperature regions, with high-temperature regions displayed in red and low-temperature regions displayed in blue. At the same time, the image will also be appropriately compressed to facilitate the transmission of image data to a remote monitoring center or storage on a cloud server.

[0114] To facilitate the understanding of the present invention, some preferred embodiments of the present invention will be further described below.

[0115] In the industrial field, especially in the maintenance of power systems, regular thermal imaging detection of equipment is crucial. By analyzing the thermal distribution on the surface of the equipment, overheated areas can be detected in a timely manner to prevent potential faults and fire risks. The method of the present invention aims to improve the quality of infrared thermal imaging images to make them more suitable for industrial detection and fault diagnosis.

[0116] Step 1: Obtain infrared thermal imaging information. Use an infrared thermal imaging device to scan key equipment in the power system, such as transformers and switchgear. The thermal imaging information captured by the equipment is presented in the form of an image, showing the temperature differences in different regions, providing the raw data basis for subsequent processing.

[0117] Step 2: Perform noise reduction operation according to the infrared thermal imaging information. Through wavelet decomposition technology, the original image is decomposed into low-frequency coefficients and high-frequency coefficients. The low-frequency coefficients contain the overall structural information of the image, while the high-frequency coefficients contain detail and noise information. Then, threshold processing is applied to the high-frequency coefficients, setting the coefficients below the threshold to zero to remove noise, while retaining the coefficients above the threshold to maintain image details.

[0118] Step 3: Perform threshold function construction operations based on the denoised wavelet coefficient values. Construct a local weight threshold function and a high threshold function, which will be used in the image reconstruction process to finely process the image details and overall structure. The local weight threshold function assigns weights according to the difference between the pixel value and its neighborhood, while the high threshold function enhances or suppresses the wavelet coefficients based on the comparison result between the value of the local energy model and the segmentation threshold.

[0119] Step 4: Perform image reconstruction operations. Use the denoised wavelet coefficient values, the local weight threshold function, and the high threshold function to reconstruct the image and obtain the reconstructed image data.

[0120] Step 5: Perform the first image enhancement operation based on the reconstructed image data. By constructing a second local energy model and extracting curve features, enhance the local details of the image to make the thermal anomaly region more obvious.

[0121] Step 6: Perform the second image enhancement operation based on the first enhanced image. Use Laplacian operator processing and exponential transformation to further improve the overall contrast of the image, so that both the details and the overall structure of the image are improved.

[0122] Step 7: Perform image output operations based on the second enhanced image. Adjust the color and contrast to adapt to different display devices or storage requirements, and finally obtain a high-quality output image.

[0123] In summary, the present invention provides an infrared thermal imaging image processing method. First, obtain the temperature distribution data on the surface of the device through an infrared thermal imaging device to form an original thermal imaging image. Then, use wavelet decomposition technology to perform noise reduction processing on the image, effectively removing noise while retaining key detail information. Next, construct a local weight threshold function and a high threshold function, which play a key role in the image reconstruction process, enabling fine processing of the image detail information and overall structure. The image reconstruction operation combines the information of the denoised wavelet coefficient values, the local weight threshold function, and the high threshold function to accurately restore the image details and structure. The subsequent image enhancement operations further improve the local details and overall contrast of the image, making the thermal anomaly region more obvious. Finally, a high-quality image is obtained through the image output operation. It realizes noise smoothing processing while maintaining image details, provides accurate image data support for applications in industrial detection and other fields, and improves the accuracy and reliability of image analysis.

[0124] Referring to Figure 2 , the second embodiment of the present invention provides a data processing system for infrared thermal imaging, including:

[0125] A data acquisition module for acquiring infrared thermal imaging information;

[0126] A noise reduction module, configured to perform a noise reduction operation based on the infrared thermal imaging information to obtain denoised wavelet coefficient values;

[0127] A threshold function construction module, configured to perform a threshold function construction operation based on the denoised wavelet coefficient values to obtain a local weight threshold function and a high threshold function;

[0128] An image reconstruction module, configured to perform an image reconstruction operation based on the denoised wavelet coefficient values, the local weight threshold function, and the high threshold function to obtain reconstructed image data;

[0129] A first image enhancement module, configured to perform a first image enhancement operation based on the reconstructed image data to obtain a first enhanced image;

[0130] A second image enhancement module, configured to perform a second image enhancement operation based on the first enhanced image to obtain a second enhanced image;

[0131] An image output module, configured to perform an image output operation based on the second enhanced image to obtain an output image.

[0132] It should be noted that a data processing system for infrared thermal imaging provided in an embodiment of the present invention is used to execute all the process steps of a data processing method for infrared thermal imaging in the above embodiment. The working principles and beneficial effects of the two correspond one by one, and thus will not be elaborated herein.

[0133] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a second image enhancement program. When the processor executes the computer program, the steps in the above embodiments of various data processing methods for infrared thermal imaging are implemented, such as Figure 1 the step S11 shown. Alternatively, when the processor executes the computer program, the functions of each module / unit in the above device embodiments are implemented, such as the image output module.

[0134] Exemplarily, the computer program may be divided into one or more modules / units. The one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0135] The electronic device may be a computing device such as a desktop computer, a notebook, a palm computer, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine some components, or different components. For example, the electronic device may further include an input / output device, a network access device, a bus, etc.

[0136] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects various parts of the entire electronic device through various interfaces and lines.

[0137] The memory may be used to store the computer programs and / or modules. The processor realizes various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. Among them, the program storage area may store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area may store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory may include high-speed random access memory and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0138] Among them, if the modules / units integrated in the electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0139] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that they have a communication connection, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative effort.

[0140] The above-described specific embodiments have further elaborated on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above description is only for the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A data processing method for infrared thermal imaging, characterized in that: Executed by a computer, including: Obtain infrared thermal imaging information; Perform a noise reduction operation according to the infrared thermal imaging information to obtain a denoised wavelet coefficient value; According to the denoised wavelet coefficient value, a threshold function construction operation is performed to obtain a local weight threshold function and a high threshold function, including: Performing a first local energy model construction operation according to the denoised wavelet coefficient values ​​to obtain a first local energy model; Performing a high threshold function construction operation according to the denoised wavelet coefficient value and the first local energy model to obtain a high threshold function; According to the denoised wavelet coefficient value, a local weight threshold function construction operation is performed to obtain a local weight threshold function; The calculation formula of the first local energy model construction operation is as follows: The calculation formula for the high threshold function construction operation is as follows: The calculation formula for the local weight threshold function construction operation is as follows: in, represents the first local energy model, Indicates The weight of the first local energy model corresponding to pixels, represents the Gaussian distribution function of the first local energy model, represents the mean value in the Gaussian distribution function of the first local energy model, represents the standard deviation in the Gaussian distribution function of the first local energy model, Represents the denoised wavelet coefficient value; represents the high threshold function, represents the first amplitude parameter, represents the second amplitude parameter, represents the segmentation threshold; represents the local weight threshold function, express pixel values, express pixel value, represents the regularization parameter, Indicates The weight parameter of each pixel, represents the smoothing coefficient, Indicates The denoised wavelet coefficient value corresponding to the pixel, Represents the maximum value of the denoised wavelet coefficients; Performing an image reconstruction operation according to the denoising wavelet coefficient value, the local weight threshold function and the high threshold function to obtain reconstructed image data; Performing a first image enhancement operation according to the reconstructed image data to obtain a first enhanced image; Performing a second image enhancement operation according to the first enhanced image to obtain a second enhanced image; An image output operation is performed according to the second enhanced image to obtain an output image.

2. The data processing method for infrared thermal imaging according to claim 1, characterized in that: A denoising operation is performed according to the infrared thermal imaging information to obtain a denoised wavelet coefficient value, including: Performing a wavelet decomposition operation according to the infrared thermal imaging information to obtain a low-frequency coefficient and a first high-frequency coefficient; According to the first high frequency coefficient, a high frequency coefficient threshold processing operation is performed to obtain a second high frequency coefficient; A denoising wavelet coefficient value generation operation is performed according to the second high-frequency coefficient and the low-frequency coefficient to obtain a denoising wavelet coefficient value.

3. The data processing method for infrared thermal imaging according to claim 1, characterized in that: Performing an image reconstruction operation according to the denoising wavelet coefficient value, the local weight threshold function and the high threshold function to obtain reconstructed image data includes: According to the denoised wavelet coefficient value and the local weight threshold function, a local weight threshold processing operation is performed to obtain a local weight wavelet coefficient value; Performing a high threshold processing operation according to the denoised wavelet coefficient value and the high threshold function to obtain a high threshold wavelet coefficient value; Performing a first data fusion operation according to the local weighted wavelet coefficient value and the high threshold wavelet coefficient value to obtain a fused wavelet coefficient value; A wavelet reconstruction operation is performed according to the fused wavelet coefficient values ​​to obtain reconstructed image data.

4. The data processing method for infrared thermal imaging according to claim 1, characterized in that: Performing a first image enhancement operation according to the reconstructed image data to obtain a first enhanced image includes: According to the reconstructed image data, an operation of obtaining Gaussian distribution parameters is performed to obtain Gaussian distribution parameters; Performing a second local energy model construction operation according to the Gaussian distribution parameters to obtain a second local energy model; performing a curve feature extraction operation according to the second local energy model to obtain curve features; Performing a first image enhancement operation according to the curve feature to obtain a first enhanced image; The calculation formula of the second local energy model building operation is as follows: in, represents the second local energy model, Indicates The weight of the second local energy model corresponding to pixels, represents the second local energy model Gaussian distribution function, represents the mean value in the Gaussian distribution function of the second local energy model, represents the standard deviation in the Gaussian distribution function of the second local energy model, Represents the fused wavelet coefficient value.

5. The data processing method for infrared thermal imaging according to claim 1, characterized in that: Performing a second image enhancement operation according to the first enhanced image to obtain a second enhanced image includes: Performing a Laplacian operator processing operation on the first enhanced image to obtain a Laplacian image; Performing an exponential transformation operation according to the Laplace image to obtain an exponential transformation image; Performing a Gaussian distribution feature extraction operation according to the first enhanced image to obtain a Gaussian distribution feature; A second data fusion operation is performed according to the Gaussian distribution characteristics and the exponential transformation image to obtain a second enhanced image.

6. A data processing system for infrared thermal imaging, characterized in that: A data processing method for infrared thermal imaging for implementing any one of claims 1 to 5, comprising: A data acquisition module, used to acquire infrared thermal imaging information; A noise reduction module, used for performing noise reduction operation according to the infrared thermal imaging information to obtain a denoised wavelet coefficient value; A threshold function construction module is used to perform a threshold function construction operation according to the denoising wavelet coefficient value to obtain a local weight threshold function and a high threshold function; An image reconstruction module, used to perform an image reconstruction operation according to the denoising wavelet coefficient value, the local weight threshold function and the high threshold function to obtain reconstructed image data; A first image enhancement module, configured to perform a first image enhancement operation according to the reconstructed image data to obtain a first enhanced image; A second image enhancement module, configured to perform a second image enhancement operation according to the first enhanced image to obtain a second enhanced image; An image output module is used to perform an image output operation according to the second enhanced image to obtain an output image.

7. An electronic device, characterized in that: It comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements a data processing method for infrared thermal imaging as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute a data processing method for infrared thermal imaging as described in any one of claims 1 to 5.

Images