Infrared temperature difference sensing image processing method and system, and electronic device

Abstract

The application discloses an infrared temperature difference sensing image processing method and system and electronic equipment, and relates to the technical field of image processing.The processing steps are as follows: S1, acquiring surrounding images based on a three-proofing mobile phone, automatically adjusting filter parameters for the features of the images based on an adaptive filtering algorithm, and obtaining infrared images; S2, performing layered processing on the infrared images based on a guided filtering technology, dividing the images into different levels based on the local features of the images, and including background images and subject images; and S3, for the subject images, mainly enhancing the edge and detail information thereof, and for the background images.The adaptive filtering algorithm can automatically adjust parameters according to image features, can accurately remove noises of different degrees, can retain details, can avoid the problems of excessive blurring or incomplete denoising in traditional filtering, can quickly discover temperature difference abnormalities during outdoor exploration, can timely detect equipment temperature faults in industrial scenes, and can meet the application requirements in multiple scenes.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, specifically to an infrared temperature difference sensing image processing method and system, and electronic equipment. Background Technology

[0002] With the widespread adoption of smart devices, rugged phones designed for extreme environments, featuring waterproof, drop-proof, and explosion-proof capabilities, are increasingly being used in outdoor adventures, mountaineering, and outdoor work. However, current rugged phones generally lack the ability to sense ambient temperature or use thermal imaging technology to assess the approximate temperature of objects in the surrounding environment. This has led to a growing demand for high-precision, high-stability infrared temperature difference sensing image processing technology. Existing technologies focus on improving image processing speed and accuracy to meet the demands of harsh environments. Currently, rugged phones primarily use infrared sensors for temperature difference imaging. Traditional infrared temperature difference sensing image processing methods struggle to maintain high accuracy under strong light or extreme temperature conditions, and their slow processing speed prevents real-time information feedback. Therefore, we propose an infrared temperature difference sensing image processing method, system, and electronic device. Summary of the Invention

[0003] To solve the above-mentioned technical problems, an infrared temperature difference sensing image processing method and system, as well as an electronic device, are provided. This technical solution solves the above-mentioned problems.

[0004] To achieve the above objectives, the technical solution adopted by this invention is: an infrared temperature difference sensing image processing method, the processing steps of which are: S1. Based on the rugged mobile phone, acquire surrounding images, and use an adaptive filtering algorithm to automatically adjust the filtering parameters of the image features to obtain an infrared image; S2. Based on guided filtering technology, the infrared image is processed in layers. Based on the local features of the image, the image is divided into different layers, including the background image and the main image. S3. For the main image, focus on enhancing its edges and details; for the background image, adjust its brightness and contrast to highlight the difference between the target and the background. S4. Perform fusion processing on the processed subject image and background image to restore the overall image; S5. Divide the pixels in the image into different categories and divide the image into different temperature zones, and display them in real time on the rugged phone screen.

[0005] Preferably, the step of automatically adjusting the filter parameters in step S1 is as follows: Read the acquired image data, calculate the mean and variance to obtain the features that reflect the image; Based on the obtained features, the filtering parameters are adjusted. For areas with high noise, a large filtering window and strong filtering intensity are used; for areas with rich texture, a small filtering window and weak filtering intensity are used. The adjusted parameters are filtered, and the output is an infrared image.

[0006] Preferably, the filter parameter adjustment distinguishes the features of image regions by setting a judgment standard value. When the variance of a local region is greater than the judgment standard value, the region is considered to have high noise, and a large-size filter window and a strong filter intensity are used. When the variance of a local region is less than or equal to the judgment standard value, the region is considered to have rich texture, and a small-size filter window and a weak filter intensity are used.

[0007] Preferably, the filtering operation uses a weighted average method to filter the image. For each pixel in the image, based on the adjusted filtering intensity of the region where the pixel is located, 1 - the filtering intensity is multiplied by the original pixel value to calculate the average value of all pixel values ​​in the region corresponding to the adjusted filtering window size centered on the pixel. The average value is multiplied by the filtering intensity, and the two results are added together to obtain the new value of the pixel after filtering. All pixels are calculated in this way, and the output is the processed infrared image.

[0008] Preferably, the layering process in step S2 is as follows: Acquire an infrared image, set the guided filtering parameters, and perform the guided filtering algorithm on the input infrared image based on the set parameters. After guided filtering, a smoothed image is obtained, which is the background image. The subject image is obtained by subtracting the preliminary shape of the background image obtained through guided filtering from the original infrared image.

[0009] Preferably, in step S3, the main image processing is based on the Sobel edge detection algorithm to outline the target contour, and the Laplacian sharpening operator is used to enhance the gray-scale abrupt change parts and refine the texture structure. Background image processing uses histogram equalization to redistribute grayscale values, widening the grayscale dynamic range and improving the overall visual effect; adaptive gamma correction is used to dynamically adjust the gamma value based on local features, optimizing the display effect of different brightness areas, resulting in the processed background image.

[0010] Preferably, the image fusion processing step in step S4 is as follows: Determine the fusion weights, perform a weighted fusion calculation on all pixels of the image, and multiply the gray values ​​of the corresponding pixels of the subject and the background according to their weights and then sum them up; Gaussian blur smoothing transition technology is used to process the boundary region; The image fusion process is completed by using color balance algorithms and brightness equalization operations.

[0011] Preferably, in step S5, the pixels are divided into high temperature, medium temperature and low temperature categories, and the membership probability is given. Based on the clustering results, the pixel grayscale value is converted into the corresponding color through a preset color mapping table, and the pixels are divided into different temperature regions.

[0012] An infrared temperature difference sensing image processing system, applied to any of the above-mentioned infrared temperature difference sensing image processing methods, includes: The adjustment module is configured to use an adaptive filtering algorithm to automatically adjust the filtering parameters of the image features to obtain an infrared image. The layering module is configured to perform layered processing on infrared images based on guided filtering technology, dividing the image into different layers. The enhancement processing module is configured to focus on enhancing the edge and detail information of the main image, and adjust the brightness and contrast of the background image. The fusion processing module is configured to fuse the processed subject image and background image to restore the overall image. The display module is configured to divide the image into different temperature zones and display them in real time on the screen of a rugged phone.

[0013] An infrared temperature difference sensing image processing electronic device, applied to any of the above-mentioned infrared temperature difference sensing image processing methods, includes a mobile terminal, a camera, a display screen, a software terminal, and an infrared temperature measurement module, wherein the infrared temperature measurement module adopts a vanadium oxide uncooled detector, performs full-screen temperature measurement, and has two temperature measurement ranges, namely -20℃-150℃ and 0℃-550℃.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention proposes an adaptive filtering algorithm that automatically adjusts parameters based on image features, accurately removing noise of varying degrees while preserving details, avoiding the problems of excessive blurring or incomplete noise removal found in traditional filtering. Utilizing guided filtering layering, it effectively preserves subject edges while smoothing the background, preventing background interference with subject information. Differential processing of the subject and background enhances subject edge details and optimizes background brightness and contrast, significantly improving the distinction between the target and background. It also prevents the target from becoming detached from the background, providing users with intuitive temperature distribution information. This allows for rapid detection of abnormal temperature differences during outdoor exploration and timely detection of equipment temperature faults in industrial settings, meeting the needs of multiple application scenarios. Attached Figure Description

[0015] Figure 1 This is a flowchart of the processing steps of the present invention; Figure 2 This is a system framework diagram for the present invention. Detailed Implementation

[0016] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0017] Reference Figure 1 and Figure 2 As shown, an infrared temperature difference sensing image processing method includes the following processing steps: S1. Based on the rugged mobile phone, acquire surrounding images, and use an adaptive filtering algorithm to automatically adjust the filtering parameters of the image features to obtain an infrared image; S2. Based on guided filtering technology, the infrared image is processed in layers. Based on the local features of the image, the image is divided into different layers, including the background image and the main image. S3. For the main image, focus on enhancing its edges and details; for the background image, adjust its brightness and contrast to highlight the difference between the target and the background. S4. Perform fusion processing on the processed subject image and background image to restore the overall image; S5. Divide the pixels in the image into different categories and divide the image into different temperature zones, and display them in real time on the rugged phone screen.

[0018] The adaptive filtering algorithm in this application automatically adjusts parameters based on image features, accurately removing noise of varying degrees while preserving details, avoiding the problems of excessive blurring or incomplete noise removal found in traditional filtering. Guided filtering layering effectively preserves subject edges while smoothing the background, preventing background interference with subject information. Differential processing of the subject and background enhances subject edge details and optimizes background brightness and contrast, significantly improving the distinction between the target and background. Fusion processing ensures the integrity of the overall image structure and natural transitions, preventing the target from becoming detached from the background. Clustering algorithms classify pixels, divide temperature regions, and display them in real time, providing users with intuitive temperature distribution information. This allows for rapid detection of temperature anomalies during outdoor adventures and timely detection of equipment temperature faults in industrial settings, meeting the needs of multiple application scenarios. The entire method is specifically designed for rugged phones, with optimized algorithms for high computational efficiency, enabling rapid image processing and real-time output in extreme environments. Furthermore, it adapts hardware performance requirements to phone configurations, ensuring stable operation.

[0019] The automatic adjustment of filter parameters in step S1 is as follows: Read the acquired image data, calculate the mean and variance to obtain the features that reflect the image; Based on the obtained features, the filtering parameters are adjusted. For areas with high noise, a large filtering window and strong filtering intensity are used; for areas with rich texture, a small filtering window and weak filtering intensity are used. The adjusted parameters are filtered, and the output is an infrared image.

[0020] This application obtains features by calculating the local mean and variance of the image, which can accurately identify noisy areas and textured areas. For noisy areas, a large filtering window and strong filtering intensity are used to effectively remove various types of noise, such as salt and pepper and Gaussian noise. In different scenarios such as outdoor exploration and industrial operations, the noise situation faced by the image is complex and varied. This application avoids the blindness caused by using fixed parameter filtering and reduces unnecessary calculations.

[0021] The filter parameter adjustment distinguishes the features of image regions by setting judgment standard values. When the variance of a local region is greater than the judgment standard value, the region is considered to have high noise, and a large-size filter window and strong filter intensity are used. When the variance of a local region is less than or equal to the judgment standard value, the region is considered to have rich texture, and a small-size filter window and weak filter intensity are used.

[0022] The specific calculation formula is as follows: Let the input infrared image be ,in Represents the coordinates of pixels in an image. , , and These are the height and width of the image, respectively. Local variance calculation, for images in pixels A size centered on Local window Its local variance The calculation formula is: in, This is the mean of the local window, calculated as follows: Set a standard value for judgment, and set a standard value for variance judgment. ; Adjusting the filter window size, let the maximum filter window size be... The minimum filter window size is ( ), then the size of the filtering window The adjustment formula is: Filter strength adjustment, assuming the maximum filter strength is... The minimum filter strength is ( Then the filter strength The adjustment formula is: The filtering operation uses a weighted average filtering method for each pixel in the image. Filtered pixel values The calculation formula is: in This indicates a floor operation; after this filtering operation, the output is... This is the processed infrared image; Based on the variance characteristics of local image regions, the above formula automatically adjusts the filter window size and filter intensity to achieve adaptive filtering.

[0023] The filtering operation uses a weighted average method to filter the image. For each pixel in the image, the average value of all pixel values ​​in the region corresponding to the adjusted filter window size centered on the pixel is calculated by multiplying the adjusted filter intensity by 1 based on the filter intensity of the region where the pixel is located. The average value is then multiplied by the filter intensity, and the two results are added together to obtain the new value of the pixel after filtering. All pixels are processed in this way, and the output is the processed infrared image.

[0024] The specific calculation formula is as follows: Let the input infrared image be... ,in Represents the coordinates of pixels in an image. , , and These are the height and width of the image, representing the number of pixels in the image. Centered on, the adjusted filter window size is ( (The adjusted window side length based on local features), and the adjusted filter intensity are: Calculation based on pixels Centered on, size is The average value of all pixels within the region : in This indicates a round-down operation; Calculate the filtered pixels new value : By analyzing all pixels in the image , , All calculations are performed according to the above formulas to obtain the final processed infrared image. .

[0025] The layering process in step S2 is as follows: Acquire an infrared image, set the guided filtering parameters, and perform the guided filtering algorithm on the input infrared image based on the set parameters. After guided filtering, a smoothed image is obtained, which is the background image. The subject image is obtained by subtracting the preliminary shape of the background image obtained through guided filtering from the original infrared image.

[0026] This application obtains a smoothed background image through guided filtering, which can effectively remove noise and detail interference in the image and highlight the overall background information of the image. At the same time, by subtracting the background image from the original image to obtain the subject image, the edge and texture details of the target object can be highlighted, making the subject and background features in the image clearer. This facilitates subsequent targeted processing of images at different levels. Dividing the image into two levels, the background image and the subject image, is equivalent to a preliminary decomposition and simplification of the image. In subsequent processing, different processing methods can be adopted according to the characteristics of different levels. For example, the brightness and contrast of the background image can be adjusted, and the edge and detail information of the subject image can be enhanced. This can improve the efficiency and accuracy of image processing and avoid information confusion or loss that may occur when processing the entire image uniformly.

[0027] In step S3, the main image processing is based on the Sobel edge detection algorithm to outline the target contour, and the Laplacian sharpening operator is used to enhance the gray-level abrupt changes and refine the texture structure. Background image processing uses histogram equalization to redistribute grayscale values, widening the grayscale dynamic range and improving the overall visual effect; adaptive gamma correction is used to dynamically adjust the gamma value based on local features, optimizing the display effect of different brightness areas, resulting in the processed background image.

[0028] The Sobel edge detection algorithm in this application can accurately locate the edges of target objects by calculating pixel gradients, and can clearly outline the contours and define the target boundaries even in complex backgrounds. The Laplacian sharpening operator focuses on enhancing the gray-level abrupt changes, which can effectively refine the target texture structure, making the texture and shape details of the main object clearer, making the target more eye-catching in the image, and making it easier for users to quickly capture key information. Histogram equalization redistributes the gray values ​​of the background image, uniformly expanding the originally concentrated gray-level range, greatly widening the gray-level dynamic range, and making the background image clearer and brighter.

[0029] The image fusion processing steps in step S4 are as follows: Determine the fusion weights, perform a weighted fusion calculation on all pixels of the image, and multiply the gray values ​​of the corresponding pixels of the subject and the background according to their weights and then sum them up; Gaussian blur smoothing transition technology is used to process the boundary region; The image fusion process is completed by using color balance algorithms and brightness equalization operations.

[0030] This application determines the fusion weights and performs weighted fusion calculations. Based on the differences in importance between the subject and the background, it scientifically allocates their proportions in the image, so that the gray values ​​of the subject and background pixels are organically combined according to their weights. This avoids a sense of separation between the subject and the background, ensures the coherence and integrity of the overall image structure, and allows the target object to blend naturally into the background. Gaussian blur is used to smooth the transition of the boundary areas, which can effectively eliminate obvious splicing marks and jagged defects produced after the subject image and background image are merged. By blurring the boundary pixels, the transition between different areas is more gentle and natural, enhancing the visual smoothness of the image, improving the overall visual experience, and avoiding distraction of the user's attention due to abrupt boundaries.

[0031] In step S5, pixels are divided into high temperature, medium temperature and low temperature categories, and the membership probability is given. Based on the clustering results, the pixel grayscale values ​​are converted into corresponding colors through a preset color mapping table, and the pixels are divided into different temperature regions.

[0032] This application transforms the abstract grayscale values ​​of infrared images into intuitive temperature information by dividing pixels into high-temperature, medium-temperature, and low-temperature categories and assigning them different colors. The membership probability can further quantify the possibility that each pixel belongs to a different temperature category, making the image information richer and more accurate.

[0033] An infrared temperature difference sensing image processing system, applied to any of the above-mentioned infrared temperature difference sensing image processing methods, includes: The adjustment module is configured to use an adaptive filtering algorithm to automatically adjust the filtering parameters of the image features to obtain an infrared image. The layering module is configured to perform layered processing on infrared images based on guided filtering technology, dividing the image into different layers. The enhancement processing module is configured to focus on enhancing the edge and detail information of the main image, and adjust the brightness and contrast of the background image. The fusion processing module is configured to fuse the processed subject image and background image to restore the overall image. The display module is configured to divide the image into different temperature zones and display them in real time on the screen of a rugged phone.

[0034] The adjustment module of this application utilizes an adaptive filtering algorithm to automatically adjust the filtering parameters based on image features. In complex environments, such as strong light or electromagnetic interference that causes a lot of image noise, it can specifically remove noise, preserve image details, and make the image clearer, providing a high-quality foundation for subsequent processing. The adaptive filtering algorithm enables the system to adapt to different scenarios and environments. Under different outdoor lighting and temperature conditions, or in indoor environments with different equipment interference, it can automatically adjust the filtering parameters to ensure stable image quality.

[0035] An infrared temperature difference sensing image processing electronic device, applied to any of the above-mentioned infrared temperature difference sensing image processing methods, includes a mobile terminal, a camera, a display screen, a software terminal, and an infrared temperature measurement module, wherein the infrared temperature measurement module adopts a vanadium oxide uncooled detector, performs full-screen temperature measurement, and has two temperature measurement ranges, namely -20℃-150℃ and 0℃-550℃.

[0036] This application integrates a mobile terminal, camera, display screen, software, and infrared temperature measurement module into a single unit, forming a complete infrared temperature difference sensing image processing system. Users no longer need to carry multiple separate devices, making it convenient for use in various scenarios and improving the portability and efficiency of the device. It adopts a vanadium oxide uncooled detector, which has the advantages of high sensitivity, high resolution, and good stability. It can accurately detect the infrared radiation emitted by an object, convert it into an electrical signal, and then accurately measure the temperature of the object, providing a reliable data foundation for subsequent image processing and temperature analysis.

[0037] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. An infrared temperature difference sensing image processing method, characterized in that, The processing steps are as follows: S1. Based on the rugged mobile phone, acquire surrounding images, and use an adaptive filtering algorithm to automatically adjust the filtering parameters of the image features to obtain an infrared image; S2. Based on guided filtering technology, the infrared image is processed in layers. Based on the local features of the image, the image is divided into different layers, including the background image and the main image. S3. For the main image, focus on enhancing its edges and details; for the background image, adjust its brightness and contrast to highlight the difference between the target and the background. S4. Perform fusion processing on the processed subject image and background image to restore the overall image; S5. Divide the pixels in the image into different categories and divide the image into different temperature zones, and display them in real time on the rugged phone screen.

2. The infrared temperature difference sensing image processing method according to claim 1, characterized in that, The automatic adjustment of filter parameters in step S1 is as follows: Read the acquired image data, calculate the mean and variance to obtain the features that reflect the image; Based on the obtained features, the filtering parameters are adjusted. For areas with high noise, a large filtering window and strong filtering intensity are used; for areas with rich texture, a small filtering window and weak filtering intensity are used. The adjusted parameters are filtered, and the output is an infrared image.

3. The infrared temperature difference sensing image processing method according to claim 2, characterized in that, The filter parameter adjustment distinguishes the features of image regions by setting judgment standard values. When the variance of a local region is greater than the judgment standard value, the region is considered to have high noise, and a large-size filter window and strong filter intensity are used. When the variance of a local region is less than or equal to the judgment standard value, the region is considered to have rich texture, and a small-size filter window and weak filter intensity are used.

4. The infrared temperature difference sensing image processing method according to claim 2, characterized in that, The filtering operation uses a weighted average method to filter the image. For each pixel in the image, the average value of all pixel values ​​in the region corresponding to the adjusted filter window size centered on the pixel is calculated by multiplying the adjusted filter intensity by 1 based on the filter intensity of the region where the pixel is located. The average value is then multiplied by the filter intensity, and the two results are added together to obtain the new value of the pixel after filtering. All pixels are processed in this way, and the output is the processed infrared image.

5. The infrared temperature difference sensing image processing method according to claim 1, characterized in that, The layering process in step S2 is as follows: Acquire an infrared image, set the guided filtering parameters, and perform the guided filtering algorithm on the input infrared image based on the set parameters. After guided filtering, a smoothed image is obtained, which is the background image. The subject image is obtained by subtracting the preliminary shape of the background image obtained through guided filtering from the original infrared image.

6. The infrared temperature difference sensing image processing method according to claim 1, characterized in that, In step S3, the main image processing is based on the Sobel edge detection algorithm to outline the target contour, and the Laplacian sharpening operator is used to enhance the gray-level abrupt changes and refine the texture structure. Background image processing uses histogram equalization to redistribute grayscale values, widening the grayscale dynamic range and improving the overall visual effect; adaptive gamma correction is used to dynamically adjust the gamma value based on local features, optimizing the display effect of different brightness areas, resulting in the processed background image.

7. The infrared temperature difference sensing image processing method according to claim 1, characterized in that, The image fusion processing steps in step S4 are as follows: Determine the fusion weights, perform a weighted fusion calculation on all pixels of the image, and multiply the gray values ​​of the corresponding pixels of the subject and the background according to their weights and then sum them up; Gaussian blur smoothing transition technology is used to process the boundary region; The image fusion process is completed by using color balance algorithms and brightness equalization operations.

8. The infrared temperature difference sensing image processing method according to claim 1, characterized in that, In step S5, pixels are divided into high temperature, medium temperature and low temperature categories, and the membership probability is given. Based on the clustering results, the pixel grayscale values ​​are converted into corresponding colors through a preset color mapping table, and the pixels are divided into different temperature regions.

9. An infrared temperature difference sensing image processing system, applied to an infrared temperature difference sensing image processing method according to any one of claims 1-8, characterized in that, include: The adjustment module is configured to use an adaptive filtering algorithm to automatically adjust the filtering parameters of the image features to obtain an infrared image. The layering module is configured to perform layered processing on infrared images based on guided filtering technology, dividing the image into different layers. The enhancement processing module is configured to focus on enhancing the edge and detail information of the main image, and adjust the brightness and contrast of the background image. The fusion processing module is configured to fuse the processed subject image and background image to restore the overall image. The display module is configured to divide the image into different temperature zones and display them in real time on the screen of a rugged phone.

10. An infrared temperature difference sensing image processing electronic device, applied to an infrared temperature difference sensing image processing method according to any one of claims 1-8, characterized in that, It includes a mobile terminal, camera, display screen, software and infrared temperature measurement module. The infrared temperature measurement module uses a vanadium oxide uncooled detector, which measures the temperature of the entire screen. The temperature measurement range includes two sets, namely -20℃-150℃ and 0℃-550℃.

Images