CN116452589A - Intelligent detection method for surface defects of artificial board based on image processing - Google Patents
Intelligent detection method for surface defects of artificial board based on image processing Download PDFInfo
- Publication number
- CN116452589A CN116452589A CN202310712573.6A CN202310712573A CN116452589A CN 116452589 A CN116452589 A CN 116452589A CN 202310712573 A CN202310712573 A CN 202310712573A CN 116452589 A CN116452589 A CN 116452589A
- Authority
- CN
- China
- Prior art keywords
- texture
- radian
- image
- value
- determining
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 47
- 230000007547 defect Effects 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 claims abstract description 16
- 238000000605 extraction Methods 0.000 claims abstract description 11
- 239000002023 wood Substances 0.000 claims description 41
- 239000011295 pitch Substances 0.000 claims description 38
- 230000002159 abnormal effect Effects 0.000 claims description 24
- 230000005856 abnormality Effects 0.000 claims description 22
- 238000010606 normalization Methods 0.000 claims description 13
- 230000009466 transformation Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 description 6
- 238000004519 manufacturing process Methods 0.000 description 5
- 230000001419 dependent effect Effects 0.000 description 4
- 230000000739 chaotic effect Effects 0.000 description 3
- 230000007423 decrease Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000013499 data model Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 241000771208 Buchanania arborescens Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000049 pigment Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000002194 synthesizing effect 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
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/10—Image enhancement or restoration by non-spatial domain filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/54—Extraction of image or video features relating to texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20061—Hough transform
-
- 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
Abstract
The invention relates to the technical field of image processing, in particular to an intelligent detection method for surface defects of an artificial board based on image processing. The method comprises the steps of obtaining a plate gray level image, and obtaining a texture image through texture feature extraction processing; determining the complexity degree of the texture; determining radian values of textures, determining radian coefficients according to the radian values of all textures, and further determining radian similarity indexes; determining the texture interval of adjacent textures according to the distances between corresponding texture pixel points of the adjacent textures in the texture image, and determining the interval similarity index of the texture image according to the texture intervals of all the adjacent textures and preset interval values; determining the comprehensive texture index of the texture image according to the radian similarity index and the interval similarity index; and performing defect detection on the plate gray level image according to the texture complexity and the comprehensive texture index to obtain a detection result. The invention can effectively improve the accuracy of defect detection.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to an intelligent detection method for surface defects of an artificial board based on image processing.
Background
With the increasing shortage of wood resources and environmental protection requirements, artificial boards have grown as extensions and alternatives to wood. The artificial board is various board products obtained by processing and synthesizing by using the residues and wastes of wood and light wood. The artificial board with simulated wood grain has annual ring textures similar to wood grain, and is greatly concerned as a substitute product of a solid wood board, and the surface textures of the simulated wood grain are usually the annual ring textures obtained by pigment printing or wood splicing, but the artificial board is easy to generate cracks in the processing process, so that the confusion between the cracks and the surface textures of the wood grain is difficult to identify.
In the related art, by means of feature extraction and detection by using a big data model, textures inconsistent with normal texture distribution are used as crack textures to be detected for analysis, and in the mode, due to the fact that the artificial board is varied, the applicability of the big data model is low, the recognition effect is poor, and then the accuracy of the detection of the surface defects of the artificial board is low.
Disclosure of Invention
In order to solve the technical problem of lower accuracy of artificial board surface defect detection, the invention provides an intelligent artificial board surface defect detection method based on image processing, which adopts the following technical scheme:
the invention provides an intelligent detection method for surface defects of an artificial board based on image processing, which comprises the following steps:
obtaining a plate gray image of the artificial plate with simulated wood grains, and carrying out texture feature extraction processing on the plate gray image to obtain a texture image; determining the texture complexity degree of the texture image according to the gradient information of the pixel points in the texture image;
determining texture endpoint pixel points in the texture image, determining radian values of the texture according to tangential directions of the texture endpoint pixel points in the same texture, determining radian coefficients according to all radian values of the texture, and determining radian similarity indexes according to all radian values and preset standard radian polar differences, wherein the radian coefficients and the preset standard radian coefficients;
determining the texture interval of adjacent textures according to the distance between corresponding texture pixel points of the adjacent textures in the texture image, and determining the interval similarity index of the texture image according to the texture interval and preset interval values of all the adjacent textures; determining a comprehensive texture index of the texture image according to the radian similarity index and the interval similarity index;
and performing defect detection on the plate gray level image according to the texture complexity degree and the comprehensive texture index to obtain a detection result.
Further, the performing texture feature extraction processing on the plate gray level image to obtain a texture image includes:
performing binary division on the plate gray image based on an Ojin threshold algorithm to obtain a binary image;
and detecting a wood grain texture region in the binary image based on Hough transformation, and extracting the wood grain texture region from the plate gray level image to obtain a texture image.
Further, the gradient information includes a gradient magnitude and a gradient direction, and the determining the texture complexity of the texture image according to the gradient information of the pixel points in the texture image includes:
constructing a direction gradient histogram according to gradient amplitude values and gradient directions of all pixel points in the texture image;
taking a normalized value of the sum value of gradient amplitude values in each direction region in the direction gradient histogram as a direction distribution probability;
and calculating the information entropy of the texture image according to the direction distribution probability of the direction region by using an information entropy formula, and taking the information entropy as the texture complexity.
Further, the determining the radian value of the texture according to the tangential direction of the endpoint pixel point of the texture in the same texture includes:
and obtaining normals of the texture endpoint pixel points in the tangential direction, wherein the normals of the two endpoint pixel points intersect at the center of the texture, and the normalized value of the included angle between the center of the circle and the two normals is used as the radian value of the texture.
Further, the determining radian coefficients according to the radian values of all the textures includes:
and calculating the average value of all radian values of the textures as radian coefficients.
Further, determining the radian similarity index according to all the radian values and the preset standard radian polar differences, wherein the radian coefficient and the preset standard radian coefficient comprise:
calculating the difference value of the maximum value and the minimum value in the radian values to serve as the radian range, and calculating the absolute value of the difference value of the radian range and the preset standard radian range to serve as the radian range difference;
calculating the absolute value of the difference between the radian coefficient and a preset standard radian coefficient as the radian coefficient difference;
and determining the radian similarity index according to the radian range difference and the radian coefficient difference, wherein the radian range difference and the radian similarity index are in negative correlation, the radian coefficient difference and the radian similarity index are in negative correlation, and the numerical value of the radian similarity index is a normalized numerical value.
Further, the determining the texture interval of the adjacent texture according to the distance between the corresponding texture pixel points of the adjacent texture in the texture image includes:
taking the average value of the line segment lengths formed by the pixel points of the texture end points of any texture and the corresponding circle centers as the radius distance of the texture;
and taking two adjacent textures as adjacent textures, and calculating the absolute value of the difference value of the radius distances of the adjacent textures as the texture distance of the adjacent textures.
Further, the determining a pitch similarity index of the texture image according to the texture pitches of all adjacent textures and a preset pitch value includes:
calculating the average value of the texture intervals of all the adjacent textures as an interval average value;
and calculating an inverse proportion normalization value of the absolute value of the difference value between the pitch mean value and the preset pitch value as a pitch similarity index.
Further, the radian similarity index and the comprehensive texture index are in positive correlation, and the interval similarity index and the comprehensive texture index are in positive correlation.
Further, the detecting the defect of the plate gray level image according to the texture complexity and the comprehensive texture index to obtain a detection result includes:
obtaining a texture abnormality degree according to the texture complexity degree and the comprehensive texture index, wherein the texture complexity degree and the texture abnormality degree are in a positive correlation, the comprehensive texture index and the texture abnormality degree are in an inverse correlation, and the value of the texture abnormality degree is a normalized value;
when the texture abnormality degree is larger than a preset abnormality degree threshold value, determining that the detection result is abnormal texture of the surface of the artificial board;
and when the abnormal texture degree is smaller than or equal to a preset abnormal degree threshold value, determining that the detection result is that the surface texture of the artificial board is normal.
The invention has the following beneficial effects:
according to the invention, the texture image is obtained through feature extraction, and then the texture complexity is determined according to the pixel point gradient information in the texture image, and as the texture corresponding to the artificial board with the cracks is more complex, the surface defect of the artificial board can be effectively detected according to the texture complexity; determining the radian value of each texture according to the tangential direction of each texture endpoint pixel point, further determining radian similarity indexes, and taking the radian of wood grains as the judgment basis of anomaly detection, so that effective distinction can be carried out on the radian based on the distinction between the texture of the simulated wood grains and the crack texture; the pitch similarity index is determined through the texture pitch of adjacent textures, so that the pitch of the textures is analyzed, the comprehensive texture index of the texture image is determined according to the radian similarity index and the pitch similarity index, the interference of cracks to the textures can be represented through the comprehensive texture index, and therefore the severity of the cracks can be accurately identified.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of an intelligent detection method for surface defects of an artificial board based on image processing according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a gray scale image of a plate according to an embodiment of the present invention.
Detailed Description
In order to further explain the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of specific implementation, structure, characteristics and effects of the intelligent detection method for the surface defects of the artificial board based on image processing, which is provided by the invention, with reference to the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a specific scheme of an intelligent detection method for the surface defects of an artificial board based on image processing, which is specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of an intelligent detection method for surface defects of an artificial board based on image processing according to an embodiment of the present invention is shown, where the method includes:
s101: obtaining a plate gray image of the artificial plate with simulated wood grains, and carrying out texture feature extraction treatment on the plate gray image to obtain a texture image; and determining the texture complexity degree of the texture image according to the gradient information of the pixel points in the texture image.
An industrial camera may be disposed directly above the artificial wood grain board production line for capturing board images, and then, image preprocessing is performed on the board images, where the image preprocessing includes image denoising processing and image graying processing to obtain board gray scale images, as shown in fig. 2, and fig. 2 is a schematic diagram of the board gray scale images provided by an embodiment of the present invention.
Further, in some embodiments of the present invention, the texture feature extraction processing is performed on the panel gray scale image to obtain a texture image, including: performing binary division on the plate gray image based on an Ojin threshold algorithm to obtain a binary image; and detecting a wood grain texture region in the binary image based on Hough transformation, and extracting the wood grain texture region from the plate gray level image to obtain a texture image.
The features in the plate gray image mainly comprise texture features corresponding to simulated wood grains, in the embodiment of the invention, the pixel points in the plate gray image can be subjected to binary processing by using an OGn threshold algorithm to obtain a binary image, then, the wood grain texture region in the binary image is detected based on Hough transformation, and the wood grain texture region is extracted from the plate gray image to obtain a texture image.
Further, in some embodiments of the present invention, determining the texture complexity of the texture image according to gradient information of pixel points in the texture image includes: constructing a direction gradient histogram according to gradient amplitude values and gradient directions of all pixel points in the texture image; taking a normalized value of the sum value of gradient amplitude values in each direction region in the direction gradient histogram as a direction distribution probability; and calculating the information entropy of the texture image according to the direction distribution probability of the direction region by using an information entropy formula, and taking the information entropy as the texture complexity.
The direction gradient histogram is a histogram constructed by classifying the pixel points according to gradient directions of all the pixel points in the texture image, obtaining direction distribution probability of the corresponding class according to a sum value normalization value of gradient amplitude values in each class, taking the class of the gradient direction as an abscissa and the direction distribution probability as an ordinate, wherein the direction gradient histogram is a technology well known to a person skilled in the art, and the process of obtaining the direction gradient histogram is not repeated.
In the embodiment of the present invention, the information entropy of the texture image is obtained by calculating the class corresponding to the gradient direction and the direction distribution probability corresponding to each class, and the corresponding information entropy formula may specifically be, for example:
where H represents the information entropy of the texture image, L represents the total number of gradient direction categories, L represents the index of the gradient direction category,the direction distribution probability representing the class i gradient direction category,representing a logarithmic function.
In the embodiment of the invention, the information entropy formula is a technology well known in the art, the texture characteristics of all pixel points in the texture image can be calculated through the information entropy formula, the texture distribution in the texture image is characterized as more complex when the information entropy is larger, the texture distribution in the texture image is characterized as more simple when the information entropy is smaller, the information entropy is taken as the texture complexity degree, and therefore the texture complexity degree can be ensured to be capable of representing the distribution condition of the texture in the image.
S102: determining texture endpoint pixel points in a texture image, determining radian values of textures according to tangential directions of the texture endpoint pixel points in the same texture, determining radian coefficients according to the radian values of all the textures, and determining radian similarity indexes according to all the radian values and a preset standard radian range, and determining radian coefficients and a preset standard radian coefficient.
In the embodiment of the invention, a plurality of determining modes can be used for determining the pixel points of the texture end points, for example, all the pixel points representing the same texture are determined, then two poles distributed on two sides of the texture direction in all the pixel points are used as the pixel points of the texture end points, and the radian value of each texture is calculated by acquiring the pixel points of the texture end points, and the corresponding calculating modes are as follows.
Further, in some embodiments of the present invention, determining radian values of textures according to tangential directions of texture endpoint pixel points in the same texture includes: and obtaining normals of the pixel points of the texture end points in the tangential direction, wherein the normals of the pixel points of the two end points are intersected with the circle center of the texture, and the normalized value of the included angle between the circle center and the two normals is used as the radian value of the texture.
In the embodiment of the invention, the tangent line of the pixel point of the texture end point is the tangent line direction of the position where the texture is located, and it can be understood that because the texture can be regarded as an approximate arc shape, the embodiment of the invention can perform curve fitting on any texture, then determine the tangent line direction of two end points of the curve obtained by fitting by using feature extraction, and make normal lines along the pixel point of the texture end point perpendicular to the tangent line, the two normal lines can intersect at a point, the point is used as the circle center of the texture, and the included angle formed by the circle center of the texture and the two normal lines is used as the included angle of the texture.
After the included angle of the texture is determined, the normalization value of the included angle of the texture can be obtained, for example, the ratio of the included angle of the texture to the 360-degree angle can be calculated as the radian value of the texture, or the angle value can be directly normalized, and the normalization value is not limited.
Further, in some embodiments of the present invention, determining radian coefficients from radian values of all textures includes: and calculating the average value of the radian values of all textures as the radian coefficient.
After determining the radian value of each texture, the embodiment of the invention can calculate the average value of the radian values of all textures as the texture coefficient of the texture image, and can understand that under the condition that the defects such as cracks and the like are generated in the wood grain pattern, the texture identification is caused to generate errors, and then the radian value is changed, so that whether the abnormal condition occurs can be determined according to the radian coefficient.
Further, in some embodiments of the present invention, determining the radian similarity index according to all the radian values and the preset standard radian polar difference, the radian coefficient and the preset standard radian coefficient includes: calculating the difference value of the maximum value and the minimum value in the radian values to serve as the radian polar difference, and calculating the absolute value of the difference value of the radian polar difference and the preset standard radian polar difference to serve as the radian polar difference; calculating the absolute value of the difference between the radian coefficient and a preset standard radian coefficient as the radian coefficient difference; and determining an radian similarity index according to the radian polar difference and the radian coefficient difference, wherein the radian polar difference and the radian similarity index are in a negative correlation, the radian coefficient difference and the radian similarity index are in a negative correlation, and the value of the radian similarity index is a normalized value.
In the embodiment of the invention, the preset standard radian range may be the radian range corresponding to the standard simulated wood grain artificial board, alternatively, the preset standard radian range may be specifically, for example, 0.1, or may be adjusted according to the actual production situation, which is not limited.
In the embodiment of the invention, the maximum value and the minimum value of the radian values of all textures in the texture image are calculated, and the difference between the maximum value and the minimum value is calculated to be used as the radian range, and it can be understood that the maximum value and the minimum value of the radian values can represent the radian range of all textures in the texture image, and the absolute value of the difference between the calculated radian range and the preset standard radian range can represent the difference between the radian range and the standard condition, namely, the difference between the radian range of the textures in the texture image and the difference between the radian range and the standard condition.
The preset standard radian coefficient is a radian coefficient corresponding to the simulated wood grain artificial board under the standard condition, and optionally, the preset standard radian coefficient may be specifically, for example, 0.1, or may be adjusted according to the actual production condition, which is not limited.
In the embodiment of the invention, by calculating the absolute value of the difference between the radian coefficient and the preset standard radian coefficient, it can be understood that, because the radian coefficient can represent the overall radian condition of all textures in the texture image, the absolute value of the difference between the calculated radian coefficient and the preset standard radian coefficient can represent the difference between the radian distribution and the standard condition, that is, the difference between the radian coefficient and the radian distribution of the textures in the texture image can represent the difference between the radian distribution and the standard condition.
It can be understood that in the embodiment of the present invention, the positive correlation relationship indicates that the dependent variable increases with the increase of the independent variable, and the dependent variable decreases with the decrease of the independent variable, and the specific relationship may be a multiplication relationship, an addition relationship, an idempotent of an exponential function, which is determined by practical application; the negative correlation indicates that the dependent variable decreases with increasing independent variable, and the dependent variable increases with decreasing independent variable, which may be a subtraction relationship, a division relationship, or the like, and is determined by the actual application.
Thus, a calculation formula for determining the radian similarity index from the radian difference and the radian coefficient difference may be, for example:
wherein F represents a radian similarity index,representing the maximum value of the radian value,representing the minimum value of the radian value,the standard radian range is shown to be preset,the radian factor is represented by,the preset standard radian coefficient is indicated,in one embodiment of the present invention, the normalization process may be specifically, for example, maximum and minimum normalization processes, and the normalization in the subsequent steps may be performed by using the maximum and minimum normalization processes, and in other embodiments of the present invention, other normalization methods may be selected according to a specific range of values, which will not be described herein.The representation takes absolute value.
As shown in the calculation formula of the radian similarity index,can be expressed as a very bad arc degree,the radian range difference is represented, and as the difference between the radian range and the preset standard radian range is larger, the difference between the currently detected artificial wood grain board and the artificial wood grain board produced under the standard condition can be represented, namely, the similarity is lower, the corresponding radian similarity index is lower, and the radian range difference and the radian similarity index are in a negative correlation.
As shown in the calculation formula of the radian similarity index,the radian coefficient difference is represented, and the radian coefficient is the average value of radian values of all textures in the texture image, namely, the radian coefficient represents the distribution of the textures, and the larger the difference between the radian coefficient and a preset standard radian coefficient is, the larger the difference between the distribution of the corresponding textures can be represented, and crack textures with abnormal angles can appear, so that the radian coefficient difference and the radian similarity index form a negative correlation.
That is, the greater the radian similarity index, the higher the degree of similarity between the texture in the characterization texture image and the standard texture, and the more normal the corresponding texture shape.
S103: determining the texture interval of adjacent textures according to the distances between corresponding texture pixel points of the adjacent textures in the texture image, and determining the interval similarity index of the texture image according to the texture intervals of all the adjacent textures and preset interval values; and determining the comprehensive texture index of the texture image according to the radian similarity index and the interval similarity index.
It can be understood that if the simulated wood grain artificial board contains crack defects, the corresponding texture image can misjudge the cracks as normal textures due to the existence of cracks, so that the texture distribution is different.
Further, in some embodiments of the present invention, determining a texture pitch of an adjacent texture according to a distance between corresponding texture pixel points of the adjacent texture in a texture image includes: taking the average value of the line segment lengths formed by the pixel points of the texture end points of any texture and the corresponding circle centers as the radius distance of the texture; and taking two adjacent textures as the adjacent textures, and calculating the absolute value of the difference value of the radius distances of the adjacent textures as the texture distance of the adjacent textures.
That is, in the embodiment of the present invention, the length of a line segment formed by the endpoint pixel points of the texture and the corresponding center is calculated, and since each texture has two endpoint pixel points, the embodiment of the present invention calculates the average value of the lengths of the line segments formed by the endpoint pixel points and the corresponding center as the radius distance of the texture, and calculates the absolute value of the difference value of the radius distances of adjacent textures as the texture distance of the adjacent textures, under the simulated wood grain texture, the difference may be referred to as the texture distance because the wood grain is usually annual ring wood grain, that is, the radius distances corresponding to the adjacent wood grains may be different due to the characteristics of annual rings.
Of course, when applied to other types of artificial boards, the distance between adjacent textures may be directly used as the texture pitch, for example, by calculating the center points of the textures, and the distance between the center points may be used as the texture pitch, which is not limited.
Further, in some embodiments of the present invention, determining a pitch similarity index of a texture image according to a texture pitch and a preset pitch value of all neighboring textures includes: calculating the average value of the texture intervals of all adjacent textures as an interval average value; and calculating an inverse proportion normalization value of the difference absolute value of the pitch mean value and the preset pitch value as a pitch similarity index.
In the embodiment of the present invention, the texture intervals between all textures and adjacent textures can be calculated, and due to the complex distribution of textures, one texture may be adjacent to at least two other textures, so that the embodiment of the present invention can count the texture intervals of all adjacent textures, then calculate the average value of the texture intervals of all adjacent textures as the interval average value, and determine the interval similarity index according to the interval average value and the preset interval value, and the corresponding calculation formula may specifically be, for example:
in the method, in the process of the invention,represents a pitch similarity index, K represents the total number of texture pitches, K represents an index of texture pitches,a value representing the kth texture pitch,the mean value of the distance is represented,representing a preset spacing value, alternatively, the preset spacing value may be specifically for example 0.75,the normalization process is represented.
In the embodiment of the invention, when the difference between the pitch average value and the preset pitch value is larger, the pitch of textures in the artificial board with simulated wood grains can be influenced by cracks or other abnormal factors, so that the pitch similarity index is obtained through calculation, and the more serious the crack influence is, the smaller the corresponding pitch similarity index is.
Therefore, the comprehensive texture index of the texture image is determined according to the radian similarity index and the spacing similarity index, and as the radian similarity index is larger, the similarity degree of the texture in the texture image and the standard texture is higher, the corresponding texture shape is more normal, the spacing similarity index is larger, the difference between the texture spacing in the texture image and the standard texture spacing is smaller, and the corresponding texture spacing is more normal.
Therefore, in the embodiment of the invention, the radian similarity index and the comprehensive texture index are in positive correlation, the interval similarity index and the comprehensive texture index are in positive correlation, the sum of the radian similarity index and the interval similarity index can be calculated to serve as the comprehensive texture index, or the product of the radian similarity index and the interval similarity index can be calculated to serve as the comprehensive texture index, and the method is not limited.
S104: and performing defect detection on the plate gray level image according to the texture complexity and the comprehensive texture index to obtain a detection result.
It can be understood that the comprehensive texture index is an index for identifying texture distribution by combining radian and spacing, and represents a coefficient of the normal degree of the texture distribution, and the texture complexity degree is an index for representing the chaotic degree of the texture distribution.
Further, in some embodiments of the present invention, performing defect detection on a gray image of a plate according to a texture complexity degree and a comprehensive texture index to obtain a detection result, including: obtaining the texture abnormality degree according to the texture complexity degree and the comprehensive texture index, wherein the texture complexity degree and the texture abnormality degree are in positive correlation, and the comprehensive texture index and the texture abnormality degree are in inverse correlation; when the abnormal degree of the texture is larger than a preset abnormal degree threshold value, determining that the detection result is abnormal texture of the surface of the artificial board; and when the abnormal degree of the texture is smaller than or equal to a preset abnormal degree threshold value, determining that the detection result is that the surface texture of the artificial board is normal.
In the embodiment of the present invention, the calculation formula of the texture anomaly degree may specifically be, for example:
in the method, in the process of the invention,represents the degree of texture abnormality, H represents the degree of texture complexity, W represents the comprehensive texture index,represents a constant coefficient, a safety value set to prevent the denominator from being 0,the normalization process is represented.
In the embodiment of the invention, the abnormal state of the texture in the texture image is represented by the abnormal degree of the texture, and the comprehensive texture index is a coefficient representing the normal degree of the texture distribution, so that the comprehensive texture index and the abnormal degree of the texture are in an inverse correlation relation, and the texture complexity index represents the index of the chaotic degree of the texture distribution, and the more the chaotic represents the abnormal texture distribution, the more the texture complexity and the abnormal degree of the texture are in a positive correlation relation.
After the texture abnormality degree is calculated, the texture abnormality degree is compared with the preset abnormality degree threshold, and when the texture abnormality degree is larger than the preset abnormality degree threshold, the corresponding crack defect and serious abnormal conditions of the textures generated in the manufacturing process of the simulated wood grain artificial board can be represented, and then the detection result of the simulated wood grain artificial board can be determined to be the surface texture abnormality of the artificial board. When the abnormal texture degree is less than or equal to a preset abnormal degree threshold, the corresponding crack defect and the abnormal condition of the texture generated in the manufacturing process of the artificial board with simulated wood grains can be represented to have small influence, and in an acceptable error range, the detection result of the artificial board with simulated wood grains can be determined to be that the surface texture of the artificial board is normal, and optionally, the preset abnormal degree threshold can be specifically, for example, 0.8, which is not limited.
According to the invention, the texture image is obtained through feature extraction, and then the texture complexity is determined according to the pixel point gradient information in the texture image, and as the texture corresponding to the artificial board with the cracks is more complex, the surface defect of the artificial board can be effectively detected according to the texture complexity; determining the radian value of each texture according to the tangential direction of each texture endpoint pixel point, further determining radian similarity indexes, and taking the radian of wood grains as the judgment basis of anomaly detection, so that effective distinction can be carried out on the radian based on the distinction between the texture of the simulated wood grains and the crack texture; the pitch similarity index is determined through the texture pitch of adjacent textures, so that the pitch of the textures is analyzed, the comprehensive texture index of the texture image is determined according to the radian similarity index and the pitch similarity index, the interference of cracks to the textures can be represented through the comprehensive texture index, and therefore the severity of the cracks can be accurately identified.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (10)
1. An intelligent detection method for surface defects of an artificial board based on image processing is characterized by comprising the following steps:
obtaining a plate gray image of the artificial plate with simulated wood grains, and carrying out texture feature extraction processing on the plate gray image to obtain a texture image; determining the texture complexity degree of the texture image according to the gradient information of the pixel points in the texture image;
determining texture endpoint pixel points in the texture image, determining radian values of the texture according to tangential directions of the texture endpoint pixel points in the same texture, determining radian coefficients according to all radian values of the texture, and determining radian similarity indexes according to all radian values and preset standard radian polar differences, wherein the radian coefficients and the preset standard radian coefficients;
determining the texture interval of adjacent textures according to the distance between corresponding texture pixel points of the adjacent textures in the texture image, and determining the interval similarity index of the texture image according to the texture interval and preset interval values of all the adjacent textures; determining a comprehensive texture index of the texture image according to the radian similarity index and the interval similarity index;
and performing defect detection on the plate gray level image according to the texture complexity degree and the comprehensive texture index to obtain a detection result.
2. The intelligent detection method for surface defects of an artificial board based on image processing as claimed in claim 1, wherein the step of performing texture feature extraction processing on the gray level image of the board to obtain a texture image comprises the steps of:
performing binary division on the plate gray image based on an Ojin threshold algorithm to obtain a binary image;
and detecting a wood grain texture region in the binary image based on Hough transformation, and extracting the wood grain texture region from the plate gray level image to obtain a texture image.
3. The intelligent detection method for surface defects of an artificial board based on image processing as claimed in claim 1, wherein the gradient information includes a gradient magnitude and a gradient direction, and the determining the texture complexity of the texture image according to the gradient information of the pixel points in the texture image includes:
constructing a direction gradient histogram according to gradient amplitude values and gradient directions of all pixel points in the texture image;
taking a normalized value of the sum value of gradient amplitude values in each direction region in the direction gradient histogram as a direction distribution probability;
and calculating the information entropy of the texture image according to the direction distribution probability of the direction region by using an information entropy formula, and taking the information entropy as the texture complexity.
4. The intelligent detection method for surface defects of an artificial board based on image processing as claimed in claim 1, wherein said determining the radian value of the texture according to the tangential direction of the endpoint pixel points of the texture in the same texture comprises:
and obtaining normals of the texture endpoint pixel points in the tangential direction, wherein the normals of the two endpoint pixel points intersect at the center of the texture, and the normalized value of the included angle between the center of the circle and the two normals is used as the radian value of the texture.
5. The intelligent detection method for surface defects of an artificial board based on image processing as claimed in claim 1, wherein the determining radian factor according to radian values of all textures comprises:
and calculating the average value of all radian values of the textures as radian coefficients.
6. The intelligent detection method for surface defects of an artificial board based on image processing as claimed in claim 1, wherein the determining the radian similarity index according to all the radian values and the preset standard radian range, the radian coefficient and the preset standard radian coefficient comprises:
calculating the difference value of the maximum value and the minimum value in the radian values to serve as the radian range, and calculating the absolute value of the difference value of the radian range and the preset standard radian range to serve as the radian range difference;
calculating the absolute value of the difference between the radian coefficient and a preset standard radian coefficient as the radian coefficient difference;
and determining the radian similarity index according to the radian range difference and the radian coefficient difference, wherein the radian range difference and the radian similarity index are in negative correlation, the radian coefficient difference and the radian similarity index are in negative correlation, and the numerical value of the radian similarity index is a normalized numerical value.
7. The intelligent detection method for surface defects of an artificial board based on image processing as claimed in claim 4, wherein determining the texture interval of adjacent textures according to the distance between corresponding texture pixel points of adjacent textures in the texture image comprises:
taking the average value of the line segment lengths formed by the pixel points of the texture end points of any texture and the corresponding circle centers as the radius distance of the texture;
and taking two adjacent textures as adjacent textures, and calculating the absolute value of the difference value of the radius distances of the adjacent textures as the texture distance of the adjacent textures.
8. The intelligent detection method for surface defects of an artificial board based on image processing according to claim 1, wherein the determining a pitch similarity index of the texture image according to the texture pitches of all adjacent textures and a preset pitch value comprises:
calculating the average value of the texture intervals of all the adjacent textures as an interval average value;
and calculating an inverse proportion normalization value of the absolute value of the difference value between the pitch mean value and the preset pitch value as a pitch similarity index.
9. The intelligent detection method for surface defects of artificial boards based on image processing according to claim 1, wherein the radian similarity index and the comprehensive texture index are in positive correlation, and the interval similarity index and the comprehensive texture index are in positive correlation.
10. The intelligent detection method for surface defects of an artificial board based on image processing as claimed in claim 1, wherein the defect detection is performed on the gray level image of the board according to the texture complexity and the comprehensive texture index to obtain a detection result, and the method comprises the following steps:
obtaining a texture abnormality degree according to the texture complexity degree and the comprehensive texture index, wherein the texture complexity degree and the texture abnormality degree are in a positive correlation, the comprehensive texture index and the texture abnormality degree are in an inverse correlation, and the value of the texture abnormality degree is a normalized value;
when the texture abnormality degree is larger than a preset abnormality degree threshold value, determining that the detection result is abnormal texture of the surface of the artificial board;
and when the abnormal texture degree is smaller than or equal to a preset abnormal degree threshold value, determining that the detection result is that the surface texture of the artificial board is normal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310712573.6A CN116452589B (en) | 2023-06-16 | 2023-06-16 | Intelligent detection method for surface defects of artificial board based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310712573.6A CN116452589B (en) | 2023-06-16 | 2023-06-16 | Intelligent detection method for surface defects of artificial board based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116452589A true CN116452589A (en) | 2023-07-18 |
| CN116452589B CN116452589B (en) | 2023-10-20 |
Family
ID=87134158
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310712573.6A Active CN116452589B (en) | 2023-06-16 | 2023-06-16 | Intelligent detection method for surface defects of artificial board based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116452589B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116993742A (en) * | 2023-09-28 | 2023-11-03 | 中色(天津)新材料科技有限公司 | Nickel alloy rolling defect detection method based on machine vision |
| CN117830298A (en) * | 2024-03-01 | 2024-04-05 | 苏州翰微材料科技有限公司 | Visual inspection system for defects of turbine guide vane of aero-engine |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070104359A1 (en) * | 2005-11-10 | 2007-05-10 | Mauricio Telias | Digital inspection of the physical quality of plain surfaces |
| US20150324998A1 (en) * | 2014-05-06 | 2015-11-12 | Nant Holdings Ip, Llc | Image-based feature detection using edge vectors |
| CN106952258A (en) * | 2017-03-23 | 2017-07-14 | 南京汇川图像视觉技术有限公司 | A kind of bottle mouth defect detection method based on gradient orientation histogram |
| CN110223296A (en) * | 2019-07-08 | 2019-09-10 | 山东建筑大学 | A kind of screw-thread steel detection method of surface flaw and system based on machine vision |
| CN113989280A (en) * | 2021-12-28 | 2022-01-28 | 武汉市鑫景诚路桥钢模有限公司 | Steel structure welding crack defect detection method based on image processing technology |
| WO2022027949A1 (en) * | 2020-08-04 | 2022-02-10 | 湖南大学 | Machine vision-based detecting method and system for glass bottle bottom defects |
| CN114881965A (en) * | 2022-04-29 | 2022-08-09 | 蔡志刚 | Wood board joint detection method based on artificial intelligence and image processing |
| CN114998227A (en) * | 2022-05-20 | 2022-09-02 | 江苏博腾家用纺织品有限公司 | Cloth printing and dyeing defect detection method and system based on image processing |
| CN115601296A (en) * | 2022-08-25 | 2023-01-13 | 淮南师范学院（Cn） | Surface defect detection device and detection method based on deep learning |
| CN116012357A (en) * | 2023-02-09 | 2023-04-25 | 河南职业技术学院 | Automobile circular trailer hole air line defect detection method based on machine vision |
| CN116030058A (en) * | 2023-03-29 | 2023-04-28 | 无锡斯达新能源科技股份有限公司 | Quality evaluation method for surface roughness of polishing pad |
-
2023
- 2023-06-16 CN CN202310712573.6A patent/CN116452589B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070104359A1 (en) * | 2005-11-10 | 2007-05-10 | Mauricio Telias | Digital inspection of the physical quality of plain surfaces |
| US20150324998A1 (en) * | 2014-05-06 | 2015-11-12 | Nant Holdings Ip, Llc | Image-based feature detection using edge vectors |
| CN106952258A (en) * | 2017-03-23 | 2017-07-14 | 南京汇川图像视觉技术有限公司 | A kind of bottle mouth defect detection method based on gradient orientation histogram |
| CN110223296A (en) * | 2019-07-08 | 2019-09-10 | 山东建筑大学 | A kind of screw-thread steel detection method of surface flaw and system based on machine vision |
| WO2022027949A1 (en) * | 2020-08-04 | 2022-02-10 | 湖南大学 | Machine vision-based detecting method and system for glass bottle bottom defects |
| CN113989280A (en) * | 2021-12-28 | 2022-01-28 | 武汉市鑫景诚路桥钢模有限公司 | Steel structure welding crack defect detection method based on image processing technology |
| CN114881965A (en) * | 2022-04-29 | 2022-08-09 | 蔡志刚 | Wood board joint detection method based on artificial intelligence and image processing |
| CN114998227A (en) * | 2022-05-20 | 2022-09-02 | 江苏博腾家用纺织品有限公司 | Cloth printing and dyeing defect detection method and system based on image processing |
| CN115601296A (en) * | 2022-08-25 | 2023-01-13 | 淮南师范学院（Cn） | Surface defect detection device and detection method based on deep learning |
| CN116012357A (en) * | 2023-02-09 | 2023-04-25 | 河南职业技术学院 | Automobile circular trailer hole air line defect detection method based on machine vision |
| CN116030058A (en) * | 2023-03-29 | 2023-04-28 | 无锡斯达新能源科技股份有限公司 | Quality evaluation method for surface roughness of polishing pad |
Non-Patent Citations (1)
| Title |
|---|
| 宋开勇;王传旭;董向鹏;: "轮胎纹理提取与缺陷检测方法", 计算机工程, no. 18, pages 291 - 293 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116993742A (en) * | 2023-09-28 | 2023-11-03 | 中色(天津)新材料科技有限公司 | Nickel alloy rolling defect detection method based on machine vision |
| CN116993742B (en) * | 2023-09-28 | 2023-12-22 | 中色(天津)新材料科技有限公司 | Nickel alloy rolling defect detection method based on machine vision |
| CN117830298A (en) * | 2024-03-01 | 2024-04-05 | 苏州翰微材料科技有限公司 | Visual inspection system for defects of turbine guide vane of aero-engine |
| CN117830298B (en) * | 2024-03-01 | 2024-05-07 | 苏州翰微材料科技有限公司 | Visual inspection system for defects of turbine guide vane of aero-engine |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116452589B (en) | 2023-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116452589B (en) | Intelligent detection method for surface defects of artificial board based on image processing | |
| CN116168026B (en) | Water quality detection method and system based on computer vision | |
| CN115311292B (en) | Strip steel surface defect detection method and system based on image processing | |
| CN116205919B (en) | Hardware part production quality detection method and system based on artificial intelligence | |
| CN116977358B (en) | Visual auxiliary detection method for corrugated paper production quality | |
| CN108550166B (en) | Spatial target image matching method | |
| CN116091504B (en) | Connecting pipe connector quality detection method based on image processing | |
| CN115311275B (en) | Method and system for detecting surface defects of electronic component | |
| CN116137036B (en) | Gene detection data intelligent processing system based on machine learning | |
| CN115311629B (en) | Abnormal bending precision monitoring system of bending machine | |
| CN114332026A (en) | Visual detection method and device for scratch defects on surface of nameplate | |
| CN110619146A (en) | Polycrystalline silicon cell crack defect detection method based on structural similarity measurement | |
| CN115082477B (en) | Semiconductor wafer processing quality detection method based on light reflection removing effect | |
| CN117689655B (en) | Metal button surface defect detection method based on computer vision | |
| CN115311267A (en) | Method for detecting abnormity of check fabric | |
| CN115018835A (en) | Automobile starter gear detection method | |
| CN115359047A (en) | Abnormal defect detection method for intelligent welding of PCB (printed circuit board) | |
| CN115861987A (en) | Intelligent electric power fitting defect identification method for on-line monitoring of power transmission line | |
| CN115272350A (en) | Method for detecting production quality of computer PCB mainboard | |
| CN112950594B (en) | Method, device and storage medium for detecting surface defects of product | |
| CN114332081A (en) | Textile surface abnormity determination method based on image processing | |
| CN116758056B (en) | Electrical terminal production defect detection method | |
| CN115855961B (en) | Distribution box fault detection method used in operation | |
| CN116883408A (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
| CN111126371B (en) | Coarse pointer dial reading method based on image processing |
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 |