An evaluation system for dust pollution
Through image processing technology, dust pollution levels are generated, which solves the problem of large errors in single variable evaluation in the existing technology, and achieves more accurate dust pollution monitoring and management.
Patent Information
- Application Number
- CN202111549340.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-12-17
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2041-12-17
AI Technical Summary
In the prior art, there are large errors or distortions in the evaluation of dust pollution conditions using univariates, especially in complex working conditions, which has a great impact on sensor arrangement and environmental conditions, resulting in inaccurate determination of dust concentration.
Dust images are collected through the image acquisition device, a three-dimensional image digital matrix is generated, gradient calculation, image sharpening, grayscale conversion and binarization processing are performed, fractal dimensions and grayscale mean are obtained, and dust pollution levels are generated by combining the dust pollution index model. Implement multivariate evaluation.
It improves the accuracy and reliability of dust pollution evaluation, reduces errors, and can more effectively monitor and manage dust pollution conditions.
Smart Images

Figure CN114240884B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of image processing, and in particular to an evaluation system for dust pollution. Background Art
[0002] Dust pollution has always been a common problem in industrial production, especially in the process of mineral resource development. In the mining process of coal mines and some metal mines, some mining processes: drilling, blasting, cutting, concrete spraying, transportation, ore lifting, loading, belt transmission and other working areas are all high sources of dust. Most dust has unfavorable properties such as easy diffusion and flammability. The hazards brought by dust to production and life are mainly divided into:
[0003] (1) Workers who work in a high-concentration dust environment for a long time are prone to occupational diseases: keratitis, rhinitis, pharyngitis, bronchitis, silicosis, chronic obstructive pulmonary disease, pneumoconiosis, etc.;
[0004] (2) Coal mines have a high tendency to spontaneous combustion. When the air circulation is sufficient and the concentration conditions meet the dust explosion limit, an explosion accident is very likely to occur under the action of the ignition source. Dust explosions are very destructive. For coal dust explosions, the shock wave generated by dust explosions is as high as 2340m / s, and the explosion temperature can reach 2300-2500℃;
[0005] (3) Under the influence of dust, workers’ workplaces are seriously polluted, visibility is low, and production efficiency is reduced. At the same time, dust has a wear effect on production equipment. Under long-term wear, the accuracy and safety requirements of the equipment cannot be fully maintained, which ultimately affects the economic benefits of the enterprise.
[0006] At present, the dust concentration evaluation method is generally used to evaluate dust pollution at home and abroad, that is, to measure the dust concentration of the working surface and analyze and judge the measurement results. The methods for measuring dust concentration generally include handheld sampling equipment, rapid concentration detectors and dust online detectors. Handheld dust detection equipment mainly includes filter membrane weighing; rapid detectors mainly include photoelectric conversion method, β-ray attenuation method, piezoelectric crystal frequency change method, etc.; dust concentration online detectors mainly use the acoustic and optical effect to achieve dust online concentration detection as a whole by setting up multiple sensors. In the application of digital image technology in dust concentration detection, the dust concentration fitting curve is generally obtained by studying the relationship between dust concentration and digital image related parameters, such as grayscale, morphological characteristics, etc., but in field applications, the maturity is relatively low.
[0007] Existing dust pollution monitoring and evaluation technologies include:
[0008] CN103089291B discloses an image-based coal mine dust pollution evaluation and spray dust removal linkage system, including an image evaluation part and a state control part. The image evaluation part monitors the entire dust-generating space through an image sensor arranged at the dust-generating point, and obtains an evaluation result by comparing with a set background image; the state control part controls the spray device arranged at the dust-generating point according to the result of the image evaluation, and starts the spray dust reduction system when the evaluation result shows that the current dust concentration exceeds the reference value; the spray dust reduction system is turned off when the evaluation result shows that the current concentration is lower than the reference value; the state control part mainly controls the spray dust reduction system installed at each dust-generating point through a control box.
[0009] CN206138918U discloses an automatic integrated device for dust monitoring and control, including a dust sensor, a dust removal device, and a control device. The dust sensor is used to detect dust data in a target area; the dust removal device is used to reduce the dust content in the target area; the control device is connected to the dust sensor and the dust removal device in communication, and controls the opening and closing of the dust removal device according to the dust data. The control device simultaneously exchanges data with the dust sensor and the dust removal device, thereby realizing the synchronous management and control of the dust monitoring system and the dust control system.
[0010] The dust pollution monitoring and evaluation methods in the above-mentioned prior art can, to a certain extent, measure the concentration of dust sources and evaluate pollution. Traditional dust concentration measurement methods, such as heavy filter membranes, are complicated to operate and take a long time to measure; rapid concentration measuring instruments such as the β-ray attenuation method have high requirements for experimental equipment and have limited measurement ranges; online dust detectors have received more attention in recent years. They use a series of sensors and the Internet of Things technology to monitor and measure dust concentration online. However, under complex working conditions, the number and location of sensor arrangements may greatly affect normal production, and the maintenance cost of the equipment is relatively high. The relationship between image parameters and dust concentration is studied through digital imaging technology, and then the dust pollution status is evaluated. This is easily affected by environmental conditions (light, wind speed, etc.), resulting in large errors.
[0011] In view of this, this application is filed. Summary of the invention
[0012] The invention discloses a dust pollution evaluation system, aiming to solve the problem of large error or distortion in the prior art when using a single variable to evaluate the dust pollution status.
[0013] An embodiment of the present invention provides a dust pollution evaluation system, comprising: a transparent box, a dust removal device arranged above the transparent box, a fan device arranged at a first side of the transparent box, a movable door arranged at a second side of the transparent box, an image acquisition device arranged outside the transparent box, and a controller electrically connected to the image acquisition device;
[0014] The controller is configured to implement the following steps by executing a computer program stored therein:
[0015] Receiving a plurality of dust images collected by the image acquisition device at preset time intervals, and processing the plurality of dust images to generate a three-dimensional image digital matrix;
[0016] Performing gradient calculation on the three-dimensional image digital matrix with pixels as gradient step length to generate a gradient image;
[0017] Performing image sharpening, grayscale conversion, and binarization processing on the gradient image to generate a dust denoising image;
[0018] The dust denoising image is operated to obtain a fractal dimension value and a grayscale mean value, and a dust pollution index model is called to operate the fractal dimension and the grayscale mean value to generate a dust pollution level.
[0019] Preferably, the step of performing gradient calculation on the three-dimensional image digital matrix with pixels as gradient step to generate a gradient image is specifically as follows:
[0020] Call the gradient calculation model to calculate the pixels {I i,j (x,y),I i+Δi,j (x,y),I i,j+Δj (x, y)} to perform gradient calculation to generate the gradient image; wherein the expression of the gradient calculation model is:
[0021]
[0022] Among them, Grad ij represents the image pixel value after gradient calculation, Δj and Δi represent the step sizes selected in the vertical and horizontal directions respectively.
[0023] Preferably, the step of performing image sharpening, grayscale conversion, and binarization processing on the gradient image to generate the dust denoising image is specifically as follows:
[0024] Performing image sharpening processing on the gradient image to generate a sharpened image with compensated image contour features;
[0025] Calling a grayscale average model to perform grayscale conversion on the sharpened image to generate a grayscale image;
[0026] Performing binarization calculation on the grayscale image to generate a binary image;
[0027] The noise points in the binary image are affected to generate a dust denoising image.
[0028] Preferably, the dust denoising image is operated to obtain a fractal dimension value and a grayscale mean value, and a dust pollution index model is called to operate the fractal dimension and the grayscale mean value to generate a dust pollution level.
[0029] Calling a two-dimensional image box-dimension model to calculate the dust denoising image to obtain a fractal dimension;
[0030] Calling a grayscale mean model to operate the dust denoising image to obtain a grayscale mean;
[0031] Performing linear fitting on the grayscale mean to generate a first fitting square value, and performing exponential fitting on the fractal dimension with e as the base to generate a second fitting square value;
[0032] The dust pollution index model is called to calculate the first fitting square value, the second fitting square value, the grayscale mean, and the fractal dimension to generate a dust pollution level.
[0033] Preferably, the expression of the two-dimensional image box-dimension model is:
[0034] log(N k )=D F log(K)+C;
[0035] Among them, D F Fractal dimension, N k To satisfy the total number of boxes that traverse the entire image, K is the reciprocal 1 / k of the side length k of the square box, and the side lengths of the fractal boxes are 1, 2, ... 2 n .
[0036] Preferably, the grayscale mean model is expressed as:
[0037]
[0038] Among them, M and N are the height and width of the input size of the image respectively.
[0039] Preferably, the dust pollution index model is expressed as:
[0040]
[0041] Among them, R G is the first fitting square value, R Fis the second fitting square value, h is the fitting coefficient, D F Fractal dimension, G A Grayscale mean.
[0042] Preferably, it also includes lighting devices arranged on the left and right sides of the image acquisition device.
[0043] Preferably, the dust removal device comprises a plurality of funnels;
[0044] Wherein, a plurality of feeding holes are arranged on the top of the transparent box body, and the neck of the funnel is adapted to the feeding holes.
[0045] Preferably, the fan device comprises: an axial flow motor, a tachometer, and fan blades;
[0046] Wherein, the tachometer is electrically connected to the axial flow motor, the fan blades are arranged on the output shaft of the axial flow motor, and the fan blades are arranged inside the transparent box.
[0047] A dust pollution evaluation system provided by the present invention firstly processes a plurality of dust images collected by an image acquisition device at preset time intervals to generate a three-dimensional image digital matrix, then performs gradient calculation on the three-dimensional image digital matrix with pixels as gradient step to generate a gradient image; then performs image sharpening, grayscale conversion, and binarization on the gradient image to generate a dust denoising image, and finally performs calculation on the dust denoising image to obtain a fractal dimension and a grayscale mean, and calls a dust pollution index model to calculate the fractal dimension and the grayscale mean to generate a dust pollution grade, thereby solving the problem of large errors or distortions in the prior art when using a single variable to evaluate the dust pollution status. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] Figure 1 It is a structural schematic diagram of a dust pollution evaluation system provided by the present invention;
[0049] Figure 2 This is a schematic diagram of a dust pollution evaluation process provided by the present invention;
[0050] Figure 3 It is a schematic diagram of an application example process provided by the present invention;
[0051] Figure 4 is a schematic diagram of an example image processing diagram provided by the present invention;
[0052] Figure 5 It is a schematic diagram of the mean value and fractal dimension results of an example dust image provided by the present invention;
[0053] Figure 6 This is a schematic diagram of the dust pollution area division provided by the present invention. DETAILED DESCRIPTION
[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0055] In order to better understand the technical solution of the present invention, the embodiments of the present invention are described in detail below with reference to the accompanying drawings.
[0056] It should be clear that the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0057] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.
[0058] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.
[0059] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.
[0060] The "first\second" mentioned in the embodiments is only to distinguish similar objects, and does not represent a specific order for the objects. It is understandable that the "first\second" can be interchanged with the specific order or sequence where permitted. It should be understood that the objects distinguished by "first\second" can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than those illustrated or described herein.
[0061] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.
[0062] The invention discloses a dust pollution evaluation system, aiming to solve the problem of large error or distortion in the prior art when using a single variable to evaluate the dust pollution status.
[0063] See also Figure 1 The embodiment of the present invention provides a dust pollution evaluation system, comprising: a transparent box 1, a dust removal device arranged above the transparent box 1, a fan device arranged at a first side of the transparent box 1, a movable door 6 arranged at a second side of the transparent box 1, an image acquisition device 2 arranged outside the transparent box 1, and a controller 8 electrically connected to the image acquisition device 2;
[0064] It should be noted that the transparent box 1 can be made of organic glass to facilitate the image acquisition device 2 to clearly acquire images, and the movable door 6 can be moved in the horizontal direction to change the inlet and outlet air flow;
[0065] In this embodiment, the dust removal device includes a plurality of funnels 3; wherein, the top of the transparent box 1 is provided with a plurality of discharge holes 5, and the neck of the funnel 3 is adapted to the discharge hole 5. wherein, each of the funnels 3 can be equipped with dust 4 of different masses, wherein the funnel 3 can be made of stainless steel, and under the guarantee of diameter and material, the adhesion effect of the falling dust and the wall of the device can be ignored, and its diameter can be 25 cm to ensure the early dispersion of the dust particles, and at the same time rely on gravity to generate a certain initial velocity, so that the dust sample can achieve a more sufficient diffusion effect under the action of wind pressure when it exits the hole.
[0066] In this embodiment, an illumination device 7 may be further included, which is arranged on the left and right sides of the image acquisition device 2. The illumination device 7 may adjust the intensity and color temperature of the spotlight; adjust the horizontal position and the height of the lampshade according to the placement position of the funnel 33; and make the illumination device 7 focus the light on the long neck of the funnel 3.
[0067] In this embodiment, the fan device includes: an axial flow motor 9, a speed controller 10, and fan blades 11;
[0068] The tachometer 10 is electrically connected to the axial flow motor 9 , the fan blades 11 are arranged on the output shaft of the axial flow motor 9 , and the fan blades 11 are arranged inside the transparent box 1 .
[0069] The axial flow motor 9 is equipped with a tachometer 10. By adjusting the tachometer 10, a speed regulation range of 0 to 1500 r / min is achieved, providing different levels of wind pressure to act on the dust sample. In this embodiment, reverse ventilation can be achieved by adjusting the speed regulator circuit; the fan blades 11 can be a five-blade fan, which is fixedly connected to the axial flow motor 9.
[0070] Specifically, the specific experimental steps in this embodiment may include:
[0071] Step 1. Dry the dust sample for 24 hours at 50°C. Weigh the dust sample with an electronic balance and divide the dust sample into 30 groups (10mg to 300mg) at 10mg weight intervals. Put them in weighing paper and label them. Fully drying the dust reduces the adverse effects of dust shielding or adhesion on subsequent dust diffusion and image acquisition;
[0072] Step 2, turn on the lighting device 7 to adjust the spotlight intensity and color temperature; adjust the horizontal position of the spotlight and the height of the lampshade according to the position of the funnel 3; turn on the image acquisition device 2, adjust the image acquisition angle so that it is directly facing the dust falling position, and can obtain the dust diffusion trajectory to the greatest extent; observe the image range quality, and fine-tune the image acquisition device 2 parameter settings as needed;
[0073] Step 3: Turn on the power of the axial flow motor 9, adjust the speed controller 10 to reach a stable speed to ensure the stability of the air flow; push the movable doors 6 on the left and right sides to reach the air inlet and outlet areas required by the test plan; place the funnel 3 at the determined feeding hole 5. After completing the above steps, conduct a preliminary experiment, let a small amount of dust fall into the transparent box through the funnel 3, observe the dust image quality status collected by the camera under the above conditions, and make the final equipment fine-tuning;
[0074] Step 4: The grouped dust is dropped into the box through the funnel 3 in sequence. The image acquisition device 2 performs real-time image acquisition and transmits the acquired image to the controller 8. The controller 8 processes and calculates the dust image through a program to obtain an image result of the target attribute and visualizes the experimental data on the display.
[0075] See also Figure 2 The controller 8 is configured to implement the following steps by executing the computer program stored therein:
[0076] S101, receiving a plurality of dust images collected by the image collection device at preset time intervals, and processing the plurality of dust images to generate a three-dimensional image digital matrix, the dust images such as Figure 3 (a) as shown;
[0077] It should be noted that the dust images P1, P2, ..., Pn are collected by the image acquisition device, and the collection time interval Tm is determined according to the actual test plan. The collected dust image Pn is transmitted to the controller, and the controller reads the dust original image, forms a three-dimensional image digital matrix An, and stores it;
[0078] S102, performing gradient calculation on the three-dimensional image digital matrix with pixels as gradient step length to generate a gradient image;
[0079] Specifically, in this embodiment, the gradient calculation model is called to calculate the pixels {I i,j (x,y),I i+Δi,j (x,y),I i,j+Δj (x, y)} to perform gradient calculation to generate the gradient image; wherein the expression of the gradient calculation model is:
[0080]
[0081] Among them, Grad ij represents the image pixel value after gradient calculation, Δj and Δi represent the step sizes selected in the vertical and horizontal directions respectively.
[0082] It should be noted that the edge features of dust particles can be enhanced through gradient calculation, and the interference of larger dust particles or other impurities on the calculation results can be reduced.
[0083] S103, performing image sharpening, grayscale conversion, and binarization processing on the gradient image to generate a dust denoising image;
[0084] For the gradient image (such as Figure 3 b) performing image sharpening processing to generate a sharpened image with compensated image contour features;
[0085] Call the gray average model Gray ij =[B(x, y)+G(x, y)+R(x, y)] / 3], grayscale conversion is performed on the sharpened image to generate a grayscale image (like Figure 3 c);
[0086] Binarize the grayscale image to generate a binary image (like Figure 3 d);
[0087] The noise points in the binary image are affected to generate a dust denoising image.
[0088] It should be noted that the gradient image is preprocessed, and the preprocessing operations include: image sharpening, grayscale conversion, and binarization. Image sharpening further compensates for image contour features; sharpening the image After grayscale average model: Gray ij = [B(x, y) + G(x, y) + R(x, y)] / 3], grayscale conversion is performed on the dust image; the generated grayscale image Perform binarization calculation to form a binary image Separate the dust particle object and the image background, determine the threshold T according to the brightness characteristics of the collected image, and obtain a binary image of the dust object after the grayscale image pixels are processed by the threshold T. right Perform opening operation denoising to remove the noise in the binary image caused by the image acquisition process or other electronic noise, and generate a dust denoising image
[0089] S104, performing calculations on the dust denoising image to obtain a fractal dimension value and a grayscale mean value, and calling a dust pollution index model to calculate the fractal dimension and the grayscale mean value to generate a dust pollution level, specifically:
[0090] The two-dimensional image box-dimension model is called to operate the dust denoising image to generate a fractal dimension; wherein the expression of the two-dimensional image box-dimension model is: log(N k )=D F log(K)+C;
[0091] Among them, D F Fractal dimension, N k To satisfy the total number of boxes that traverse the entire image, K is the reciprocal 1 / k of the side length k of the square box, and the side lengths of the fractal boxes are 1, 2, ... 2 n .
[0092] The grayscale mean model is called to operate the dust denoising image to generate a grayscale mean, wherein the grayscale mean model is expressed as:
[0093] Among them, M and N are the height and width of the input size of the image respectively;
[0094] The grayscale mean is linearly fitted (G A =a1+b1P) to generate the first fitting square value, and perform an exponential fitting (D F =a2+b2e hP ) generates the second fitting square value; where h is the exponential fitting coefficient, P is the reference dust concentration
[0095] The dust pollution index model is called to calculate the first fitting square value, the second fitting square value, the grayscale mean, and the fractal dimension to generate the dust pollution level. The expression of the dust pollution index model is: Among them, R G is the first fitting square value, R F is the second fitting square value, h is the fitting coefficient, D F Fractal dimension, G A Grayscale mean.
[0096] The following is an example to further illustrate this embodiment:
[0097] The generated dust denoising images were analyzed by computer program for fractal dimension D F , grayscale mean G A Calculate the fractal dimension D F Use the Box-dimension method to count the box dimensions of a two-dimensional image:
[0098] log(N k )=D F log(K)+C
[0099] N k To satisfy the total number of boxes that traverse the entire image, K is the inverse of the side length k of the square box 1 / k; in this method, the side length of the fractal box is 1, 2, ... 2n, and the image visualization result is as follows Figure 4 As shown. The input size of the image is 1920*1080, corresponding to the height and width respectively, then the grayscale mean G A The method is expressed as:
[0100]
[0101] The weighed dust is used as a reference for dust pollution and is divided into four levels according to quality: heavy pollution, high pollution, medium pollution, and slight pollution. The experimental results of each group of dust are processed by image processing, and the corresponding fractal number D is calculated respectively. F and gray mean G A Some of the calculation results are shown in the following table:
[0102] Table 1 Grayscale mean and fractal dimension results in coal powder experiment
[0103]
[0104] The dust samples of 10mg to 300mg are divided into four levels from high to low: M1 (75, 75), M2 (150, 150), M3 (225, 225), M4 (300, 300); the corresponding fractal numbers D are calculated respectively. F and gray mean GA Based on the concentration analysis data, the four level boundary values of concentration determination were replaced by the dust pollution index value Fp, and the regional boundary values of Fp were obtained as Fp1 (64.86, 1.306), Fp2 (149.34, 1.319), Fp3 (234.39, 1.322), and Fp4 (299.71, 1.342). According to the pollution level boundary, the dust pollution level of the underground monitoring point was finally determined as follows: Figure 6 The calculation method of dust pollution index Fp is: according to the calculated G A and D F Values, respectively, linear fitting and exponential fitting with e as the base:
[0105] G A =0.503+0.00020P (4)
[0106] D F =1.66181-0.42671e -0.01755P (5)
[0107] Where k is the exponential fitting coefficient and P is the reference dust concentration. The fitting results are shown in Figure 5 (a) Figure 5 (b) is G A and D F A conventional residual plot of .
[0108] Get the fitting R square value R G : 0.895, R F :0.841. The exponential fitting curve obtained was linearized, that is, the corresponding D F The value is used as the independent variable of the fitting equation. Combined with the above fitting R-square value, the following equation for the dust pollution index Fp is obtained:
[0109]
[0110] Based on the above confirmation equation of dust pollution index Fp, the dust pollution level and corresponding regional division method are as follows:
[0111]
[0112] After completing the dust pollution assessment area division Figure 6 As shown in the figure, based on the obtained evaluation parameters, the actual working face of the mine is monitored in real time, thereby obtaining the dust pollution level of different dust-generating points, or the dust diffusion of different distribution points of the same dust-generating point. The dust pollution level is determined to facilitate the formulation of efficient prevention and control plans or the timely implementation of dust reduction and removal measures.
[0113] A dust pollution evaluation system provided by the present invention firstly processes a plurality of dust images collected by an image acquisition device at preset time intervals to generate a three-dimensional image digital matrix, then performs gradient calculation on the three-dimensional image digital matrix with pixels as gradient step to generate a gradient image; then performs image sharpening, grayscale conversion, and binarization on the gradient image to generate a dust denoising image, and finally performs calculation on the dust denoising image to obtain a fractal dimension value and a grayscale mean, and calls a dust pollution index model to calculate the fractal dimension and the grayscale mean to generate a dust pollution grade, thereby solving the problem of large errors or distortions in the prior art when using a single variable to evaluate the dust pollution status.
[0114] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A dust pollution evaluation system, characterized in that: include: A transparent box, a dust removal device arranged above the transparent box, a fan device arranged at a first side of the transparent box, a movable door arranged at a second side of the transparent box, an image acquisition device arranged outside the transparent box, and a controller electrically connected to the image acquisition device; The controller is configured to implement the following steps by executing a computer program stored therein: Receiving a plurality of dust images collected by the image collection device at preset time intervals, and processing the plurality of dust images to generate a three-dimensional image digital matrix; Performing gradient calculation on the three-dimensional image digital matrix with pixels as gradient step length to generate a gradient image; Performing image sharpening, grayscale conversion, and binarization processing on the gradient image to generate a dust denoising image; The dust denoising image is operated to generate a fractal dimension value and a grayscale mean value, and the dust pollution index model is called to operate the fractal dimension and the grayscale mean value to generate a dust pollution level, wherein specifically: Calling a two-dimensional image box-dimension model to calculate the dust denoising image to obtain a fractal dimension; Calling a grayscale mean model to operate the dust denoising image to obtain a grayscale mean; Performing linear fitting on the grayscale mean to generate a first fitting square value, and performing exponential fitting on the fractal dimension with e as the base to generate a second fitting square value; The dust pollution index model is called to calculate the first fitting square value, the second fitting square value, the grayscale mean, and the fractal dimension to generate the dust pollution level; the expression of the two-dimensional image box dimension model is: ; Among them, D F Fractal dimension, N k To satisfy the total number of boxes that traverse the entire image, K is the reciprocal 1 / k of the side length k of the square box, and the side lengths of the fractal boxes are 1, 2, ... 2 n ; The expression of the grayscale mean model is: ; Wherein, M and N are the height and width of the input size of the image respectively; the expression of the dust pollution index model is: ; Among them, R G is the first fitting square value, R F is the second fitting square value, h is the fitting coefficient, D F Fractal dimension, G A Grayscale mean.
2. A dust pollution evaluation system according to claim 1, characterized in that: The step length of performing gradient calculation on the three-dimensional image digital matrix with pixels as gradient step to generate a gradient image is specifically as follows: Call the gradient calculation model to calculate the pixels in the digital matrix of the three-dimensional image in the horizontal and vertical directions respectively. Perform gradient calculation to generate the gradient image; wherein the expression of the gradient calculation model is: ; Among them, th represents the image pixel value after gradient calculation, Δ and Δi represent the step sizes selected in the vertical and horizontal directions respectively.
3. The dust pollution evaluation system according to claim 1, characterized in that: The steps of performing image sharpening, grayscale conversion, and binarization on the gradient image to generate a dust denoising image are as follows: 5 10 15 20 Performing image sharpening on the gradient image to generate a sharpened image with compensated image contour features; Calling a grayscale average model to perform grayscale conversion on the sharpened image to generate a grayscale image; Performing binarization calculation on the grayscale image to generate a binary image; Eliminating the influence of noise points in the binary image to generate a dust denoising image.
4. The dust pollution evaluation system according to claim 1, characterized in that: It also includes lighting devices arranged on the left and right sides of the image acquisition device.
5. The dust pollution evaluation system according to claim 1, characterized in that: The dust removal device includes a plurality of funnels; wherein, a plurality of discharge holes are arranged on the top of the transparent box, and the neck of the funnel is adapted to the discharge holes.
6. The dust pollution evaluation system according to claim 1, characterized in that: The fan device includes: an axial flow motor, a speed controller, and fan blades; wherein the speed controller is electrically connected to the axial flow motor, the fan blades are arranged on the output shaft of the axial flow motor, and the fan blades are arranged inside the transparent box.
Citation Information
Patent Citations
An image-based coal mine dust pollution assessment and spray dust suppression linkage system
CN103089291B
A automatic integrated device that is used for dust monitoring and administers
CN206138918U
Underground dust monitoring and prevention system and method based on opencv technology
CN112924348A
Method for controlling oven temperature of coke oven and operation of coke guide dust-collector in discharging of coke oven
JP1996337781A
Method for predicting pulmonary disease using fractal dimension values and apparatus for the same
KR101991250B1
