CN116580024B - Coke quality detection method based on image processing - Google Patents
Coke quality detection method based on image processing Download PDFInfo
- Publication number
- CN116580024B CN116580024B CN202310848578.1A CN202310848578A CN116580024B CN 116580024 B CN116580024 B CN 116580024B CN 202310848578 A CN202310848578 A CN 202310848578A CN 116580024 B CN116580024 B CN 116580024B
- Authority
- CN
- China
- Prior art keywords
- coke
- pixel point
- image
- calculating
- window
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000000571 coke Substances 0.000 title claims abstract description 144
- 238000001514 detection method Methods 0.000 title claims abstract description 23
- 238000001914 filtration Methods 0.000 claims abstract description 36
- 230000003044 adaptive effect Effects 0.000 claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 14
- 239000011148 porous material Substances 0.000 claims description 43
- 238000004364 calculation method Methods 0.000 claims description 13
- 239000011159 matrix material Substances 0.000 claims description 13
- 229910052704 radon Inorganic materials 0.000 claims description 5
- SYUHGPGVQRZVTB-UHFFFAOYSA-N radon atom Chemical compound [Rn] SYUHGPGVQRZVTB-UHFFFAOYSA-N 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000003708 edge detection Methods 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 claims description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 abstract description 2
- 229910052799 carbon Inorganic materials 0.000 abstract description 2
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000003723 Smelting Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 239000000945 filler Substances 0.000 description 3
- 229910003481 amorphous carbon Inorganic materials 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000011230 binding agent Substances 0.000 description 1
- 239000011335 coal coke Substances 0.000 description 1
- 238000003703 image analysis method Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 239000011347 resin Substances 0.000 description 1
- 229920005989 resin Polymers 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Coke Industry (AREA)
- Image Processing (AREA)
Abstract
The application relates to the field of image processing, and provides a coke quality detection method based on image processing, which comprises the following steps: collecting an image of a coke air hole structure to obtain an image to be detected; calculating the interference degree of dark spots of each pixel point in the image to be detected; calculating the size of an adaptive filter window of each pixel point based on the dark spot interference degree of each pixel point in the image to be detected; and filtering each pixel point based on the size of the adaptive filtering window of each pixel point, and detecting the quality based on the filtered image. The method adaptively calculates the size of the filtering window of each pixel point, so that the filtering accuracy is high when the filtering is performed, and the accuracy of the image-based focusing carbon quality detection index is improved.
Description
Technical Field
The application relates to the field of image processing, in particular to a coke quality detection method based on image processing.
Background
In recent years, with the continuous development of the steel industry, the blast furnace smelting technology is continuously improved, and coke is used as an important raw material for blast furnace smelting, so that the production quality plays an important role in the physicochemical change of the smelting process. Therefore, the requirements on the quality of the coke are continuously increased, and the development of the technology for detecting the quality of the coke is also promoted.
At present, the detection indexes of coke quality mainly comprise detection in aspects of crack degree, porosity and the like, an image analysis method can be adopted for detecting the air holes of the coal coke, the air holes of the coke are detected through collected images, the collected images are generally gray images of the structure of the air holes of the coke under a microscope, the collected images in the mode are poor in quality and have more interference, the accuracy of image processing is low by adopting a traditional Gaussian filtering algorithm, the accuracy of segmentation of a detection object (air holes) is also directly influenced, and therefore the accuracy of the calculated coke quality detection indexes is low.
Disclosure of Invention
The application provides a coke quality detection method based on image processing, which is used for adaptively calculating the size of a filtering window of each pixel point, so that the filtering accuracy is high when filtering is performed, and the accuracy of a coke quality detection index based on images is further improved.
In a first aspect, the present application provides a coke quality detection method based on image processing, including:
collecting an image of a coke air hole structure, and obtaining an image to be detected after graying treatment;
calculating the interference degree of dark spots of each pixel point in the image to be detected;
calculating the size of an adaptive filter window of each pixel point based on the dark spot interference degree of each pixel point in the image to be detected;
and filtering each pixel point based on the size of the adaptive filtering window of each pixel point, and detecting the quality based on the filtered image.
Optionally, calculating the interference degree of the dark spots of each pixel point in the image to be detected includes:
calculating a coke detail loss coefficient of each pixel point;
and calculating the dark spot interference degree of each pixel point in the image to be detected based on the coke detail loss coefficient of each pixel point.
Optionally, calculating a coke detail loss coefficient of each pixel point includes:
constructing a first preset window by taking each pixel point in the image to be detected as a central pixel point;
processing the first preset window by using a radon transform algorithm to obtain a regional coke characteristic sequence of pixel points in the first preset window, and calculating the information entropy of the regional coke characteristic sequence; the information entropy of the regional coke characteristic sequence is the information entropy of the first preset window;
and calculating a coke detail loss coefficient of each pixel point based on the number of the edge pixel points in the first preset window and the information entropy of the first preset window, wherein the number of the edge pixel points in the first preset window is detected by an edge detection algorithm.
Optionally, calculating the dark spot interference degree of each pixel point in the image to be detected based on the coke detail loss coefficient of each pixel point includes:
constructing a second preset window by taking each pixel point in the image to be detected as a central pixel point, forming a dark spot interference degree judgment matrix corresponding to the second preset window based on a coke detail loss coefficient of each pixel point in the second preset window, wherein each row of elements in the dark spot interference degree judgment matrix form a first sequence, and each column of elements form a second sequence;
calculating a first difference value between two adjacent first sequences and calculating a second difference value between two adjacent second sequences; obtaining a coke detail loss coefficient difference set based on the first difference value and the second difference value;
and calculating the dark spot interference degree of each pixel point based on the coke detail loss coefficient difference set, the coke detail loss coefficient of the pixel points in the second preset window and the number of the pixel points in the second preset window.
Optionally, calculating the dark spot interference degree of each pixel point based on the coke detail loss coefficient difference set, the coke detail loss coefficient of the pixel point in the second preset window and the number of the pixel points in the second preset window includes:
calculating a first average value of numerical values in the coke detail loss coefficient difference set;
calculating a second average value of the coke detail loss coefficients of the central pixel point of the second preset window based on the sum of the coke detail loss coefficients of all the pixel points in the second preset window and the number of the pixel points in the second preset window;
and calculating the dark spot interference degree of the central pixel point based on the first average value and the second average value, and further calculating the dark spot interference degree of each pixel point.
Optionally, the following formula is used to calculate the dark spot interference degree of each pixel point:
in the aboveFirst mean value of values in the difference set of coke detail loss coefficients>Indicating +.>Coke detail loss coefficient of individual pixels, < ->Calculating the number of pixels in the second preset window to obtain +.>Dark spot interference degree of central pixel point of second preset window is represented by +.>The normalization processing is performed on the calculation result.
Optionally, calculating the adaptive filter window size of each pixel point based on the dark spot interference degree of each pixel point in the image to be detected includes:
calculating the adaptive filter window size of each pixel point by using the following formula：
In the above-mentioned method, the step of,for the initial filter window length size, +.>For maximum filter window length size, +.>For the dark spot interference degree of the pixel point, +.>To get the calculation result equal to or less than +.>Is the largest even number of (c).
Optionally, filtering each pixel based on the adaptive filter window size of each pixel includes:
calculating the standard deviation of gray values of pixel points in the adaptive filter window;
and performing deblurring processing on each pixel point in the image to be detected based on the adaptive filtering window size and the standard deviation to obtain a Gaussian kernel matrix.
Optionally, performing quality detection based on the filtered image includes:
dividing the filtered image by using a self-adaptive threshold segmentation algorithm to obtain a coke pore structure image segmentation result diagram;
the porosity is calculated based on the filtered image using the following formula:
in the aboveRepresenting the +.>The area of the pore region in each of the coke pores, u represents the number of pores in the coke pore structure image, +.>For the area of the coke pore structure image, calculated +.>The porosity is shown.
The application has the beneficial effects that the coke quality detection method based on image processing, which is different from the prior art, comprises the following steps: collecting an image of a coke air hole structure to obtain an image to be detected; calculating the interference degree of dark spots of each pixel point in the image to be detected; calculating the size of an adaptive filter window of each pixel point based on the dark spot interference degree of each pixel point in the image to be detected; and filtering each pixel point based on the size of the adaptive filtering window of each pixel point, and detecting the quality based on the filtered image. The method adaptively calculates the size of the filtering window of each pixel point, so that the filtering accuracy is high when the filtering is performed, and the accuracy of the image-based focusing carbon quality detection index is improved.
Drawings
FIG. 1 is a schematic flow chart of an embodiment of a coke quality detection method based on image processing according to the present application;
fig. 2 is a flow chart of an embodiment of step S12 in fig. 1.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The application adopts an image processing mode to detect the quality of coke, mainly uses a produced coke finished product to manufacture a coke polished section, collects a coke pore structure gray level image under a microscope, adopts a Gaussian filtering mode to process interference information generated by the collected image, and because the collected image is a microscopic image, more detail information needs to be reserved, the processing precision of the coke pore structure gray level image is lower by a traditional Gaussian filtering algorithm, so that the analysis of the coke pore structure gray level image is inaccurate, the accuracy of coke quality detection based on image processing is reduced, and therefore, the darkness interference degree of pixel points is calculated based on characteristic information of the coke pore structure gray level image surface, and the image is subjected to Gaussian filtering deblurring processing through the darkness interference degree, so that the coke pore structure gray level image with more accurate image detail information is obtained. The present application will be described in detail with reference to the accompanying drawings and examples.
Referring to fig. 1, fig. 1 is a schematic flow chart of an embodiment of a coke quality detection method based on image processing according to the present application, which specifically includes:
step S11: and acquiring an image of the coke pore structure to obtain an image to be detected.
According to the produced coke product, corresponding coke Jiao Guangpian is manufactured, observation focusing is carried out by a 5-time dry objective lens under a microscope, the scanning interval of the object table is 2.5mm transversely and 2.5mm longitudinally, and the scanning area is 2And after focusing is finished, adopting Optmas software to automatically acquire a coke pore structure gray level image every 20s, and acquiring 20 images of each coke sheet.
Step S12: and calculating the interference degree of the dark spots of each pixel point in the image to be detected.
In order to facilitate observation, in the process of manufacturing the focal plate, the filler is filled with a binder for image processing, the filler is generally resin, the surface structure of the filler is greatly different from that of the pore wall (coke), and the influence on pore detection is small.
Because the coke pore structure gray level image is acquired through the coke sheet, the surface of the coke sheet is provided with more pores and amorphous carbon areas, the acquired image can generate Fresnel bright spots (formed by diffraction of light in the pores and the amorphous carbon areas), dark stripes, dark spots and other forms appear in the image, and due to the existence of the influence factors, more detail information is lost when the Gaussian filtering is adopted to process the coke pore structure gray level image, and the filtering effect is poor.
The application calculates the self-adaptive filter window size of each pixel based on the dark spot interference degree of the pixel; and filtering each pixel point based on the size of the adaptive filtering window of each pixel point, and detecting the quality of the filtered image to improve the filtering effect.
Based on this, the step needs to calculate the interference degree of the dark spots of each pixel point in the image to be detected. Referring to fig. 2, step S12 specifically includes:
step S21: and calculating the coke detail loss coefficient of each pixel point.
Specifically, a first preset window is constructed by taking each pixel point in the image to be detected as a central pixel point. In a specific embodiment, each pixel point in the image to be detected is taken as a central pixel point, a window of 9*9 is constructed, and a larger window area is set to reflect the influence of the area where the pixel point is located according to the influence of the dark spots and the like possibly occurring in the image to be detected due to pixels.
And detecting an edge pixel point of the coke pore structure gray level image, namely an image to be detected, by adopting a Canny edge detection algorithm, and marking the edge pixel point according to the obtained coke edge binary image so as to further determine the edge pixel point in the image to be detected.
Processing the first preset window by using a radon transform algorithm to obtain a regional coke characteristic sequence of pixel points in the first preset window, and calculating the regional coke characteristic sequenceInformation entropy; the information entropy of the regional coke characteristic sequence is the information entropy of the first preset window. Specifically, when edge pixel points appear in a first preset window area constructed according to the pixel points, counting the number of the edge pixel points in the first preset window, wherein the more the number of the edge pixel points is, the closer the position of the pixel point is to the edge. Because the boundary of the gray image of the coke pore structure is influenced by dark spots, the area generates more interference factors, and the smaller filter window can lead the pore boundary information of the coke dark spot influence area to be lost. The pixel points close to the boundary carry more boundary information, so that the characteristics of the pixel points in the neighborhood of the first preset window are calculated, and the amount of coke detail information carried by the pixel points is obtained. Inputting a two-dimensional image of the area where the first preset window is located, processing the two-dimensional image by adopting a Radon transformation Radon algorithm, and obtaining one-dimensional data of the projected area of the first preset window, wherein the projection angle is 45 degrees clockwise in the horizontal direction, and the sequence can be defined as an area coke characteristic sequence [ the sequence of pixels ]]The number of one-dimensional data after projection is +.>. The information entropy of the region coke characteristic sequence is calculated as follows: />If the entropy value is larger, the pixel is more disturbed, and the loss of detail information is more serious. The larger the randomness of the sequence is, the larger the influence of factors such as dark spots on the region is, and the more the information entropy of the region coke characteristic sequence is required to be described.
And calculating the coke detail loss coefficient of each pixel point based on the number of the pixel points at the inner edge of the first preset window and the information entropy of the first preset window. Specifically, the coke detail loss coefficientThe calculation method is as follows:
in the aboveGray level image representing coke pore structure, namely first +.in image to be detected>The number of edge pixel points in the first preset window area of each pixel point, +.>Representing calculation of +.>Information entropy obtained in a first preset window area of each pixel point, and calculated +.>Indicate->Coke detail loss coefficient for each pixel. Calculated +.>The larger the pixel point is, the more edge pixel points are contained in the first preset window area, and the calculated +.>The larger the value of (2) is, the closer the first preset window area where the pixel is located is to the edge of the air hole, and the larger the area is affected by interference, the larger the coke detail loss coefficient of the pixel is.
Step S22: and calculating the dark spot interference degree of each pixel point in the image to be detected based on the coke detail loss coefficient of each pixel point.
In one embodiment, a second preset window is constructed by taking each pixel point in the image to be detected as a central pixel point,and forming a dark spot interference degree judgment matrix corresponding to the second preset window based on the coke detail loss coefficient of each pixel point in the second preset window, wherein each row of elements in the dark spot interference degree judgment matrix form a first sequence, and each column of elements form a second sequence. Specifically, due to factors such as dark spots, the discontinuity of boundary pixel points in the image is serious, and the degree of lost information influenced by interference factors can be analyzed according to the coke detail loss coefficient of the pixel points. And setting an interference judging window, namely a second preset window, as a central pixel point by taking each pixel point in the coke pore structure image, namely the image to be detected, wherein the size of the second preset window is 5*5, and when more interference factors appear in the second preset window, the gray value of the pixels in the second preset window is reduced to different degrees due to the influence of the interference factors, so that the influence on the coke structure in the image is larger. Therefore, the dark spot interference degree of the central pixel point of the area where the second preset window is located can be calculated according to the coke detail loss coefficient of the pixel point. The dark spot interference degree judgment matrix can be formed according to the coke detail loss coefficient of the pixel points：
The matrix described aboveMiddle->Represents the coke detail loss coefficient of the 1 st row and 1 st column pixel points in the second preset window,for the size of the second preset window (+)>Taking the empirical value 5), each row of elements of the matrix constitutes a first sequence +.>Second sequence of each column +.>。
Calculating a first difference value between two adjacent first sequences and calculating a second difference value between two adjacent second sequences; and obtaining a coke detail loss coefficient difference set based on the first difference value and the second difference value. Specifically, a first difference value between adjacent first sequences is calculated using the following formula:. Wherein (1)>Representing the calculation of the difference value between the two first sequences, is->Representation matrix->Middle->A first sequence of row elements, +.>Representation matrix->Middle->A first sequence of row elements, +.>Indicate->First sequence of row elements and +.>A first difference value between a first sequence of row elements. Calculating a second difference value between adjacent second sequences using the formula: />. Wherein (1)>Representing the calculation of the difference value between the two second sequences, is->Representation matrix->Middle->A second sequence of column elements, +.>Representation matrix->Middle->A second sequence of column elements, +.>Indicate->Second sequence of column elements and +.>A second difference value between the second sequences of column elements.
It should be noted that the number of the substrates,(Dynamic Time Warping) is a dynamic time warping algorithm, which is a dynamic programming cost-effective for computing similarity of 2 time sequences, especially sequences of different lengthsA method of manufacturing the same.
The difference values of two adjacent rows can be calculated from top to bottom in the mode, and the sequence formed by each column of elements sequentially calculates the difference values of the adjacent sequences from left to right. Based on the calculated matrixThe first difference value of the middle row and the second difference value of the column form a coke detail loss coefficient difference set +.>Can obtain a coke detail loss coefficient difference setK represents the number of calculated difference values.
Specifically, as the dark spot interference exists in the coke region more, the image detail loss characteristic value is obtained through the dark spot, namely the change of the coke detail loss coefficient of the pixel point is calculated, and the interference degree generated in the region can be judged. And calculating the dark spot interference degree of each pixel point based on the coke detail loss coefficient difference set, the coke detail loss coefficient of the pixel points in the second preset window and the number of the pixel points in the second preset window.
In one embodiment, a first average of values in the set of coke detail loss coefficient differences is calculatedA second preset window is based on the sum of coke detail loss coefficients of all pixel points in the second preset window>And the number of pixels in the second preset window +.>Calculating a second average value of coke detail loss coefficients of the central pixel point of a second preset window +.>Based on the firstMean->And said second mean->Calculating to obtain dark spot interference degree of the central pixel point>Further obtain the dark spot interference degree of each pixel point>。
In one embodiment, the following formula is used to calculate the dark spot interference level of each pixel:
in the aboveFirst mean value of values in the difference set of coke detail loss coefficients>Indicating +.>Coke detail loss coefficient of individual pixels, < ->Calculating the number of pixels in the second preset window to obtain +.>Dark spot interference degree of central pixel point of second preset window is represented by +.>The normalization processing is performed on the calculation result.
It can be appreciated that in the defined interference determinationThe window is the average value of coke detail loss coefficients of the pixel points in a second preset windowThe larger the dark spot influence of the second preset window area is, the larger the dark spot influence is, the larger the change of the transverse and longitudinal coke detail loss coefficients is, the larger the dark spot influence of the dark spot influence is, the larger the calculated dark spot interference degree is, the dark spot interference degree of a central pixel point in an interference judging window can be obtained, and the interference degree of the coke air hole boundary information of the area where the pixel point is located is indicated.
Step S13: and calculating the self-adaptive filter window size of each pixel point based on the dark spot interference degree of each pixel point in the image to be detected.
Specifically, the size of the adaptive filter window of each pixel point can be calculated according to the interference degree of the dark spots of the pixel points in sequence, wherein the size of the initial filter window is as follows（/>The value is 3, the minimum filter window size), the image blurring degree is increased due to the fact that the selected window is not too large, and therefore the maximum filter window is as follows: />。
Calculating the adaptive filter window size of each pixel point by using the following formula：
In the above-mentioned method, the step of,for the initial filter window length size, +.>For maximum filter window length size, +.>For the dark spot interference degree of the pixel point, +.>To get the calculation result equal to or less than +.>The maximum even number of each pixel point is that the size of a filtering window when each pixel point is calculated is finally obtained according to the size of the coke surface influenced by dark spots>In the present application->The value of (2) is +.>。
Step S14: and filtering each pixel point based on the size of the adaptive filtering window of each pixel point, and detecting the quality based on the filtered image.
Specifically, according to the size of the adaptive filter windowAnd calculating the standard deviation of the gray value of the pixel point in the filter window, and finally obtaining a Gaussian kernel matrix according to the size and the standard deviation of the filter window to perform deblurring calculation, wherein the specific calculation process can refer to the existing Gaussian filter algorithm, and details are not repeated here.
And further, segmenting the filtered image by using an adaptive threshold segmentation algorithm to obtain a coke pore structure image segmentation result graph. In one embodiment, the partitions of the adaptive threshold partitioning algorithmThe domain size is 5*5 and the constant C takes 6. The pixel point set which is input as the air hole boundary can obtain a group of invariant moment (zero-order moment, first-order moment and sixth-order moment) of the area by adopting the Hu invariant moment algorithm, wherein the zero-order moment represents the area of the area, and the air hole area set [ in the gray level image of the coke air hole structure can be obtained by selecting the zero-order moment of each area]The number of pores in the gray level image of the coke pore structure is。
Specifically, the porosity is calculated based on the filtered image using the following formula:
in the aboveRepresenting the +.>The area of the pore region in each of the coke pores, u represents the number of pores in the coke pore structure image, +.>For the area of the coke pore structure image, calculated +.>The specific index of qualified coke porosity can be set according to the actual industrial requirements.
According to the coke quality detection method based on image processing, dark spot interference degree is constructed through dark spot interference information in a gray level image of a coke pore structure and structure detail loss information of a region where a pixel point is located and is used as a reference coefficient for Gaussian filter window adjustment. The size of the Gaussian filter window is calculated based on the dark spot interference degree, and the size of the Gaussian filter window can be adaptively adjusted according to the dark spot influence of different areas of the gray level image of the coke pore structure, so that the condition that the deblurring effect is low due to the influence of the dark spot is avoided, and the information of the pore boundary structure in the coke pore structure is lost, so that the calculation error of the porosity is increased. Therefore, gaussian filtering is carried out on the gray level image of the coke pore structure, the corresponding filtering window size is obtained according to the difference of the regional characteristics, the deblurring precision of Gaussian filtering is improved, and the accuracy of detecting the quality of the coke based on image processing is further improved.
The foregoing is only the embodiments of the present application, and therefore, the patent scope of the application is not limited thereto, and all equivalent structures or equivalent processes using the descriptions of the present application and the accompanying drawings, or direct or indirect application in other related technical fields, are included in the scope of the application.
Claims (4)
1. The coke quality detection method based on image processing is characterized by comprising the following steps:
collecting an image of a coke air hole structure, and obtaining an image to be detected after graying treatment;
calculating the interference degree of dark spots of each pixel point in the image to be detected;
calculating the size of an adaptive filter window of each pixel point based on the dark spot interference degree of each pixel point in the image to be detected;
filtering each pixel point based on the size of the adaptive filtering window of each pixel point, and detecting the quality based on the filtered image;
the calculating of the dark spot interference degree of each pixel point in the image to be detected comprises the following steps:
calculating a coke detail loss coefficient of each pixel point;
calculating the interference degree of dark spots of each pixel point in the image to be detected based on the coke detail loss coefficient of each pixel point;
calculating a coke detail loss coefficient of each pixel point comprises:
constructing a first preset window by taking each pixel point in the image to be detected as a central pixel point;
processing the first preset window by using a radon transform algorithm to obtain a regional coke characteristic sequence of pixel points in the first preset window, and calculating the information entropy of the regional coke characteristic sequence; the information entropy of the regional coke characteristic sequence is the information entropy of the first preset window;
calculating a coke detail loss coefficient of each pixel point based on the number of the edge pixel points in the first preset window and the information entropy of the first preset window, wherein the number of the edge pixel points in the first preset window is detected by an edge detection algorithm;
calculating the dark spot interference degree of each pixel point in the image to be detected based on the coke detail loss coefficient of each pixel point, wherein the dark spot interference degree comprises the following steps:
constructing a second preset window by taking each pixel point in the image to be detected as a central pixel point, forming a dark spot interference degree judgment matrix corresponding to the second preset window based on a coke detail loss coefficient of each pixel point in the second preset window, wherein each row of elements in the dark spot interference degree judgment matrix form a first sequence, and each column of elements form a second sequence;
calculating a first difference value between two adjacent first sequences and calculating a second difference value between two adjacent second sequences; obtaining a coke detail loss coefficient difference set based on the first difference value and the second difference value;
calculating the dark spot interference degree of each pixel point based on the coke detail loss coefficient difference set, the coke detail loss coefficient of the pixel points in the second preset window and the number of the pixel points in the second preset window;
calculating the dark spot interference degree of each pixel point based on the coke detail loss coefficient difference set, the coke detail loss coefficient of the pixel points in the second preset window and the number of the pixel points in the second preset window, wherein the dark spot interference degree comprises the following steps:
calculating a first average value of numerical values in the coke detail loss coefficient difference set;
calculating a second average value of the coke detail loss coefficients of the central pixel point of the second preset window based on the sum of the coke detail loss coefficients of all the pixel points in the second preset window and the number of the pixel points in the second preset window;
calculating the dark spot interference degree of the central pixel point based on the first average value and the second average value, and further calculating the dark spot interference degree of each pixel point;
the dark spot interference degree of each pixel point is calculated by using the following formula:
in the aboveFirst mean value of values in the difference set of coke detail loss coefficients>Indicating +.>Coke detail loss coefficient of individual pixels, < ->Calculating the number of pixels in the second preset window to obtain +.>Dark spot interference degree of central pixel point of second preset window is represented by +.>The normalization processing is performed on the calculation result.
2. The method for detecting coke quality based on image processing according to claim 1, wherein calculating the adaptive filter window size of each pixel based on the darkness interference level of each pixel in the image to be detected comprises:
calculating the adaptive filter window size of each pixel point by using the following formula：
In the above-mentioned method, the step of,for the initial filter window length size, +.>For maximum filter window length size, +.>For the dark spot interference degree of the pixel point, +.>To get the calculation result equal to or less than +.>Is the largest even number of (c).
3. The method for detecting coke quality based on image processing according to claim 1, wherein the filtering process is performed for each pixel based on the adaptive filter window size of each pixel, comprising:
calculating the standard deviation of gray values of pixel points in the adaptive filter window;
and performing deblurring processing on each pixel point in the image to be detected based on the adaptive filtering window size and the standard deviation to obtain a Gaussian kernel matrix.
4. The coke quality detection method based on image processing according to claim 1, wherein the quality detection based on the filtered image comprises:
dividing the filtered image by using a self-adaptive threshold segmentation algorithm to obtain a coke pore structure image segmentation result diagram;
the porosity is calculated based on the filtered image using the following formula:
in the aboveRepresenting the +.>The area of the pore region in each of the coke pores, u represents the number of pores in the coke pore structure image, +.>For the area of the coke pore structure image, calculated +.>The porosity is shown.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310848578.1A CN116580024B (en) | 2023-07-12 | 2023-07-12 | Coke quality detection method based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310848578.1A CN116580024B (en) | 2023-07-12 | 2023-07-12 | Coke quality detection method based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116580024A CN116580024A (en) | 2023-08-11 |
| CN116580024B true CN116580024B (en) | 2023-09-15 |
Family
ID=87534454
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310848578.1A Active CN116580024B (en) | 2023-07-12 | 2023-07-12 | Coke quality detection method based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116580024B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116735612B (en) * | 2023-08-15 | 2023-11-07 | 山东精亿机械制造有限公司 | Welding defect detection method for precise electronic components |
| CN117557569B (en) * | 2024-01-12 | 2024-04-02 | 吉林交通职业技术学院 | Road pavement construction quality detection method and system |
| CN117952958B (en) * | 2024-03-21 | 2024-06-07 | 陕西中环机械有限责任公司 | Drum filter screen intelligent health detection method based on machine learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115406808A (en) * | 2022-08-29 | 2022-11-29 | 武汉顶力康自动化有限公司 | Coke particle size diameter and proportion analysis detector based on vision |
| CN115496693A (en) * | 2022-11-17 | 2022-12-20 | 南通鼎勇机械有限公司 | Sintering flame image smog removing method based on dark channel algorithm |
| CN115631191A (en) * | 2022-11-28 | 2023-01-20 | 南京北新智能科技有限公司 | Coal blockage detection algorithm based on gray level features and edge detection |
| CN116188462A (en) * | 2023-04-24 | 2023-05-30 | 深圳市翠绿贵金属材料科技有限公司 | Noble metal quality detection method and system based on visual identification |
| CN116309570A (en) * | 2023-05-18 | 2023-06-23 | 山东亮马新材料科技有限公司 | Titanium alloy bar quality detection method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11270429B2 (en) * | 2019-06-13 | 2022-03-08 | Tata Consultancy Services Limited | Method and system for impurity detection using multi-modal imaging |
-
2023
- 2023-07-12 CN CN202310848578.1A patent/CN116580024B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115406808A (en) * | 2022-08-29 | 2022-11-29 | 武汉顶力康自动化有限公司 | Coke particle size diameter and proportion analysis detector based on vision |
| CN115496693A (en) * | 2022-11-17 | 2022-12-20 | 南通鼎勇机械有限公司 | Sintering flame image smog removing method based on dark channel algorithm |
| CN115631191A (en) * | 2022-11-28 | 2023-01-20 | 南京北新智能科技有限公司 | Coal blockage detection algorithm based on gray level features and edge detection |
| CN116188462A (en) * | 2023-04-24 | 2023-05-30 | 深圳市翠绿贵金属材料科技有限公司 | Noble metal quality detection method and system based on visual identification |
| CN116309570A (en) * | 2023-05-18 | 2023-06-23 | 山东亮马新材料科技有限公司 | Titanium alloy bar quality detection method and system |
Non-Patent Citations (2)
| Title |
|---|
| Application of Improved Canny Edge Detection Algorithm in Coal-rock Interface Recognition;Lihong Dong 等;《 Applied Mechanics and Materials》;第220-223卷;第1279-1283页 * |
| 多特征融合的锅炉燃烧状态监测方法研究;何雨晨;《中国博士学位论文全文数据库 工程科技Ⅱ辑》(第03期);第C042-76页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116580024A (en) | 2023-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116580024B (en) | Coke quality detection method based on image processing | |
| CN115829883B (en) | Surface image denoising method for special-shaped metal structural member | |
| CN116385448B (en) | Alumina ceramic surface machining defect detection method based on machine vision | |
| JP6099479B2 (en) | Crack detection method | |
| CN109489566B (en) | Lithium battery diaphragm material slitting width detection method, detection system and device | |
| CN116664559B (en) | Machine vision-based memory bank damage rapid detection method | |
| CN114820773B (en) | Silo transport vehicle carriage position detection method based on computer vision | |
| JP5852919B2 (en) | Crack detection method | |
| CN113109368A (en) | Glass crack detection method, device, equipment and medium | |
| CN112308854B (en) | Automatic detection method and system for chip surface flaws and electronic equipment | |
| CN115861291A (en) | Patch circuit board production defect detection method based on machine vision | |
| CN115830033A (en) | Automobile hub surface defect detection method based on machine vision | |
| CN116703251B (en) | Rubber ring production quality detection method based on artificial intelligence | |
| CN117315289B (en) | Aeroengine blade contour edge detection method based on image processing | |
| CN115294099A (en) | Method and system for detecting hairline defect in steel plate rolling process | |
| CN117893527B (en) | New energy storage box surface defect detection method | |
| CN116309577A (en) | Intelligent detection method and system for high-strength conveyor belt materials | |
| CN116823827B (en) | Ore crushing effect evaluation method based on image processing | |
| CN117649451A (en) | Visual positioning method for positioning edge finder wafer | |
| CN116416268A (en) | Method and device for detecting edge position of lithium battery pole piece based on recursion dichotomy | |
| CN116468689A (en) | Flaw identification method based on gray scale characteristics | |
| CN110827272B (en) | Tire X-ray image defect detection method based on image processing | |
| CN116703930B (en) | Automobile rearview mirror mold forming detection method | |
| CN116740065B (en) | Quick tracing method and system for defective products of artificial board based on big data | |
| CN117314899B (en) | Carbon fiber plate quality detection method based on image characteristics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |