Circular coal yard temperature monitoring method, system and storage medium based on infrared image

Abstract

The present invention relates to a circular coal yard temperature monitoring method, system, and storage medium based on infrared images. The method comprises the following steps: S1: acquiring infrared images of different locations in the circular coal yard in real time, where the image content of the infrared images at different locations covers the entire circular coal yard; S2: obtaining the total number of images in the entire circular coal yard puzzle based on the principle of overlapping matching, and identifying the infrared images to be spliced; S3: splicing the infrared images to be spliced ​​to form a complete circular coal yard thermal image; and S4: extracting temperature data within the circular coal yard based on the circular coal yard thermal image. Compared with existing technologies, the present invention has the advantages of being effectively applicable to circular coal yards and having high temperature recognition accuracy.

Description

Technical Field

[0001] The present invention relates to the field of coal yard temperature monitoring, and in particular to a circular coal yard temperature monitoring method, system and storage medium based on infrared images. Background Art

[0002] Coal is a flammable material that is prone to spontaneous combustion during long-term storage. In thermal power generation companies, coal is the main production raw material and accounts for about 70% of their operating costs. If the problem of heat and spontaneous combustion during coal storage is not well solved, it will easily cause a huge loss of calorific value and extreme decomposition and spontaneous combustion, bringing immeasurable economic losses to the company. At the same time, the spontaneous combustion process will also produce toxic and harmful gases and even explosion hazards in the stacking site, which will have a significant impact on the company's personnel and production safety.

[0003] By installing multiple infrared thermal imaging devices to scan the entire coal yard in real time, each infrared temperature measuring device has a limited coverage area and cannot observe the entire coal yard. Its output data only shows the temperature extremes and averages in the current coverage area, making continuous temperature detection impossible. Furthermore, multiple infrared thermal imaging devices only present real-time infrared video, which cannot form an intuitive temperature graph for the entire coal yard. The spontaneous combustion process of a coal pile is quite complex, especially when approaching the auto-ignition point. It also involves high-order chemical reactions, making the calculation model very complex. When the temperature of the coal pile approaches the auto-ignition point, its temperature changes rapidly, and the coal pile will spontaneously combust very quickly. Conventional temperature monitoring methods cannot continuously monitor changes in the temperature of the coal pile, resulting in poor detection results and limited accuracy. Summary of the Invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a circular coal yard temperature monitoring method, system and storage medium based on infrared images.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A circular coal yard temperature monitoring method based on infrared images includes the following steps:

[0007] S1: Acquire infrared images of different positions of the circular coal yard in real time, where the image contents of the infrared images of different positions cover the entire circular coal yard;

[0008] S2: Based on the principle of overlapping matching, the total number of images of the entire circular coal yard puzzle is obtained to confirm the infrared images to be stitched:

[0009] S3: stitching the infrared images to be stitched to form a complete circular thermal image of the coal yard;

[0010] S4: extracting temperature data in the circular coal yard based on the thermal image of the circular coal yard.

[0011] Preferably, in step S1, a plurality of infrared thermal imaging devices are arranged in the circular coal yard to acquire infrared images, and the shooting ranges of the plurality of infrared thermal imaging devices overlap to cover the entire circular coal yard.

[0012] Preferably, the calculation formula for obtaining the total number of pictures in step S2 is:

[0013] L(x,y,σ)=G(x,y,σ)*I(x i ,y i ,a,b,c,h)*N

[0014]

[0015]

[0016] Among them, G(x,y,σ) is a Gaussian function with a predefined variation scale, I(x i ,y i ,a,b,c,h) is the original image, x i is the x-coordinate value of the image, y i is the y-direction coordinate value of the image, L(x,y,σ) is the sum of G(x,y,σ) and I(x i ,y i , a, b, c, h), x, y are the position parameters of the infrared thermal imaging device, σ is the standard deviation of the Gaussian function, a, b, c are the current pan / tilt angles of the infrared thermal imaging device, h is the installation height of the infrared thermal imaging device, N is the total number of images, m, n are the dimensional information of the Gaussian template, I p is the gray value of any point on the image, I i is the gray value of any point on the circle with a radius of 3 around the point, C i For I p with I i The relationship between the difference and the predefined threshold t.

[0017] Preferably, the steps of step S3 specifically include:

[0018] S31: performing texture recognition stitching, stitching images with a texture matching degree higher than a preset value according to the texture of the images to be stitched;

[0019] S32: Perform similarity recognition stitching, extract features from images that cannot be stitched by texture recognition, and stitch similar images together.

[0020] Preferably, in step S32, feature extraction and similar image stitching are performed on images that cannot be stitched by texture recognition based on the Ransac algorithm.

[0021] Preferably, the step S32 specifically includes:

[0022] S321: Gridding the images that cannot be texture-recognized and stitched, and establishing a central coordinate system within each grid;

[0023] S322: Calculate grayscale values ​​in four dimensional directions, match grayscale values ​​between four adjacent grids, and use grid intersections with high similarity as feature points;

[0024] S323: Performing linear transformation matrix transformation on the feature points, substituting the transformation matrix result into the position information of each image for fine-tuning, and forming an image that can be stitched for graphic stitching.

[0025] Preferably, the linear transformation matrix in step S323 is:

[0026]

[0027] Among them, k is the weight of the matching point in image 1, α' and β' are the image coordinate values ​​of the matching point in image 1, α and β are the image coordinate values ​​of the pixel point in image 2, and α τ , β τ is the image coordinate value of the transformed new image, H is the weight of the matching point in image 2, and h1, h2, h3, h4, h5, h6, h7, and h8 are the transformation matrices.

[0028] Preferably, the formula for the image position information in step S323 is:

[0029]

[0030] in, For the stitched image, are the images 1 and 2 to be stitched, is the x-coordinate value of the image, γ is the y-coordinate value of the image, w1 and w2 are the identification weights of image 1 and image 2 respectively, which are used to identify the coincidence requirements of feature point matching in each image.

[0031] A circular coal yard temperature monitoring system based on infrared images includes a memory and a processor. The memory stores a computer program, and the processor calls the computer program to execute the steps of the above-mentioned circular coal yard temperature monitoring method based on infrared images.

[0032] A computer-readable storage medium stores a program that can be loaded and executed by a processor to implement the above-mentioned circular coal yard temperature monitoring method based on infrared images.

[0033] Compared with the prior art, the present invention has the following advantages:

[0034] (1) The present invention can monitor the temperature data of a circular coal yard in real time, obtain the temperature data in the circular coal yard, and realize the combination of temperature data and stacking data, thereby providing a data basis for studying the types of coal and temperature changes. The present invention effectively splices and fuses the acquired infrared images to form a wide-viewing angle, complete, high-definition seamless splicing image information containing the information of each image sequence, thereby overcoming the limitations of the special site of the coal yard and the infrared thermal imaging equipment itself, and realizing the infrared video image display of the entire field;

[0035] (2) When performing infrared image stitching, the present invention minimizes the influence of grayscale differences between images on the fusion results, so that the final target image graphics are accurate and natural. A multi-level stitching form is adopted, first stitching is performed based on texture conditions, and then similarity recognition stitching is performed, which effectively improves the stitching effect, and effectively improves the accuracy and reliability of the subsequent temperature data in the coal yard based on the completion of the entire image recognition, thereby realizing the continuous temperature data acquisition and monitoring of the circular coal yard temperature. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a flow chart of the present invention;

[0037] Figure 2 This is a schematic diagram of the similarity recognition and splicing process of the present invention;

[0038] Figure 3 Schematic diagram of the structure of a circular coal yard temperature monitoring system based on infrared images in an embodiment of the present invention;

[0039] Figure 4 This is a complete circular coal yard thermal image spliced ​​together by the present invention. DETAILED DESCRIPTION

[0040] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. Note that the following embodiments are merely illustrative in nature and the present invention is not intended to limit its applicable objects or uses, and the present invention is not limited to the following embodiments.

[0041] Example

[0042] A circular coal yard temperature monitoring method based on infrared images. Due to the special location of the coal yard and the limitations of the infrared thermal imaging equipment itself, in order to realize the infrared video image display of the entire field, it is necessary to splice the images of all devices. The real-time image data of different locations in the coal yard collected by multiple devices in the coal yard are spliced ​​together to obtain the complete circular coal yard thermal image for temperature monitoring. Figure 1 As shown, the following steps are included:

[0043] S1: Acquire infrared images of different positions of the circular coal yard in real time, where the image contents of the infrared images of different positions cover the entire circular coal yard.

[0044] In step S1, multiple infrared thermal imaging devices are set up in the circular coal yard to acquire infrared images. The shooting ranges of the multiple infrared thermal imaging devices overlap and cover the entire circular coal yard.

[0045] In this embodiment, a maintenance platform within the enclosed circular storage yard structure serves as a support structure. Three dual-visible, positionable, high-precision infrared thermal imaging temperature detection devices are equally spaced on this platform, each responsible for monitoring the temperature of a portion of the coal pile within the yard. The infrared thermal imaging devices feature 320x240 pixel pixels, a 25µm pixel size, a response band of 7.5µm to 13µm, an operating temperature range of -40°C to 60°C, a storage temperature range of -45°C to 70°C, a 360° pan / tilt angle, a 180° tilt angle, and an accuracy of 0.01°. The devices can be installed at a height of 50 meters.

[0046] S2: Based on the overlapping matching principle, the total number of images of the entire circular coal yard puzzle is obtained to confirm the infrared images to be stitched.

[0047] The calculation formula for obtaining the total number of pictures in step S2 is:

[0048] L(x,y,σ)=G(x,y,σ)*I(x i ,y i ,a,b,c,h)*N

[0049]

[0050]

[0051] Among them, G(x,y,σ) is a Gaussian function with a predefined variation scale, I(x i ,y i ,a,b,c,h) is the original image, x i is the x-coordinate value of the image, y i is the y-direction coordinate value of the image, L(x,y,σ) is the sum of G(x,y,σ) and I(x i ,y i , a, b, c, h), x, y are the position parameters of the infrared thermal imaging device, σ is the standard deviation of the Gaussian function, a, b, c are the current pan / tilt angles of the infrared thermal imaging device, h is the installation height of the infrared thermal imaging device, N is the total number of images, m, n are the dimensional information of the Gaussian template, I p is the gray value of any point on the image, I i is the gray value of any point on the circle with a radius of 3 around the point, C i For I pwith I i The relationship between the difference and the predefined threshold t.

[0052] Infrared thermal imaging equipment converts visible light image information into thermal imaging information through infrared radiation, which passes through the atmosphere, optical imaging, photoelectric conversion, and grayscale processing. Therefore, the real-time image information obtained by infrared thermal imaging is grayscale image, with low resolution and unclear outline edges. In order to ensure that multiple adjacent images can truly restore the scene during the splicing process, it is necessary to preprocess the infrared image and match the multiple adjacent photos to achieve the maximum possible overlap between adjacent images, thus completing the splicing of multiple images. During the image acquisition process, the more overlap between adjacent images, the better the splicing effect after the images overlap. However, more image overlap will result in an increase in the number of infrared images, which will in turn cause a decrease in image processing speed and increase system operation time. Therefore, in order to quickly achieve image information matching during the preprocessing process without reducing the overall operating efficiency of the system, the present invention obtains image information according to the principle of 50% overlap matching between adjacent images. By utilizing the resolution, focal length, coal yard size, and equipment installation location of the infrared thermal imaging equipment, a data model of the photos that need to be matched is derived. Based on this model, each infrared thermal imaging device is required to capture a specified number of images within the coverage area of ​​the device within a specified time, and the total number of images required to complete the entire coal yard puzzle is calculated. Furthermore, based on the situation of each device, the amplitude and frequency of the pan / tilt movement of each device are calculated to confirm the infrared images to be spliced.

[0053] S3: stitch the infrared images to be stitched to form a complete circular thermal image of the coal yard, such as Figure 4 shown.

[0054] The purpose of image stitching is to superimpose multiple thermal images to form a complete thermal image of the storage yard. During this process, the stitching process needs to be smoothly transitioned to complete the grayscale fusion. The images acquired during the shooting process will show slight differences in the grayscale of the picture due to differences in direction, angle, illumination, etc. Therefore, there will be matching errors in the image matching process, and it is impossible to achieve accurate matches. The process of image stitching and fusion is to minimize the impact of grayscale differences between images on the fusion results, so that the final target image graphics are accurate and natural. To this end, it is necessary to analyze various situations in the stitched image and deal with different situations separately. The present invention adopts the form of first performing texture recognition stitching and then performing similarity recognition stitching to improve the stitching effect.

[0055] The steps of S3 include:

[0056] S31: Perform texture recognition stitching, and stitch images with a texture matching degree higher than a preset value based on the texture of the images to be stitched. Texture is the main element that constitutes a picture and is widely used in visible light image processing. However, since the spatial resolution and detection capability of the thermal imaging system are lower than those of the visible light imaging system, the infrared image has a low contrast compared to the visible light image, the target scene occupies a small proportion in the image, and the target is not easy to identify. At the same time, the interference of the external environment and the noise source of the thermal imaging system itself cause a variety of noises to exist in the infrared image, which directly affects the extraction of texture in the image. In the present invention, during the image stitching process, by comparing multiple overlapping graphics and using the matching degree of texture recognition, graphics with more matches are directly spliced, and for graphics with problems, image matching is performed by building a reasonable similarity model.

[0057] S32: Figure 2 As shown, similarity recognition and splicing are performed, feature extraction is performed on images that cannot be spliced ​​by texture recognition, and similar images are spliced.

[0058] In the special environment of the coal yard, the pan-tilt head is used to drive the thermal imager. The changes in the viewing angle and posture of the imaging lens will cause spatial geometric changes between images. The main geometric transformations include displacement, rotation and zoom caused by different lens focal lengths between images. In addition, due to different shooting times and angles, these factors lead to partial or no matching of image splicing. In the present invention, reasonable similarity feature points are found to determine the similarity of the photos to be matched. Reasonable similarity feature points are analyzed for rationality through the coal types in the coal yard. First, the picture that cannot be texture analyzed and its surrounding pictures are obtained, and features are extracted from the picture. The Ransac algorithm is used to find the reasonable feature points in the picture, and the feature points in several pictures are combined to match the graphics. The specific steps are as follows:

[0059] S321: Gridding the images that cannot be texture-recognized and stitched, and establishing a central coordinate system within each grid;

[0060] S322: Calculate grayscale values ​​in four dimensional directions, match grayscale values ​​between four adjacent grids, and use grid intersections with high similarity as feature points;

[0061] S323: Perform linear transformation matrix transformation on the feature points, substitute the transformation matrix result into the position information of each image for fine-tuning, and form an image that can be stitched for graphic stitching, wherein the linear transformation matrix is:

[0062]

[0063] Among them, k is the weight of the matching point in image 1, α' and β' are the image coordinate values ​​of the matching point in image 1, α and β are the image coordinate values ​​of the pixel point in image 2, and α τ , β τ is the image coordinate value of the transformed new image, H is the weight of the matching point in image 2, and h1, h2, h3, h4, h5, h6, h7, and h8 are the transformation matrices.

[0064] The formula for the image position information is:

[0065]

[0066] in, For the stitched image, are the images 1 and 2 to be stitched, is the x-coordinate value of the image, γ is the y-coordinate value of the image, w1 and w2 are the identification weights of image 1 and image 2 respectively, which are used to identify the coincidence requirements of feature point matching in each image.

[0067] S4: extracting temperature data in the circular coal yard based on the thermal image of the circular coal yard.

[0068] By splicing complete infrared thermal images of coal temperature, continuous and complete temperature monitoring results can be obtained for the entire circular coal yard and coal storage area. Although the temperature figure is only the surface temperature of the coal pile, under normal circumstances, when the temperature rises at any position of the coal pile (the bottom or the middle of the coal seam), the temperature rises vertically to the surface of the coal seam. Combined with manual measurement of the temperature inside the coal pile, according to the data of the monitoring results over a continuous period of time, it is found that when the temperature change of the coal surface reaches 2 degrees, it can be basically judged that the maximum temperature inside the coal seam has changed by about 20 degrees. Based on this situation, a linear correlation change model between the internal temperature and surface temperature of the coal pile is made, and the continuous monitoring data of the coal pile temperature is introduced into the model data. Combined with the consideration of external factors such as the storage time, season, and coal type of the coal pile in the coal storage yard, long-term continuous temperature monitoring big data in the coal yard can be formed. Through the analysis of these big data, the temperature change model is revised, which can provide data support for the subsequent research on the calorific value change law, temperature rise law, and spontaneous combustion characteristics law of different types of coal under the same storage conditions in the power plant, the calorific value change law, temperature rise law, and spontaneous combustion characteristics law of the same type of coal under different storage methods, and the long-term storage method and spontaneous combustion prevention technology of coal prone to spontaneous combustion.

[0069] The present invention also provides a circular coal yard temperature monitoring system based on infrared images, such as Figure 3 As shown, it includes a memory and a processor, wherein the memory stores a computer program, and the processor calls the computer program to execute the steps of the above-mentioned method for monitoring the temperature of a circular coal yard based on infrared images. Specifically, as Figure 3The monitoring system includes multiple infrared thermal imagers and an infrared monitoring client. The infrared monitoring client is equipped with the aforementioned memory and processor. Infrared images captured by the infrared thermal imagers are sent to the infrared monitoring client via an on-site control cabinet and a central switch for processing, splicing, and temperature data acquisition. Furthermore, the present application also includes a display and a video monitor that, under the control of the infrared monitoring client, performs video monitoring and displays infrared images of the circular coal yard. The present invention also includes a video server and a data server to store and process the monitoring videos and acquired temperature data.

[0070] In addition, the present invention also provides a computer-readable storage medium storing a program that can be loaded and executed by a processor to implement the above-mentioned method for monitoring circular coal yard temperature based on infrared images.

[0071] The above embodiments are merely examples and do not limit the scope of the present invention. These embodiments can be implemented in various other ways, and various omissions, replacements, and changes can be made without departing from the technical concept of the present invention.

Claims

1. A circular coal yard temperature monitoring method based on infrared images, characterized in that: The following steps are involved: S1: Acquire infrared images of different positions of the circular coal yard in real time, where the image contents of the infrared images of different positions cover the entire circular coal yard; S2: Based on the principle of overlapping matching, the total number of images of the entire circular coal yard puzzle is obtained to confirm the infrared images to be stitched: S3: stitching the infrared images to be stitched to form a complete circular thermal image of the coal yard; S4: Extract the temperature data in the circular coal yard based on the thermal image of the circular coal yard. The calculation formula for obtaining the total number of pictures in step S2 is: L(x,y,σ)=G(x,y,σ)*I(x i ,y i ,a,b,c,h)*N Among them, G(x,y,σ) is a Gaussian function with a predefined variation scale, I(x i ,y i ,a,b,c,h) is the original image, x i is the x-coordinate value of the image, y i is the y-direction coordinate value of the image, L(x,y,σ) is the sum of G(x,y,σ) and I(x i ,y i , a, b, c, h), x, y are the position parameters of the infrared thermal imaging device, σ is the standard deviation of the Gaussian function, a, b, c are the current pan / tilt angles of the infrared thermal imaging device, h is the installation height of the infrared thermal imaging device, N is the total number of images, m, n are the dimensional information of the Gaussian template, I p is the gray value of any point on the image, I i is the gray value of any point on the circle with a radius of 3 around the point, C i For I p with I i The relationship between the difference and the predefined threshold t.

2. The method for monitoring temperature of a circular coal yard based on infrared images according to claim 1, characterized in that: In the step S1, a plurality of infrared thermal imaging devices are set up in the circular coal yard to acquire infrared images, and the shooting ranges of the plurality of infrared thermal imaging devices overlap to cover the entire circular coal yard.

3. The method for monitoring temperature of a circular coal yard based on infrared images according to claim 1, characterized in that: The steps of step S3 specifically include: S31: performing texture recognition stitching, stitching images with a texture matching degree higher than a preset value according to the texture of the images to be stitched; S32: Perform similarity recognition stitching, extract features from images that cannot be stitched by texture recognition, and stitch similar images together.

4. The method for monitoring temperature of a circular coal yard based on infrared images according to claim 3 is characterized in that: In step S32, feature extraction and similar image stitching are performed on images that cannot be texture-recognized and stitched based on the Ransac algorithm.

5. The method for monitoring temperature of a circular coal yard based on infrared images according to claim 3 is characterized in that: The step S32 specifically includes: S321: Gridding the images that cannot be texture-recognized and stitched, and establishing a central coordinate system within each grid; S322: Calculate grayscale values ​​in four dimensional directions, match grayscale values ​​between four adjacent grids, and use grid intersections with high similarity as feature points; S323: Performing linear transformation matrix transformation on the feature points, substituting the transformation matrix result into the position information of each image for fine-tuning, and forming an image that can be stitched for graphic stitching.

6. The method for monitoring temperature of a circular coal yard based on infrared images according to claim 5, characterized in that: The linear transformation matrix in step S323 is: Among them, k is the weight of the matching point in image 1, α' and β' are the image coordinate values ​​of the matching point in image 1, α and β are the image coordinate values ​​of the pixel point in image 2, and α τ , β τ is the image coordinate value of the transformed new image, H is the weight of the matching point in image 2, and h1, h2, h3, h4, h5, h6, h7, and h8 are the transformation matrices.

7. The method for monitoring temperature of a circular coal yard based on infrared images according to claim 5, characterized in that: The formula for the image position information in step S323 is: in, For the stitched image, are the images 1 and 2 to be stitched, is the x-coordinate value of the image, γ is the y-coordinate value of the image, w1 and w2 are the identification weights of image 1 and image 2 respectively, which are used to identify the coincidence requirements of feature point matching in each image.

8. A circular coal yard temperature monitoring system based on infrared images, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor calls the computer program to execute the steps of a circular coal yard temperature monitoring method based on infrared images as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that The device stores a program that can be loaded and executed by a processor to implement a circular coal yard temperature monitoring method based on infrared images as described in any one of claims 1 to 7.

Images