Monitoring data processing method, device, equipment and medium based on industrial Internet of Things

Abstract

This application discloses a monitoring data processing method, apparatus, device, and medium based on the industrial Internet of Things, including: obtaining an original image of factory monitoring and determining the brightness value of each pixel in the original image; determining a brightness evaluation index for each pixel based on the brightness value of each pixel using a preset formula, and determining a pixel sequence based on the brightness evaluation index of each pixel; starting from the first pixel in the pixel sequence, performing multiple region divisions according to a preset strategy using each pixel in the pixel sequence as a reference point to obtain a plurality of regions; calculating the atmospheric light value corresponding to each region, processing the original image based on the atmospheric light value of each region and using a dark channel prior algorithm to obtain an enhanced image. This application can reduce the interference caused by the highlight areas generated by strong light sources in the factory on the dehazing and enhancement of industrial monitoring images.

Description

Technical Field

[0001] The present application relates to the field of monitoring data processing, and in particular to a monitoring data processing method, device, equipment and medium based on the Industrial Internet of Things. Background Art

[0002] Because factory production is prone to generating smoke and dust, existing industrial surveillance systems are severely impacted when detecting equipment anomalies and human activity. Smoke not only obscures the view of surveillance cameras but can also degrade image quality, leading to false or missed alerts and hindering the timely identification of potential safety hazards. Therefore, to ensure the accurate and effective operation of surveillance systems, enhanced dehazing of captured images is necessary.

[0003] The Dark Channel Prior (DCP) algorithm is a technique used for image enhancement and dehazing. It uses the darkest pixel values ​​in an image to estimate the concentration of haze, thereby restoring a clear image. However, when using the DCP algorithm to process factory surveillance images, due to the complex background, the DCP algorithm's estimation of atmospheric light can be biased, resulting in color distortion in the restored image.

[0004] The above content is only used to assist in understanding the technical solution of this application and does not constitute an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of this application is to provide a monitoring data processing method, device, equipment and medium based on the Industrial Internet of Things, aiming to solve the technical problem that when using a dark channel prior algorithm to process factory monitoring images, due to the complex background in the factory monitoring images, the estimation of atmospheric light in the dark channel prior algorithm will be biased, resulting in distortion of the restored image color.

[0006] To achieve the above-mentioned purpose, the present application provides a monitoring data processing method based on the industrial Internet of Things, including: obtaining an original image of factory monitoring and determining the brightness value of each pixel in the original image; using a preset formula and based on the brightness value of each pixel to determine the brightness evaluation index of each pixel, and determining a pixel sequence based on the brightness evaluation index of each pixel, wherein each pixel in the pixel sequence is arranged in descending order according to the brightness evaluation index; starting from the first pixel in the pixel sequence, taking each pixel in the pixel sequence as a reference point, performing multiple region divisions according to a preset strategy to obtain a plurality of regions, wherein after each region division is performed, the pixel sequence is updated to eliminate the pixels falling into the divided region, and the region division is performed again based on the updated sequence; the atmospheric light value of each region is calculated respectively, and the original image is processed based on the atmospheric light value of each region and using a dark channel prior algorithm to obtain an enhanced image.

[0007] Optionally, determining the brightness value of each pixel in the original image includes: converting the original image into an HSV image; and using the V channel value of each pixel in the HSV image as the brightness value of each pixel in the original image.

[0008] Optionally, the brightness evaluation index of each pixel point is determined based on the brightness value of each pixel point using a preset formula, including: determining the brightness distribution degree of each pixel point based on the difference between the brightness value of each pixel point and the brightness value of its adjacent pixel points; and using the ratio of the brightness value of each pixel point to the brightness distribution degree as the brightness evaluation index of each pixel point.

[0009] Optionally, determining the brightness distribution degree of each pixel based on the difference between the brightness value of each pixel and the brightness value of its adjacent pixels includes: using the following formula (1) to respectively calculate the upper continuity, lower continuity, left continuity and right continuity of each pixel:

[0010]

[0011] Where: P i Indicates the continuity of the i-th pixel in the P direction, Indicates the distance between the i-th pixel and the x-th pixel in the P direction, V i Represents the brightness value of the i-th pixel, V i Px Indicates the brightness value of the x-th pixel in the P direction of the i-th pixel, X iP Represents the number of pixels in the P direction of the i-th pixel point, where the P direction is any of the upper, lower, left, and right directions; the brightness distribution degree of each pixel point is obtained by summing the upper continuity, lower continuity, left continuity, and right continuity of each pixel point.

[0012] Optionally, the process of single area division includes: selecting a reference point; calculating the brightness difference between the reference point and the pixel points adjacent to it to obtain a first growth point, wherein the first growth point is the pixel point with the largest brightness difference between the reference point and the adjacent growth points, and the reference point is the parent node of the first growth point; taking the first growth point as the child growth point, calculating the brightness difference of each pixel point adjacent to it and after excluding the parent node to obtain a second growth point, wherein the second growth point is the pixel point with the largest brightness difference between the child growth point and the adjacent growth points after excluding the parent node; repeating the growth process of the child growth point until, among the obtained growth points, the number of growth points with brightness values ​​greater than a preset brightness average is the same as the number of growth points with brightness values ​​less than the brightness average.

[0013] Optionally, the preset brightness mean is the brightness mean of all pixels in the original image.

[0014] Optionally, the separately calculating the atmospheric light value of each area includes: for any area, calculating the dark channel value of each pixel point in the area, and determining a dark channel image of the area based on the dark channel value of each pixel point; determining a number of candidate pixel points of the area based on the dark channel image of the area; and determining the atmospheric light value of the area based on the number of candidate pixel points.

[0015] In addition, to achieve the above-mentioned purpose, the present application also provides a monitoring data processing device based on the industrial Internet of Things, including: an image acquisition module, used to acquire the original image of factory monitoring and determine the brightness value of each pixel in the original image; a pixel sequence acquisition module, used to use a preset formula and determine the brightness evaluation index of each pixel based on the brightness value of each pixel, and determine the pixel sequence based on the brightness evaluation index of each pixel, wherein the pixels in the pixel sequence are arranged in descending order according to the brightness evaluation index; a region division module, used to start from the first pixel in the pixel sequence, and use each pixel in the pixel sequence as a reference point to perform multiple region divisions according to a preset strategy to obtain several regions, wherein after each region division is performed, the pixel sequence is updated to eliminate the pixels falling into the divided region, and the region division is performed again based on the updated sequence; an image processing enhancement module, used to calculate the atmospheric light value of each region respectively, and process the original image based on the atmospheric light value of each region and using a dark channel prior algorithm to obtain an enhanced image.

[0016] The present application also provides a monitoring data processing device based on the Industrial Internet of Things, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned monitoring data processing method based on the Industrial Internet of Things.

[0017] The present application also provides a computer-readable storage medium, comprising: a computer program stored therein, wherein when the computer program is executed by a processor, the computer program implements the above-mentioned monitoring data processing method based on the industrial Internet of Things.

[0018] This application proposes a monitoring data processing method, apparatus, device, and medium based on the Industrial Internet of Things. This method divides the original factory monitoring image based on the brightness values ​​and brightness distribution of the pixels in the image. The resulting small regions include both highlight areas and other areas, resolving the problem of color distortion in the restored image due to interference from highlight areas with the atmospheric light estimation of the entire image. By combining the highlight areas with other areas within the resulting small regions, this application also allows for greater detail to be retained during image enhancement and restoration, more consistent with actual lighting scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a flowchart of a monitoring data processing method based on the Industrial Internet of Things according to one embodiment of the present application;

[0020] Figure 2 This is a structural block diagram of a monitoring data processing device based on the Industrial Internet of Things according to one embodiment of the present application;

[0021] Figure 3 The present invention is a schematic structural diagram of a monitoring data processing device based on the Industrial Internet of Things according to one embodiment of the present application.

[0022] The realization of the objectives, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0023] It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0024] The dark channel prior algorithm assumes that atmospheric light is relatively uniform. However, in a factory environment, there are many strong light sources (such as bright light from welding and cutting). This is reflected in some bright areas in the factory monitoring images. These bright areas will make it impossible to accurately estimate the atmospheric light when processing factory monitoring images using the dark channel prior algorithm, and the resulting enhanced restored image will have color distortion.

[0025] To solve the above problems, the present application provides a monitoring data processing method, device, equipment and medium based on the Industrial Internet of Things. The present application solution is introduced in detail below.

[0026] Figure 1 This is a flowchart of a monitoring data processing method based on the industrial Internet of Things according to an embodiment of the present application, referring to Figure 1 , the monitoring data processing method based on industrial Internet of Things may include the following steps:

[0027] S1, obtain the original image of the factory monitoring and determine the brightness value of each pixel in the original image.

[0028] During the specific implementation process, factory monitoring images are acquired frame by frame, wherein the original images may contain a variety of information such as equipment status, personnel activity records, environmental data, production progress, etc.

[0029] In one embodiment, in step S1, determining the brightness value of each pixel in the original image may specifically include:

[0030] S11, converting the original image into an HSV image;

[0031] S12. Using the V channel value of each pixel in the HSV image as the brightness value of each pixel in the original image.

[0032] It should be noted that the light generated during welding and cutting is usually high-brightness but may be a combination of multiple colors. Compared with RGB images, selecting the V channel in HSV can ignore the influence of color and focus on brightness, thereby improving the recognition accuracy of bright light. Using a single channel can also simplify calculations and increase processing speed.

[0033] S2, using a preset formula and based on the brightness value of each pixel point to determine a brightness evaluation index of each pixel point, and determining a pixel sequence based on the brightness evaluation index of each pixel point, wherein the pixels in the pixel sequence are arranged in descending order according to the brightness evaluation index.

[0034] S3, starting from the first pixel in the pixel sequence, performing multiple region divisions according to a preset strategy using each pixel in the pixel sequence as a reference point to obtain a plurality of regions, wherein after each region division is performed, the pixel sequence is updated to eliminate pixels falling into the divided region, and region division is performed again based on the updated sequence.

[0035] In this exemplary embodiment, a dark channel a priori algorithm is used to enhance the surveillance image. The specific steps of the dark channel a priori algorithm are as follows:

[0036] First, a dark channel map is calculated for each pixel from the input image, usually by selecting the minimum color value of each pixel within a small window;

[0037] Next, the atmospheric illumination is estimated using the dark channel map, usually by selecting the brightest pixel value in the dark channel;

[0038] Then, the transmittance is calculated by introducing the relationship between atmospheric light and dark channel and performing smoothing to remove noise;

[0039] Finally, the original image, transmittance, and atmospheric light are combined to restore the image to reduce haze or other blurring effects, thereby improving image clarity and visibility.

[0040] Among them, when highlights appear in the input image, the atmospheric light estimation will be overestimated, thereby introducing excessive brightness when restoring the image, resulting in overall image color distortion and detail loss, reducing the quality and authenticity of the restored image.

[0041] Based on this, in an embodiment of the present application, the brightness evaluation index of each pixel point is first determined. The brightness evaluation index can be used to characterize the possibility of each pixel point falling into the highlight area, and then the pixel sequence is determined based on the brightness evaluation index of each pixel point. Starting from the first pixel point in the pixel sequence, multiple region divisions are performed according to a preset strategy with each pixel point in the pixel sequence as a reference point to obtain several regions, so that the highlight area is divided into other adjacent areas. In this way, after the region division, the area where the highlight area is located includes both part of the highlight area and part of other areas, that is, the highlight part and the normal part are included in the same area, which can smooth the brightness changes of the area, better simulate the normal lighting environment, and the obtained atmospheric value is more accurate.

[0042] In one embodiment, in step S2, determining the brightness evaluation index of each pixel point based on the brightness value of each pixel point using a preset formula may specifically include:

[0043] S21, determining the brightness distribution degree of each pixel based on the brightness value of each pixel and the brightness value difference of its adjacent pixels;

[0044] S22. Using the ratio of the brightness value of each pixel to the brightness distribution degree as a brightness evaluation index of each pixel.

[0045] The brightness distribution can represent the difference in brightness between a pixel and its neighboring pixels. For example, when a pixel falls within a high-brightness area, the brightness values ​​of the pixel and its neighboring pixels are both larger, and the difference between the brightness values ​​of the pixel and its neighboring pixels is smaller, indicating greater consistency. However, when a pixel is a noise point, the brightness values ​​of the pixel and its neighboring pixels will differ, and the difference between the brightness values ​​of the pixel and its neighboring pixels is larger, indicating lower consistency. Therefore, the brightness evaluation index can represent the probability that each pixel is a pixel in the high-brightness area.

[0046] In a specific implementation process, taking the i-th pixel as an example, the following formula (1) can be used to calculate the upper continuity, lower continuity, left continuity and right continuity of the i-th pixel respectively:

[0047]

[0048] Where: P i Indicates the continuity of the i-th pixel in the P direction, Indicates the distance between the i-th pixel and the x-th pixel in the P direction, V i Represents the brightness value of the i-th pixel, V i Px Indicates the brightness value of the x-th pixel in the P direction of the i-th pixel, X iPRepresents the number of pixels in the P direction of the i-th pixel, where the P direction is any one of the directions of up, down, left, and right.

[0049] Specifically, the upper continuity of the i-th pixel can be calculated using the following formula (1-1):

[0050]

[0051] Where, γ i Indicates the upper continuity of the i-th pixel, Indicates the distance between the i-th pixel and the u-th pixel above it, V i Represents the brightness value of the i-th pixel, V i u Indicates the brightness value of the u-th pixel above the i-th pixel, U i Indicates the number of pixels above the i-th pixel.

[0052] Furthermore, the bottom continuity of the i-th pixel can be calculated using the following formula (1-2):

[0053]

[0054] Where λ i Indicates the bottom continuity of the i-th pixel, Indicates the distance between the i-th pixel and the b-th pixel below it, V i Represents the brightness value of the i-th pixel, V i b Indicates the brightness value of the bth pixel below the i-th pixel, B i Indicates the number of pixels below the i-th pixel.

[0055] Furthermore, the left continuity of the i-th pixel can be calculated using the following formula (1-3):

[0056]

[0057] Where, δ i Indicates the left continuity of the i-th pixel, Indicates the distance between the i-th pixel and the l-th pixel on its left, V i Represents the brightness value of the i-th pixel, V i l Indicates the brightness value of the lth pixel to the left of the i-th pixel, L i Indicates the number of pixels to the left of the i-th pixel.

[0058] Furthermore, the right continuity of the i-th pixel can be calculated using the following formula (1-4):

[0059]

[0060] Where, τ i Indicates the right continuity of the i-th pixel, Indicates the distance between the i-th pixel and the r-th pixel to its right, V i Represents the brightness value of the i-th pixel, V i r Represents the brightness value of the rth pixel to the right of the i-th pixel, R i Indicates the number of pixels to the right of the i-th pixel.

[0061] Furthermore, the brightness distribution degree of each pixel point is obtained by summing up the upper continuity, lower continuity, left continuity and right continuity of each pixel point.

[0062] Furthermore, the ratio of the brightness value of each pixel to the brightness distribution degree is used as the brightness evaluation index of each pixel.

[0063] It should be noted that the ratio of the brightness value of each pixel to the brightness distribution degree is used as the brightness evaluation index of each pixel. When the brightness value of a pixel is large and the brightness distribution degree is small, it means that the pixel is more likely to fall into the brightness area.

[0064] Furthermore, the pixels are arranged in descending order according to the brightness evaluation index to form a pixel sequence, where each element in the pixel sequence is a pixel position. It should be noted that the positions of the pixels can be represented using a Z-shaped arrangement from left to right and from top to bottom in the original image, so that the positions of the pixels can be represented by one-dimensional data.

[0065] In a specific implementation process, if there are two or more pixels with the same brightness evaluation index, the pixels with the same brightness evaluation index are arranged in ascending order of pixel position in the pixel sequence.

[0066] Furthermore, starting from the first pixel in the pixel sequence, multiple region divisions are performed according to a preset strategy using each pixel in the pixel sequence as a reference point to obtain a plurality of regions. It should be noted that the regions obtained by the division do not overlap with each other.

[0067] It should be noted that since the pixels in the pixel sequence are arranged from large to small according to the brightness evaluation index, the first pixel in the pixel sequence can represent the highlight area in the original factory monitoring image. In this way, starting from the first pixel in the pixel sequence to perform region division, it can be ensured that the highlight area is first divided into multiple sub-areas, and the sub-areas are merged with other areas around them, that is, the several small areas obtained first include not only the highlight area but also the normal area.

[0068] In one embodiment, in step S3, the process of single area division may specifically include:

[0069] S31, select a reference point;

[0070] S32. Calculate the brightness difference between the reference point and its adjacent pixels to obtain a first growth point, wherein the first growth point is a pixel point having the maximum brightness difference between each adjacent growth point and the reference point, and the reference point is a parent node of the first growth point;

[0071] S33: Taking the first growing point as a child growing point, calculating the brightness difference of each pixel adjacent to the first growing point after excluding the parent node to obtain a second growing point, wherein the second growing point is the pixel having the maximum brightness difference with the child growing point among the adjacent growing points after excluding the parent node;

[0072] S34. Repeat the growth process of the sub-growing points until the number of the growing points having brightness values ​​greater than the preset brightness average is the same as the number of the growing points having brightness values ​​less than the brightness average.

[0073] In a specific implementation process, the preset brightness mean is the brightness mean of all pixels in the original image.

[0074] In the above step S32, growth is performed with the selected reference point as the starting point to obtain multiple growth points, wherein the growth point with the largest brightness difference with the reference point among the multiple growth points is the first growth point, and the multiple growth points are the eight-neighborhood pixel points of the reference point.

[0075] In the above step S33, growth is performed with the first growth point as the starting point to obtain multiple growth points, wherein, after removing the parent node of the first growth point, i.e., the reference point, the growth point with the largest brightness difference with the first growth point among the multiple growth points is the second growth point, and the multiple growth points are the seven-neighborhood pixel points of the first growth point after removing the parent node of the first growth point.

[0076] S4, respectively calculating the atmospheric light value of each region, and processing the original image based on the atmospheric light value of each region and using a dark channel prior algorithm to obtain an enhanced image.

[0077] The enhanced image is the original image of the factory monitoring after dehazing and enhancement.

[0078] It should be noted that when there are bright light areas in factory monitoring due to equipment welding, cutting, etc., by dividing the factory monitoring image into regions, the interference of the highlight areas on the atmospheric light estimation of the entire image can be effectively reduced. When dividing the factory monitoring image into regions, if only the highlight areas are considered, that is, the highlight areas are separated from other areas when dividing the regions, the lighting relationship between the highlight areas and other areas will be ignored when calculating the atmospheric light, which will lead to the loss or distortion of details in other areas during the image restoration process. Processing the highlight areas separately from other areas will also result in the transition of the highlight areas near the edge where the highlight areas are divided from other areas, which may affect the atmospheric light estimation of other areas, resulting in blurred edges or unnatural transitions between the highlight areas and other areas in the restored image.

[0079] Based on this, the embodiment of the present application is based on the prior condition that illumination in normal scenes is usually gradual rather than sudden. Through region division, it is equivalent to cutting the highlight area and recombining it with the non-highlight area of ​​the neighborhood to obtain several small areas. In other words, each of the obtained small areas includes both highlight pixels and non-highlight pixels, thereby effectively reducing the interference of the highlight area on the atmospheric light estimation of the entire image. In addition, combining the highlight pixels with other non-highlight pixels in the small area helps to retain more details when restoring the image, which is more consistent with the actual lighting scene.

[0080] In one embodiment, in step S4, respectively calculating the atmospheric light value corresponding to each area may specifically include:

[0081] S41. For any region, calculating a dark channel value of each pixel in the region, and determining a dark channel image of the region based on the dark channel value of each pixel;

[0082] S42, determining a number of candidate pixel points in the area based on the dark channel image of the area;

[0083] S43. Determine the atmospheric light value of the area based on a number of candidate pixel points.

[0084] It should be noted that the calculation of the atmospheric light value of each area can be performed according to the general steps of the dark channel prior algorithm. It can be understood that the dark channel prior algorithm usually estimates the atmospheric light of the entire image, while this embodiment divides the image into several areas, that is, the dark channel prior algorithm is used to estimate the atmospheric light of each area separately.

[0085] Furthermore, the original image is processed based on the atmospheric light value of each region and a dark channel prior algorithm is used to obtain an enhanced image.

[0086] Based on the above embodiments, Figure 2 This is a structural block diagram of a monitoring data processing device based on the industrial Internet of Things according to an embodiment of the present application, such as Figure 2 As shown, the monitoring data processing device based on the industrial Internet of Things may include: an image acquisition module 210, a pixel sequence acquisition module 220, a region division module 230, and an image processing enhancement module 240, wherein:

[0087] The image acquisition module 210 is used to obtain the original image of the factory monitoring and determine the brightness value of each pixel in the original image;

[0088] The pixel sequence acquisition module 220 is configured to determine a brightness evaluation index for each pixel based on the brightness value of each pixel using a preset formula, and to determine a pixel sequence based on the brightness evaluation index of each pixel, wherein the pixels in the pixel sequence are arranged in descending order according to the brightness evaluation index;

[0089] The region division module 230 is configured to perform multiple region divisions starting from the first pixel in the pixel sequence and sequentially using each pixel in the pixel sequence as a reference point according to a preset strategy to obtain a plurality of regions, wherein after each region division, the pixel sequence is updated to remove pixels falling into the divided region, and region division is performed again based on the updated sequence;

[0090] The image processing and enhancement module 240 is used to calculate the atmospheric light value of each region respectively, and process the original image based on the atmospheric light value of each region and using a dark channel priori algorithm to obtain an enhanced image.

[0091] In an exemplary embodiment, the image acquisition module 210 may further convert the original image into an HSV image; and use the V channel value of each pixel point in the HSV image as the brightness value of each pixel point in the original image.

[0092] In an exemplary embodiment, the pixel sequence acquisition module 220 can also be used to determine the brightness evaluation index of each pixel point based on the brightness value of each pixel point using a preset formula, including: determining the brightness distribution degree of each pixel point based on the difference between the brightness value of each pixel point and the brightness value of its adjacent pixel points; and using the ratio of the brightness value of each pixel point to the brightness distribution degree as the brightness evaluation index of each pixel point.

[0093] In an exemplary embodiment, the pixel sequence acquisition module 220 can use the following formula (1) to respectively

[0094]

[0095] Where: P i Indicates the continuity of the i-th pixel in the P direction, Indicates the distance between the i-th pixel and the x-th pixel in the P direction, V i Represents the brightness value of the i-th pixel, V i Px Indicates the brightness value of the x-th pixel in the P direction of the i-th pixel, X iP Represents the number of pixels in the P direction of the i-th pixel point, where the P direction is any one of the upper, lower, left and right directions; the pixel sequence acquisition module 220 can also be used to accumulate the upper continuity, lower continuity, left continuity and right continuity of each pixel point to obtain the brightness distribution degree of each pixel point.

[0096] In an exemplary embodiment, the area division module 230 can also be used to select a reference point; calculate the brightness difference between the reference point and the pixel points adjacent to it to obtain a first growth point, wherein the first growth point is the pixel point with the largest brightness difference between the reference point and the adjacent growth points, and the reference point is the parent node of the first growth point; take the first growth point as the child growth point, calculate the brightness difference of each pixel point adjacent to it and after excluding the parent node to obtain a second growth point, wherein the second growth point is the pixel point with the largest brightness difference between the child growth point and the adjacent growth points after excluding the parent node; repeat the growth process of the child growth point until, among the obtained growth points, the number of growth points with brightness values ​​greater than the preset brightness average is the same as the number of growth points with brightness values ​​less than the brightness average.

[0097] In an exemplary embodiment, the preset brightness mean value in the region division module 230 is the brightness mean value of all pixels in the original image.

[0098] In an exemplary embodiment, the image processing enhancement module 240 can also be used to calculate the dark channel value of each pixel point in any area, determine the dark channel image of the area based on the dark channel value of each pixel point; determine several candidate pixel points of the area based on the dark channel image of the area; and determine the atmospheric light value of the area based on the several candidate pixel points.

[0099] Those skilled in the art should understand that the division of the various modules in the embodiment is merely a division of logical functions, and in actual application, they can be fully or partially integrated into one or more actual carriers, and these modules can all be implemented in the form of software called through a processing unit, or all in the form of hardware, or in the form of a combination of software and hardware. It should be noted that the modules in the monitoring data processing device based on the industrial Internet of Things in this embodiment correspond one-to-one to the steps in the monitoring data processing method based on the industrial Internet of Things in the aforementioned embodiment. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned monitoring data processing method based on the industrial Internet of Things, and will not be repeated here.

[0100] Based on the above embodiments, Figure 3 FIG. 1 is a structural diagram of a monitoring data processing device based on the industrial Internet of Things according to an embodiment of the present application, as shown in FIG. Figure 3 As shown, the electronic device may include: a processor (processor) 310, a communication interface (Communications Interface) 320, a memory (memory) 330 and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 310 can call the logic instructions in the memory 330 to execute a monitoring data processing method based on the industrial Internet of Things, which includes: obtaining an original image of factory monitoring and determining the brightness value of each pixel in the original image; using a preset formula and based on the brightness value of each pixel to determine the brightness evaluation index of each pixel, and determining a pixel sequence based on the brightness evaluation index of each pixel, wherein each pixel in the pixel sequence is arranged in descending order according to the brightness evaluation index; starting from the first pixel in the pixel sequence, each pixel in the pixel sequence is used as a reference point and multiple region divisions are performed according to a preset strategy to obtain a plurality of regions, wherein after each region division is performed, the pixel sequence is updated to eliminate the pixels falling into the divided region, and the region division is performed again based on the updated sequence; the atmospheric light value of each region is calculated respectively, and the original image is processed based on the atmospheric light value of each region and using a dark channel prior algorithm to obtain an enhanced image.

[0101] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0102] On the basis of the above embodiments, on the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the monitoring data processing method based on the industrial Internet of Things provided by the above methods, the method including: obtaining an original image of factory monitoring and determining the brightness value of each pixel in the original image; using a preset formula and based on the brightness value of each pixel to determine a brightness evaluation index of each pixel, and determining a pixel sequence based on the brightness evaluation index of each pixel, wherein each pixel in the pixel sequence is arranged in descending order according to the brightness evaluation index; starting from the first pixel in the pixel sequence, using each pixel in the pixel sequence as a reference point according to a preset strategy to perform multiple region divisions to obtain a plurality of regions, wherein after each region division is performed, the pixel sequence is updated to eliminate pixels falling into the divided region, and region division is performed again based on the updated sequence; respectively calculating the atmospheric light value of each region, and processing the original image based on the atmospheric light value of each region and using a dark channel prior algorithm to obtain an enhanced image.

[0103] On the basis of the above embodiments, on another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the monitoring data processing method based on the industrial Internet of Things provided by the above methods, the method comprising: obtaining an original image of factory monitoring and determining the brightness value of each pixel in the original image; determining a brightness evaluation index of each pixel based on the brightness value of each pixel using a preset formula, and determining a pixel sequence based on the brightness evaluation index of each pixel, wherein each pixel in the pixel sequence is arranged in descending order according to the brightness evaluation index; starting from the first pixel in the pixel sequence, taking each pixel in the pixel sequence as a reference point, performing multiple region divisions according to a preset strategy to obtain a plurality of regions, wherein after each region division is performed, the pixel sequence is updated to eliminate pixels falling into the divided region, and region division is performed again based on the updated sequence; calculating the atmospheric light value of each region respectively, processing the original image based on the atmospheric light value of each region and using a dark channel prior algorithm to obtain an enhanced image.

[0104] The above are only preferred embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A monitoring data processing method based on industrial Internet of Things, characterized in that: include: Obtain the original image of the factory monitoring and determine the brightness value of each pixel in the original image; Determining a brightness evaluation index for each pixel based on the brightness value of each pixel using a preset formula, and determining a pixel sequence based on the brightness evaluation index of each pixel, wherein the pixels in the pixel sequence are arranged in descending order according to the brightness evaluation index; Starting from the first pixel in the pixel sequence, performing multiple region divisions according to a preset strategy using each pixel in the pixel sequence as a reference point to obtain a plurality of regions, wherein after each region division, the pixel sequence is updated to remove pixels falling into the divided region, and region division is performed again based on the updated sequence; Calculate the atmospheric light value of each area respectively, and process the original image based on the atmospheric light value of each area and use the dark channel prior algorithm to obtain an enhanced image; The method of determining the brightness evaluation index of each pixel point based on the brightness value of each pixel point by using a preset formula includes: Determine the brightness distribution of each pixel based on the brightness value of each pixel and the brightness value difference of its adjacent pixels; The ratio of the brightness value of each pixel to the brightness distribution degree is used as the brightness evaluation index of each pixel; The process of single zoning includes: Select a reference point; Calculating a brightness difference between the reference point and adjacent pixels to obtain a first growth point, wherein the first growth point is a pixel point having a maximum brightness difference with the reference point among the adjacent growth points, and the reference point is a parent node of the first growth point; Taking the first growing point as a child growing point, calculating the brightness difference of each pixel adjacent to the first growing point and excluding the parent node to obtain a second growing point, wherein the second growing point is the pixel with the largest brightness difference with the child growing point among the adjacent growing points excluding the parent node; The growth process of the sub-growing points is repeatedly performed until, among the obtained growing points, the number of growing points having brightness values ​​greater than a preset brightness average is the same as the number of growing points having brightness values ​​less than the brightness average.

2. The monitoring data processing method based on the industrial Internet of Things according to claim 1, characterized in that: Determining the brightness value of each pixel in the original image includes: Convert the original image to HSV image; The V channel value of each pixel in the HSV image is used as the brightness value of each pixel in the original image.

3. The monitoring data processing method based on the industrial Internet of Things according to claim 1, characterized in that: The determining of the brightness distribution degree of each pixel based on the difference between the brightness value of each pixel and the brightness values ​​of its adjacent pixels includes: The upper continuity, lower continuity, left continuity, and right continuity of each pixel are calculated using the following formula (1): Where: Indicates the Pixels in Continuity of direction, Indicates the Pixels and their Direction The distance of pixels, Indicates the The brightness value of each pixel, Indicates the pixels Direction The brightness value of each pixel, Indicates the pixels The number of pixels in the direction, where The direction is any one of the upper side, the lower side, the left side and the right side; The brightness distribution of each pixel is obtained by summing up the upper continuity, lower continuity, left continuity and right continuity of each pixel.

4. The monitoring data processing method based on the industrial Internet of Things according to claim 1, characterized in that: The preset brightness mean is the brightness mean of all pixels in the original image.

5. The monitoring data processing method based on the industrial Internet of Things according to claim 1, characterized in that: The calculating of the atmospheric light value of each area respectively includes: For any area, calculating the dark channel value of each pixel in the area, and determining a dark channel image of the area based on the dark channel value of each pixel; determining a plurality of candidate pixel points in the region based on a dark channel image of the region; An atmospheric light value of the area is determined based on a number of candidate pixel points.

6. A monitoring data processing device based on industrial Internet of Things, characterized in that: The monitoring data processing device based on the industrial Internet of Things includes: Image acquisition module, used to obtain the original image of factory monitoring and determine the brightness value of each pixel in the original image; a pixel sequence acquisition module, configured to determine a brightness evaluation index for each pixel based on the brightness value of each pixel using a preset formula, and to determine a pixel sequence based on the brightness evaluation index of each pixel, wherein the pixels in the pixel sequence are arranged in descending order according to the brightness evaluation index; a region division module, configured to perform multiple region divisions starting from the first pixel in the pixel sequence and sequentially using each pixel in the pixel sequence as a reference point according to a preset strategy to obtain a plurality of regions, wherein after each region division, the pixel sequence is updated to eliminate pixels falling into the divided region, and region division is performed again based on the updated sequence; The image processing enhancement module is used to calculate the atmospheric light value of each area respectively, and process the original image based on the atmospheric light value of each area and use the dark channel prior algorithm to obtain an enhanced image; The pixel sequence acquisition module is further configured to determine the brightness distribution of each pixel based on the brightness value of each pixel and the difference between the brightness values ​​of its adjacent pixels; The ratio of the brightness value of each pixel to the brightness distribution degree is used as the brightness evaluation index of each pixel; The process of single area division in the area division module includes: Select a reference point; Calculating a brightness difference between the reference point and adjacent pixels to obtain a first growth point, wherein the first growth point is a pixel point having a maximum brightness difference with the reference point among the adjacent growth points, and the reference point is a parent node of the first growth point; Taking the first growing point as a child growing point, calculating the brightness difference of each pixel adjacent to the first growing point and excluding the parent node to obtain a second growing point, wherein the second growing point is the pixel with the largest brightness difference with the child growing point among the adjacent growing points excluding the parent node; The growth process of the sub-growing points is repeatedly performed until, among the obtained growing points, the number of growing points having brightness values ​​greater than a preset brightness average is the same as the number of growing points having brightness values ​​less than the brightness average.

7. A monitoring data processing device based on industrial Internet of Things, characterized in that: The monitoring data processing equipment based on the industrial Internet of Things includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the monitoring data processing method based on the industrial Internet of Things as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the monitoring data processing method based on the industrial Internet of Things according to any one of claims 1 to 5 is implemented.

Images