CN115375676A - Stainless steel product quality detection method based on image recognition - Google Patents
Stainless steel product quality detection method based on image recognition Download PDFInfo
- Publication number
- CN115375676A CN115375676A CN202211298717.XA CN202211298717A CN115375676A CN 115375676 A CN115375676 A CN 115375676A CN 202211298717 A CN202211298717 A CN 202211298717A CN 115375676 A CN115375676 A CN 115375676A
- Authority
- CN
- China
- Prior art keywords
- analyzed
- area
- pixel point
- gray
- region
- 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
Images
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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- 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 a stainless steel product quality detection method based on image recognition, which comprises the following steps: acquiring a surface gray image of a stainless steel product, recording an area in a closed edge as an area to be analyzed, and calculating a shape characteristic index of the area to be analyzed; obtaining a gray characteristic index according to the gradient direction and the characteristic angle of the pixel point in the region to be analyzed; obtaining a first possibility according to the ratio of the shape characteristic index and the gray characteristic index; determining a shadow area, and obtaining a second possibility according to the area to be analyzed, the light source and the adjacent shadow area; obtaining a first significance level according to the first and second possibilities; and calculating the second significance to further obtain the global significance, enhancing the surface gray image by using the global significance, carrying out edge detection on the enhanced image to obtain a defect area, and further determining the quality of the stainless steel product according to the defect area. The invention can obtain the defect area more accurately.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a stainless steel product quality detection method based on image recognition.
Background
Stainless steel products are made into tableware, kitchenware and the like, the surfaces of the stainless steel products are easy to treat and corrosion resistant, and the stainless steel products are widely applied. However, steel quality is not pure, large-particle inclusions exist, and scratches are generated during rolling, so that many defective products are produced. When stainless steel products are produced, hard large-particle inclusions falling in during rolling are pressed into the surface of a rolled piece, and due to the fact that gaps exist between the large-particle inclusions and the rolled piece, the inclusions are prone to falling off from the surface of the rolled piece, and pits are formed in the surface of the rolled piece. The pits on the rolling surface are different in shape depth and appearance rate. This can cause a significant quality problem for the stainless steel product being produced, and therefore, it is important to detect the pit defect on the surface of the stainless steel product. In the prior art, a threshold segmentation method is usually adopted to process the surface image of the stainless steel product to obtain a pit part, but the method is easily affected by other defects and illumination, so that the segmentation result is inaccurate.
Disclosure of Invention
In order to solve the above technical problems, the present invention aims to provide a method for detecting the quality of a stainless steel product based on image recognition, which adopts the following technical scheme:
acquiring a surface gray image of the stainless steel product, carrying out edge detection on the image, and acquiring a region in a closed edge and marking as a region to be analyzed; obtaining a shape characteristic index according to the area of the region to be analyzed and the area of the minimum circumscribed circle of the region;
acquiring a connecting line of each pixel point in the area to be analyzed and the pixel point corresponding to the maximum gray value, recording an included angle between the connecting line corresponding to each pixel point and the horizontal direction as a characteristic angle, and acquiring a gray characteristic index according to the gradient direction and the characteristic angle of the pixel point in the area to be analyzed; obtaining a first possibility according to the ratio of the shape characteristic index and the gray characteristic index;
calculating the gray value mean value of pixel points in the area to be analyzed, and determining a shadow area according to the gray value mean value and the position relation of the area to be analyzed; acquiring the position of a light source, and obtaining a second possibility according to angles corresponding to the connection lines of the central pixel point of the region to be analyzed and the light source and the central pixel point of the shadow region adjacent to the light source; obtaining a first significance based on the first and second likelihoods;
and obtaining a second significance according to the texture information of the surface gray level image, obtaining a global significance according to the first and second significances, enhancing the surface gray level image by using the global significance, carrying out edge detection on the enhanced image to obtain a defect area, and further determining the quality of the stainless steel product according to the defect area.
Preferably, the method for acquiring the gray characteristic index specifically includes:
wherein U represents the gray scale characteristic index of the region to be analyzed,
representing the angle corresponding to the gradient direction of the ith pixel point in the region to be analyzed,
representing the characteristic angle of the ith pixel point in the region to be analyzed,
and representing the number of pixel points in the region to be analyzed.
Preferably, the determining the shadow region according to the gray value mean value and the position relationship of the region to be analyzed specifically includes:
and obtaining two areas to be analyzed which are in an inclusion relationship, and when the difference value of the gray value mean values of the two areas to be analyzed is larger than the gray threshold value, marking the area with the low gray value mean value in the two areas to be analyzed as a shadow area.
Preferably, the obtaining of the position of the light source obtains a second possibility according to angles corresponding to connection lines of a central pixel point of the region to be analyzed and the light source and central pixel points of shadow regions adjacent to the light source, and includes:
constructing a rectangular coordinate system by taking a central pixel point of the surface gray level image as an origin, and acquiring a coordinate of a position where a light source is positioned and coordinates of the central pixel point of a region to be analyzed and a shadow region adjacent to the region to be analyzed; and calculating the included angle between the straight line where the connecting line of the central pixel point of the area to be analyzed and the adjacent shadow area is located and the horizontal direction according to the coordinates of the central pixel point of the area to be analyzed and the coordinates of the central pixel point of the adjacent shadow area, calculating the included angle between the straight line where the connecting line of the light source and the central pixel point of the area to be analyzed and the horizontal direction in a similar way, and obtaining a second possibility according to the difference value of the two included angles.
Preferably, the obtaining the second saliency according to the texture information of the surface grayscale image is specifically:
constructing a gray level co-occurrence matrix according to the gray values of the pixel points and the neighborhood pixel points, and acquiring the entropy value and the energy value of the gray level co-occurrence matrix corresponding to the pixel points according to the gray level co-occurrence matrix; in the same way, a gray level co-occurrence matrix corresponding to the neighborhood pixel point is constructed, and the entropy value and the energy value of the gray level co-occurrence matrix corresponding to the neighborhood pixel point are obtained; and calculating a second significance according to the entropy value and the energy value corresponding to the pixel point and the entropy value and the energy value corresponding to the neighbor pixel point.
Preferably, the enhancing the surface grayscale image by using the global saliency specifically includes:
acquiring the enhanced gray value of each pixel point in each to-be-analyzed area in the surface gray image according to the global significance and the gray value mean value of each to-be-analyzed area and the gray value of each pixel point in each to-be-analyzed area; and acquiring an enhanced image according to the enhanced gray value of each pixel point.
Preferably, the quality of the stainless steel product determined according to the defect area is specifically:
acquiring the areas and the number of all defect regions in the enhanced image, and obtaining a quality evaluation index according to the areas and the number; and setting an evaluation threshold, wherein when the quality evaluation index is less than the evaluation threshold, the corresponding stainless steel product is poor.
The embodiment of the invention at least has the following beneficial effects:
the method comprises the steps of firstly obtaining an area in a closed edge in an image and recording the area as an area to be analyzed, calculating a shape characteristic index according to the area of the area to be analyzed, considering whether the shape characteristic of the area to be analyzed accords with the shape characteristic of an area corresponding to a pit defect, obtaining a gray characteristic index according to the gradient direction and the characteristic angle of a pixel point in the area to be analyzed, considering whether the gray value change of the pixel point in the area to be analyzed accords with the gray change characteristic of the area corresponding to the pit defect, further obtaining a first possibility according to the shape characteristic and the gray characteristic, obtaining a second possibility according to a light source, the area to be analyzed and a shadow area, considering the light and shadow characteristic of the area corresponding to the pit defect due to the existence of shadow, eliminating the interference factor of the bump defect, obtaining a first significance according to the possibilities of two aspects, further obtaining the global significance of the area to be analyzed, carrying out enhancement and inhibition on the image according to significance analysis, and combining the characteristic presented by the existence of the pit defect on the surface of a stainless steel product, giving a certain significance, and further effectively detecting the quality problem formed by the pit defect in the image according to significance. The method can acquire the accurate defect region, so that the quality detection result of the stainless steel product is accurate.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of the method of the invention for detecting the quality of stainless steel products based on image recognition.
Detailed Description
To further illustrate the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description will be given to a method for detecting the quality of a stainless steel product based on image recognition, the specific implementation manner, the structure, the features and the effects thereof according to the present invention, with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
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 following specifically describes a specific scheme of the stainless steel product quality detection method based on image recognition, which is provided by the invention, with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for detecting quality of a stainless steel product based on image recognition according to an embodiment of the present invention is shown, where the method includes the following steps:
acquiring a surface gray image of a stainless steel product, carrying out edge detection on the image, and acquiring a region in a closed edge and marking as a region to be analyzed; and obtaining the shape characteristic index according to the area of the region to be analyzed and the area of the minimum circumscribed circle of the region.
Specifically, the produced stainless steel product is placed on a conveyor belt, the stainless steel product on the conveyor belt is shot by using an industrial camera and a single light source, and the shot and collected image is subjected to preprocessing operations such as denoising and graying to obtain a surface grayscale image of the stainless steel product. In this embodiment, gaussian filtering is used to denoise an image, and a weighted average graying algorithm is used to graye the image, which can be selected by an implementer according to actual situations.
In this embodiment, the light source is positioned directly above any one side edge of the conveyor belt, and when the light source is used for irradiation, the irradiation of the light source and the stainless steel product on the conveyor belt have a certain angle, so that the operator can set the position of the light source according to actual conditions.
In evaluating the quality of a stainless steel product, surface defects of the stainless steel product are an important factor in evaluating the quality of the stainless steel product. When stainless steel products are produced, hard large-particle inclusions falling in during rolling are pressed into the surface of a rolled piece, and due to the fact that gaps exist between the large-particle inclusions and the rolled piece, the inclusions are prone to falling off from the surface of the rolled piece, and pits are formed in the surface of the rolled piece. The quality evaluation of the stainless steel product is obtained according to the pit defect degree by mainly detecting whether the surface of the stainless steel product has the pit defect or not.
And then carrying out edge detection on the surface gray level image of the stainless steel product to obtain a plurality of edges, and marking the area in the closed edge as an area to be analyzed. In the present embodiment, edge detection is performed on the surface gray image by using the canny algorithm. The areas within the closed edges of the detected edges are considered as areas of suspected defects, and subsequent analysis of these areas is required.
And obtaining the area of the region to be analyzed and the area of the minimum circumscribed circle of the region to be analyzed, and obtaining the shape characteristic index according to the ratio of the area of the region to be analyzed and the area of the minimum circumscribed circle of the region to be analyzed. If the surface of the stainless steel product has pit defects, the shape of the position of the pit defects is generally close to a circle or an ellipse, the area of the area suspected of the defects can be compared with the area of the minimum circumscribed circle of the area, namely, the ratio of the area to be analyzed to the area of the minimum circumscribed circle of the area is obtained, the degree that the area to be analyzed is close to the circle is reflected by the ratio, the larger the value of the shape characteristic index of the area to be analyzed is, the closer the area is to the circle, and the more the area is probably the pit defects.
The shape of the pit defect existing on the surface of the stainless steel product is often approximately circular or elliptical, but other shapes may exist, and the present invention analyzes the shape of the pit defect in most cases. Meanwhile, when the area in which each suspected defect exists is analyzed, the larger the shape characteristic index is, the closer the shape of the area is to the circular shape, but the defect presenting the circular shape not only has a pit defect, but also may be other defects such as a bubble defect, so that the area to be analyzed needs to be further analyzed.
Acquiring a connecting line of each pixel point in the region to be analyzed and the pixel point corresponding to the maximum gray value, recording an included angle between the connecting line corresponding to each pixel point and the horizontal direction as a characteristic angle, and acquiring a gray characteristic index according to the gradient direction and the characteristic angle of the pixel point in the region to be analyzed; the first likelihood is obtained from a ratio of the shape characteristic index and the gray characteristic index.
It should be noted that, when analyzing the surface defect of the stainless steel product, since some convex defects may exist on the surface of the stainless steel product, when detecting the pit defect on the surface of the stainless steel product, it is difficult to distinguish the convex defects, which causes false detection, and further analysis needs to be performed on the obtained region to be analyzed. According to the characteristics of pits on the surface of a stainless steel product, different gray features may be formed in the pits due to the influence of illumination, that is, according to the characteristics of concave converging light, in the areas with pit defects, the light converges from the edge outline of the areas to the center of the areas, and the gray values of pixel points in the images from the edge outline of the areas to the center of the areas are gradually increased, that is, the gray values of the pixel points in a part of the areas in the areas with pit defects are higher, so that the possibility that each area is a pit can be analyzed according to the characteristics of gray change in the areas with pit defects.
Specifically, for any one to-be-analyzed area, a pixel point corresponding to the maximum gray value in the to-be-analyzed area is obtained, each pixel point is connected with the pixel point corresponding to the maximum gray value, an included angle between a connecting line corresponding to the pixel point and the horizontal right direction is recorded as a characteristic angle of the pixel point, and the gradient direction of each pixel point in the to-be-analyzed area is obtained, so that an angle corresponding to the gradient direction can be obtained.
Because the pits have the characteristic of concave converging light, the gray value of the image corresponding to the area with the pit defects from the edge outline of the area to the central pixel point of the area is gradually increased, and the gradient direction of the pixel point in the area is from the position of the pixel point to the center of the area. Meanwhile, the characteristic angle of the pixel point is the angle corresponding to the connection line of the pixel point corresponding to the maximum gray value and the pixel point corresponding to the maximum gray value, the connection line of the pixel point and the pixel point corresponding to the maximum gray value represents the straight line in which the direction of increasing the gray value in the region is located, and the difference between the characteristic angle of the pixel point in the region with the pit defect and the angle corresponding to the gradient direction of the pixel point is small.
Based on the method, the gray characteristic index of the region to be analyzed is obtained according to the gradient direction and the characteristic angle of the pixel point in the region to be analyzed, and the gray characteristic index is expressed by a formula as follows:
wherein U represents the gray scale characteristic index of the region to be analyzed,
representing the angle corresponding to the gradient direction of the ith pixel point in the region to be analyzed,
representing the characteristic angle of the ith pixel point in the region to be analyzed,
and representing the number of pixel points in the region to be analyzed.
Further, the probability that the area to be analyzed is the area with the pit defects is obtained by combining the shape characteristic and the gray scale change characteristic of the area to be analyzed, a first probability is obtained according to the ratio of the shape characteristic index to the gray scale characteristic index, namely, the ratio of the shape characteristic index to the gray scale characteristic index is the first probability, when the outline shape of the area to be analyzed is closer to a circle, and the change of the gray scale value in the area to be analyzed is more consistent with the change of the gray scale value in the pit when the pit is irradiated by light, the area to be analyzed is more likely to have the pit defects, namely, the first probability of the area to be analyzed is higher, and the area to be analyzed is more likely to be the area corresponding to the pit defects.
Calculating the gray value average value of pixel points in the area to be analyzed, and determining a shadow area according to the gray value average value and the position relation of the area to be analyzed; acquiring the position of a light source, and obtaining a second possibility according to angles corresponding to the connection lines of the central pixel point of the region to be analyzed and the light source and the central pixel point of the shadow region adjacent to the light source; the first significance is derived from the first and second likelihoods.
It should be noted that, because the light source is positioned at a certain angle with respect to the stainless steel product on the conveyor belt, shadow regions, such as protrusion defects and pit defects, exist at the positions where defects exist in the surface gray scale image of the stainless steel product. The gray value of the shadow area may affect the analysis of the gray variation characteristics in the area to be analyzed, so that the shadow characteristics having defects in the area to be analyzed need to be analyzed in combination with the shadow area in the area to be analyzed.
Meanwhile, in order to distinguish the convex defects and the concave defects, the gray value change of the shadow area in the outline of each pixel point in the image is calculated by combining the light and shadow characteristics. That is, according to the image characteristics of the convex defect and the concave defect, under the irradiation of the light source at the same angle, the convex part and the concave part form shadow areas, but the shadow area formed by the concave part is on the side close to the light source, and the shadow area formed by the convex part is on the side far from the light source, so that the concave defect in the area to be analyzed is judged based on the characteristics, and the interference of the convex defect is eliminated.
Specifically, a plurality of edge contours exist in the edge image, and some edge contours are included, and two edge contours that are included are obtained for analysis. Calculating the mean value of the gray values of the pixel points in each to-be-analyzed area, acquiring two to-be-analyzed areas containing the relation, and when the difference value of the mean values of the gray values of the two to-be-analyzed areas is larger than a gray threshold, marking the low mean value of the gray values in the two to-be-analyzed areas as a shadow area. In this embodiment, the value of the gray threshold is 30, and the implementer can set the threshold according to a specific implementation scenario.
According to the characteristics of the shadow area, when the gray value of a pixel point appearing in the edge contour in the image is lower, the area corresponding to the edge contour may be the shadow area. Therefore, the gray value mean values in the two to-be-analyzed areas which are in the inclusion relationship can be respectively calculated, when the difference between the gray value mean values of the two to-be-analyzed areas is large, it is indicated that one of the to-be-analyzed areas may be a pit defect or a bump defect, namely, the pit or the bump causes the shadow to generate the gray difference, and the area with the lower average gray value is marked as a shadow area. The positions corresponding to the two regions to be analyzed, which are in the inclusion relationship, are in the adjacent relationship, that is, the shadow region and the other region to be analyzed are adjacent.
For a single light source that is fixed and invariant, the position of the light source may be acquired first. And the position of the light source does not change when images are subsequently taken. And then the position information of the pixel points in each region to be analyzed based on the light source can be obtained. In this embodiment, a rectangular coordinate system is constructed by using the central pixel point of the surface gray image as an origin, and the coordinates of the position of the light source, and the coordinates of the central pixel point of the region to be analyzed and the central pixel point of the shadow region adjacent to the central pixel point are obtained.
If the area to be analyzed is an area corresponding to a pit defect, the shadow area adjacent to the area to be analyzed should be on a side close to the light source, and if the area to be analyzed is an area corresponding to a bump defect, the shadow area adjacent to the area to be analyzed should be on a side far from the light source. And calculating the included angle between the straight line where the central pixel point of the area to be analyzed is connected with the central pixel point of the adjacent shadow area and the horizontal direction according to the coordinates of the central pixel point of the area to be analyzed and the coordinates of the central pixel point of the adjacent shadow area, and calculating the included angle between the straight line where the central pixel point of the area to be analyzed is connected with the light source and the horizontal direction in the same manner. It should be noted that an included angle between a straight line where a connecting line between a central pixel point of the area to be analyzed and a central pixel point of the adjacent shadow area is located and the horizontal direction is specifically an included angle between a direction in which the central pixel point of the area to be analyzed points to the central pixel point of the adjacent shadow area and the horizontal right direction.
If the area to be analyzed is the area corresponding to the pit defect, the difference between the two included angle angles corresponding to the area to be analyzed and the adjacent shadow area is smaller, if the area to be analyzed is the area corresponding to the bump defect, the difference between the two included angle angles corresponding to the area to be analyzed and the adjacent shadow area is larger, based on this, the second possibility that the area to be analyzed is the area corresponding to the pit defect is calculated according to the difference between the two included angle angles, and is expressed by a formula:
wherein L is a second probability representing a probability that the area to be analyzed is an area corresponding to the pit defect,
representing the angle between the connecting line of the light source and the central pixel point of the area to be analyzed and the horizontal direction,
and the included angle between a connecting line between the central pixel point of the area to be analyzed and the adjacent shadow area and the horizontal direction is represented, and exp () represents an exponential function with a natural constant e as a base.
Representing the difference between the two included angle angles, the smaller the difference, the shadow zone being situated close to the lightOn the source side, the more likely the area to be analyzed is to be the area corresponding to a pit defect.
Further, a first significance is obtained according to a product of a first possibility and a second possibility of the area to be analyzed, namely the product of the first possibility and the second possibility of the area to be analyzed is the first significance, the larger the value of the first possibility is, the more the area to be analyzed meets the shape characteristics and the gray level change characteristics corresponding to the pit defects, the larger the second possibility is, the more the area to be analyzed meets the light and shadow characteristics of the area corresponding to the pit defects, and further the larger the first significance is, the more the area to be analyzed is likely to be the pit defects.
And step four, obtaining a second significance according to the texture information of the surface gray level image, obtaining a global significance according to the first and second significance, enhancing the surface gray level image by using the global significance, carrying out edge detection on the enhanced image to obtain a defect area, and further determining the quality of the stainless steel product according to the defect area.
Firstly, because the texture of the pixel point and the texture around the pixel point are greatly changed in the contour edge and the pit area, the pixel point of the area to be analyzed in the image can be endowed with a larger significant value according to the characteristic. Constructing a gray level co-occurrence matrix according to gray values of pixel points and neighborhood pixel points, and acquiring entropy values and energy values of the gray level co-occurrence matrix corresponding to the pixel points according to the gray level co-occurrence matrix; in the same way, a gray level co-occurrence matrix corresponding to the neighborhood pixel point is constructed, and the entropy value and the energy value of the gray level co-occurrence matrix corresponding to the neighborhood pixel point are obtained; calculating a second significance according to the entropy and energy values corresponding to the pixel points and the entropy and energy values corresponding to the neighborhood pixel points, and expressing the second significance by a formula as follows:
wherein the content of the first and second substances,
representing the texture characteristic index corresponding to the o-th pixel point,
and
respectively representing the energy value and entropy value of the gray level co-occurrence matrix corresponding to the o-th pixel point,
and
respectively representing the energy value and entropy value of the gray level co-occurrence matrix corresponding to the t-th neighborhood pixel point in the 8 th neighborhood of the o-th pixel point
The ASM energy value reflects the uniformity of the image gray level distribution and the texture thickness. If the element values of the gray level co-occurrence matrix are similar, the energy is smaller, and the texture is detailed; if some of the values are large, and others are small, the energy value is large; a large energy value indicates a more uniform and regularly varying texture pattern. The ENT entropy value represents a randomness measure of the amount of information contained in the image. When all values in the gray level co-occurrence matrix are equal or the pixel value shows the maximum randomness, the entropy value is maximum; therefore, the entropy value indicates the complexity of the image gray level distribution, and the larger the entropy value is, the more complex the image is.
Representing the difference in texture, the greater this difference, the calculated
The larger the value of (a), the more likely the o-th pixel is to be a pixel in the region corresponding to the pit defect.
It should be noted that, according to the image, the pits on the metal surface are dented from the original flat portion on the metal surface, and if the pits are spread out, the density of the pixels in the area should be greater than that of the pixels on the metal surface. The texture of the region with pits is coarser than that of other regions, so that the overall texture characteristics in the region to be analyzed can be analyzed according to the entropy value and the energy value of the gray level co-occurrence matrix corresponding to the pixel points in the region to be analyzed, and the texture characteristics are expressed by a formula:
wherein the content of the first and second substances,
the textural feature index of the region u to be analyzed is represented,
and
respectively representing the energy value and the entropy value of the gray level co-occurrence matrix corresponding to the region to be analyzed, and exp () representing an exponential function with a natural constant e as a base, wherein the gray level co-occurrence matrix can be constructed according to all pixel points in the region to be analyzed, so as to obtain the energy value and the entropy value of the gray level co-occurrence matrix corresponding to the region to be analyzed. That is, the larger the energy value and the entropy value corresponding to the region to be analyzed are, the larger the texture characteristic index value of the region to be analyzed is, which indicates that the texture in the region to be analyzed has a large change, and the more likely the region to be analyzed is the region corresponding to the pit defect.
And then obtaining a second significance according to the product of the sum of the textural feature indexes of all the pixel points in the area to be analyzed and the textural feature index of the area to be analyzed, namely the product of the sum of the textural feature indexes of all the pixel points in the area to be analyzed and the textural feature index of the area to be analyzed is the second significance, when the texture difference between the pixel points in the area to be analyzed and the surrounding pixel points is larger and the texture in the area to be analyzed is rougher, the greater the significance value of the area to be analyzed is, the greater the value of the second significance is.
Then, according to the regional characteristics corresponding to the image pit defects, the gray values of regional pixel points corresponding to the pit defects are different from the gray values of most pixel points on the surface of the stainless steel product, so that the significance of different regions to be analyzed in the image can be obtained by using the existing significance algorithm, and the image is enhanced according to the significance to achieve the purpose of highlighting the regions corresponding to the pit defects in the image.
In this embodiment, a color contrast algorithm based on a color global histogram, that is, an HC algorithm, is used, and the idea of the algorithm is that the greater the difference degree of color features between a pixel point and other pixel points, the higher the significance is. Therefore, the algorithm can be used for well representing the pixel points in the area corresponding to the pit defect in the image. And acquiring the significance of each pixel point in the image by using an HC algorithm, and recording the sum of the significances acquired by all the pixel points in the area to be analyzed through the HC algorithm as a third significance.
And obtaining the global significance of the region to be analyzed according to the average values of the first significance, the second significance and the third significance, wherein the average values of the first significance, the second significance and the third significance are the global significance of the region to be analyzed, and when the region to be analyzed is closer to the characteristics of the pit defects and the texture difference of the pixel points in the region to be analyzed is larger, the global significance is larger. The global significance of the region to be analyzed is obtained by considering the shape characteristics and the gray characteristics of the region to be analyzed, shadow and other interference factors.
If the calculated global significance is larger, enhancing the pixel points in the region to be analyzed; the smaller the calculated global saliency is, the more the suppression of the region to be analyzed is required. The enhancement or suppression of pixel points is specifically: acquiring the enhanced gray value of each pixel point in each to-be-analyzed area in the surface gray image according to the global significance and the gray value mean value of each to-be-analyzed area and the gray value of each pixel point in each to-be-analyzed area; obtaining an enhanced image according to the enhanced gray value of each pixel point, and expressing the enhanced image as follows by a formula:
wherein G represents the gray value of each pixel point in the region to be analyzed after being enhanced,
expressing the gray value of the image before each pixel point in the area to be analyzed is enhanced, R expresses the global significance of the area to be analyzed,
represents a significance threshold, which in this example is taken to be
The implementer can set the operation according to actual conditions.
When the global saliency of the region to be analyzed in the surface grayscale image is greater than the saliency threshold, i.e.
If the pixel points in the area to be analyzed are enhanced, the overall significance of the area to be analyzed and the gray value of the pixel points in the area to be analyzed are passed
And obtaining the gray value of the pixel point after enhancement. When the global saliency of the region to be analyzed in the surface grayscale image is less than or equal to the saliency threshold, i.e. when the global saliency is less than or equal to the saliency threshold
If the pixel points in the area to be analyzed are restrained, the global significance of the area to be analyzed and the gray value of the pixel points in the area to be analyzed are passed
And obtaining the gray value of the pixel after the pixel is enhanced, namely inhibiting the pixel.
When the enhanced gray value is calculated to be out of range, W =255 is taken. And then obtaining an enhanced image according to the enhanced gray value of each pixel point.
And finally, carrying out edge detection on the enhanced image, wherein the obtained closed edge contour area is an area corresponding to the pit defect and marked as a defect area, the areas and the number of all the defect areas in the enhanced image are obtained, and a quality evaluation index is obtained according to the areas and the number and is expressed by a formula as follows:
wherein, T is a quality evaluation index representing quality evaluation of a stainless steel product corresponding to the enhanced image, C is the number of all defect regions in the enhanced image, D is the sum of the areas of all defect regions in the enhanced image, and can be obtained by the sum of the pixel points of the regions corresponding to all pit defects in the enhanced image, and exp () represents an exponential function with a natural constant e as a base.
That is, when the number of the regions corresponding to the pit defects in the enhanced image is larger and the area is larger, it means that the stainless steel product corresponding to the image contains more pit defects, and the larger the area is, the smaller the corresponding quality evaluation index is, and the worse the surface quality of the stainless steel product is. An evaluation threshold is set, and the evaluation threshold is 0.1 in this embodiment, that is, when the quality evaluation index corresponding to the stainless steel product is smaller than the evaluation threshold, the stainless steel product is a poor product, and the quality of the stainless steel product is considered to be poor, and corresponding treatment needs to be performed on the poor product.
The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications or substitutions do not cause the essential features of the corresponding technical solutions to depart from the scope of the technical solutions of the embodiments of the present application, and are intended to be included within the scope of the present application.
Claims (7)
1. A stainless steel product quality detection method based on image recognition is characterized by comprising the following steps:
acquiring a surface gray image of the stainless steel product, carrying out edge detection on the image, and acquiring a region in a closed edge and marking as a region to be analyzed; obtaining a shape characteristic index according to the area of the region to be analyzed and the area of the minimum circumscribed circle of the region;
acquiring a connecting line of each pixel point in the area to be analyzed and the pixel point corresponding to the maximum gray value, recording an included angle between the connecting line corresponding to each pixel point and the horizontal direction as a characteristic angle, and acquiring a gray characteristic index according to the gradient direction and the characteristic angle of the pixel point in the area to be analyzed; obtaining a first possibility according to the ratio of the shape characteristic index and the gray characteristic index;
calculating the gray value mean value of pixel points in the area to be analyzed, and determining a shadow area according to the gray value mean value and the position relation of the area to be analyzed; acquiring the position of a light source, and obtaining a second possibility according to angles corresponding to the connection lines of the central pixel point of the region to be analyzed and the light source and the central pixel point of the shadow region adjacent to the light source; obtaining a first significance based on the first and second likelihoods;
and obtaining a second significance according to the texture information of the surface gray level image, obtaining a global significance according to the first and second significances, enhancing the surface gray level image by using the global significance, carrying out edge detection on the enhanced image to obtain a defect area, and further determining the quality of the stainless steel product according to the defect area.
2. The image recognition-based stainless steel product quality detection method according to claim 1, wherein the gray scale feature index is obtained by a specific method comprising the following steps:
wherein U represents the gray scale characteristic index of the region to be analyzed,
representing the angle corresponding to the gradient direction of the ith pixel point in the region to be analyzed,
representing the characteristic angle of the ith pixel point in the region to be analyzed,
and representing the number of pixel points in the region to be analyzed.
3. The stainless steel product quality detection method based on image recognition according to claim 1, wherein the determining of the shadow region according to the gray value mean value and the position relationship of the region to be analyzed specifically comprises:
and when the difference value of the mean values of the gray values of the two areas to be analyzed is greater than the gray threshold value, marking the area with the low mean value of the gray values in the two areas to be analyzed as a shadow area.
4. The image recognition-based stainless steel product quality detection method according to claim 1, wherein the obtaining of the position of the light source obtains a second possibility according to angles corresponding to the connection lines of the central pixel point of the region to be analyzed and the light source and the central pixel point of the shadow region adjacent to the light source, and comprises:
constructing a rectangular coordinate system by taking the central pixel point of the surface gray level image as an origin, and acquiring the coordinate of the position of the light source and the coordinates of the central pixel point of the area to be analyzed and the adjacent shadow area;
and calculating the included angle between the straight line where the connecting line of the central pixel point of the area to be analyzed and the adjacent shadow area is located and the horizontal direction according to the coordinates of the central pixel point of the area to be analyzed and the coordinates of the central pixel point of the adjacent shadow area, calculating the included angle between the straight line where the connecting line of the light source and the central pixel point of the area to be analyzed and the horizontal direction in a similar way, and obtaining a second possibility according to the difference value of the two included angles.
5. The stainless steel product quality detection method based on image recognition according to claim 1, wherein the obtaining of the second saliency according to the texture information of the surface gray level image is specifically:
constructing a gray level co-occurrence matrix according to the gray values of the pixel points and the neighborhood pixel points, and acquiring the entropy value and the energy value of the gray level co-occurrence matrix corresponding to the pixel points according to the gray level co-occurrence matrix; in the same way, a gray level co-occurrence matrix corresponding to the neighborhood pixel point is constructed, and the entropy value and the energy value of the gray level co-occurrence matrix corresponding to the neighborhood pixel point are obtained; and calculating the second significance according to the entropy and energy values corresponding to the pixel points and the entropy and energy values corresponding to the neighborhood pixel points.
6. The stainless steel product quality detection method based on image recognition according to claim 1, wherein the enhancing the surface gray scale image by using the global saliency is specifically as follows:
acquiring the enhanced gray value of each pixel point in each to-be-analyzed area in the surface gray image according to the global significance and the gray value mean value of each to-be-analyzed area and the gray value of each pixel point in each to-be-analyzed area; and acquiring an enhanced image according to the enhanced gray value of each pixel point.
7. The image recognition-based stainless steel product quality detection method according to claim 1, wherein the determining of the quality of the stainless steel product according to the defect area specifically comprises:
acquiring the areas and the number of all defect regions in the enhanced image, and obtaining a quality evaluation index according to the areas and the number; and setting an evaluation threshold, wherein when the quality evaluation index is less than the evaluation threshold, the corresponding stainless steel product is poor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211298717.XA CN115375676B (en) | 2022-10-24 | 2022-10-24 | Stainless steel product quality detection method based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211298717.XA CN115375676B (en) | 2022-10-24 | 2022-10-24 | Stainless steel product quality detection method based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115375676A true CN115375676A (en) | 2022-11-22 |
| CN115375676B CN115375676B (en) | 2023-01-24 |
Family
ID=84073122
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211298717.XA Active CN115375676B (en) | 2022-10-24 | 2022-10-24 | Stainless steel product quality detection method based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115375676B (en) |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115601369A (en) * | 2022-12-16 | 2023-01-13 | 国网山东省电力公司东营供电公司（Cn） | Quality evaluation method for power transformation equipment support for power transmission and transformation engineering |
| CN115775250A (en) * | 2023-02-13 | 2023-03-10 | 惠州威尔高电子有限公司 | Golden finger circuit board defect rapid detection system based on digital image analysis |
| CN115797353A (en) * | 2023-02-08 | 2023-03-14 | 山东乾钢金属科技有限公司 | Intelligent detection system and method for quality of cold-rolled strip steel |
| CN115841491A (en) * | 2023-02-24 | 2023-03-24 | 杭州电子科技大学 | Quality detection method of porous metal material |
| CN115984836A (en) * | 2023-03-20 | 2023-04-18 | 山东杨嘉汽车制造有限公司 | Tank opening identification and positioning method for railway tank wagon |
| 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 |
| CN116091499A (en) * | 2023-04-07 | 2023-05-09 | 山东中胜涂料有限公司 | Abnormal paint production identification system |
| CN116188468A (en) * | 2023-04-27 | 2023-05-30 | 济宁市长荣电器科技有限公司 | HDMI cable transmission letter sorting intelligent control system |
| CN116258713A (en) * | 2023-05-11 | 2023-06-13 | 青岛穗禾信达金属制品有限公司 | Welding processing detection method for metal cabinet |
| CN116309577A (en) * | 2023-05-19 | 2023-06-23 | 山东晨光胶带有限公司 | Intelligent detection method and system for high-strength conveyor belt materials |
| CN116503408A (en) * | 2023-06-28 | 2023-07-28 | 曲阜远大集团工程有限公司 | Scanning technology-based steel structure surface defect detection method |
| CN116645372A (en) * | 2023-07-27 | 2023-08-25 | 汉克威(山东)智能制造有限公司 | Intelligent detection method and system for appearance image of brake chamber |
| CN116758084A (en) * | 2023-08-21 | 2023-09-15 | 金恒山电气无锡有限公司 | Intelligent detection method for welding defects of sheet metal parts based on image data |
| CN116883688A (en) * | 2023-09-08 | 2023-10-13 | 泗水县锦川花生食品有限公司 | Intelligent picking method for fried peanuts |
| CN116958127A (en) * | 2023-09-18 | 2023-10-27 | 山东九思新材料科技有限责任公司 | Silicon wafer quality assessment method based on image features |
| CN117036341A (en) * | 2023-10-07 | 2023-11-10 | 青岛奥维特智能科技有限公司 | Pavement crack detection method based on image processing |
| CN117152160A (en) * | 2023-11-01 | 2023-12-01 | 湖南辰东科技有限公司 | Airport lamp detection method and system based on image processing |
| CN117237344A (en) * | 2023-11-13 | 2023-12-15 | 深圳市帝一通讯有限公司 | Machine vision-based method for detecting appearance quality of electronic atomizer chip |
| CN117408890A (en) * | 2023-12-14 | 2024-01-16 | 武汉泽塔云科技股份有限公司 | Video image transmission quality enhancement method and system |
| CN117422705A (en) * | 2023-11-24 | 2024-01-19 | 太康精密（中山）有限公司 | Connector terminal quality detection method based on image vision |
| CN117495863A (en) * | 2024-01-03 | 2024-02-02 | 深圳宝铭微电子有限公司 | Vision-assisted triode packaging quality detection method |
| CN117689655A (en) * | 2024-01-31 | 2024-03-12 | 东莞市恒兴隆实业有限公司 | Metal button surface defect detection method based on computer vision |
| CN117764986A (en) * | 2024-02-22 | 2024-03-26 | 宝鸡世邦钛制品有限公司 | titanium plate surface defect detection method based on image processing |
| CN117764986B (en) * | 2024-02-22 | 2024-05-07 | 宝鸡世邦钛制品有限公司 | Titanium plate surface defect detection method based on image processing |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010112846A (en) * | 2008-11-07 | 2010-05-20 | Jfe Steel Corp | Method and apparatus for adjusting surface inspecting device |
| JP2010223621A (en) * | 2009-03-19 | 2010-10-07 | Sumitomo Metal Ind Ltd | Inner surface inspection method of tubular article |
| CN103745234A (en) * | 2014-01-23 | 2014-04-23 | 东北大学 | Band steel surface defect feature extraction and classification method |
| CN104655060A (en) * | 2015-03-16 | 2015-05-27 | 上海理工大学 | Detection device for steel ball surface |
| CN106097360A (en) * | 2016-06-17 | 2016-11-09 | 中南大学 | A kind of strip steel surface defect identification method and device |
| CN108169236A (en) * | 2016-12-07 | 2018-06-15 | 广州映博智能科技有限公司 | A kind of cracks of metal surface detection method of view-based access control model |
| CN109785290A (en) * | 2018-12-19 | 2019-05-21 | 刘咏晨 | Normalized steel plate defect detection method is shone based on local light |
| CN110009633A (en) * | 2019-04-19 | 2019-07-12 | 湖南大学 | A kind of detection method of surface flaw of steel rail based on reversed difference of Gaussian |
| CN110766684A (en) * | 2019-10-30 | 2020-02-07 | 江南大学 | Stator surface defect detection system and detection method based on machine vision |
| CN112139857A (en) * | 2020-08-05 | 2020-12-29 | 沈阳东能科技有限公司 | Robot flexible grinding method for steel plate surface defects |
| CN114723701A (en) * | 2022-03-31 | 2022-07-08 | 南通博莹机械铸造有限公司 | Gear defect detection method and system based on computer vision |
| CN114926407A (en) * | 2022-04-28 | 2022-08-19 | 哈尔滨理工大学 | Steel surface defect detection system based on deep learning |
| CN114943739A (en) * | 2022-07-26 | 2022-08-26 | 山东三微新材料有限公司 | Aluminum pipe quality detection method |
| CN115049656A (en) * | 2022-08-15 | 2022-09-13 | 海门市刘氏铸造有限公司 | Method for identifying and classifying defects in silicon steel rolling process |
| CN115115642A (en) * | 2022-08-30 | 2022-09-27 | 启东万惠机械制造有限公司 | Strip steel scab defect detection method based on image processing |
| CN115131356A (en) * | 2022-09-01 | 2022-09-30 | 南通市恒瑞精密机械制造有限公司 | Steel plate defect classification method based on richness |
-
2022
- 2022-10-24 CN CN202211298717.XA patent/CN115375676B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010112846A (en) * | 2008-11-07 | 2010-05-20 | Jfe Steel Corp | Method and apparatus for adjusting surface inspecting device |
| JP2010223621A (en) * | 2009-03-19 | 2010-10-07 | Sumitomo Metal Ind Ltd | Inner surface inspection method of tubular article |
| CN103745234A (en) * | 2014-01-23 | 2014-04-23 | 东北大学 | Band steel surface defect feature extraction and classification method |
| CN104655060A (en) * | 2015-03-16 | 2015-05-27 | 上海理工大学 | Detection device for steel ball surface |
| CN106097360A (en) * | 2016-06-17 | 2016-11-09 | 中南大学 | A kind of strip steel surface defect identification method and device |
| CN108169236A (en) * | 2016-12-07 | 2018-06-15 | 广州映博智能科技有限公司 | A kind of cracks of metal surface detection method of view-based access control model |
| CN109785290A (en) * | 2018-12-19 | 2019-05-21 | 刘咏晨 | Normalized steel plate defect detection method is shone based on local light |
| CN110009633A (en) * | 2019-04-19 | 2019-07-12 | 湖南大学 | A kind of detection method of surface flaw of steel rail based on reversed difference of Gaussian |
| CN110766684A (en) * | 2019-10-30 | 2020-02-07 | 江南大学 | Stator surface defect detection system and detection method based on machine vision |
| CN112139857A (en) * | 2020-08-05 | 2020-12-29 | 沈阳东能科技有限公司 | Robot flexible grinding method for steel plate surface defects |
| CN114723701A (en) * | 2022-03-31 | 2022-07-08 | 南通博莹机械铸造有限公司 | Gear defect detection method and system based on computer vision |
| CN114926407A (en) * | 2022-04-28 | 2022-08-19 | 哈尔滨理工大学 | Steel surface defect detection system based on deep learning |
| CN114943739A (en) * | 2022-07-26 | 2022-08-26 | 山东三微新材料有限公司 | Aluminum pipe quality detection method |
| CN115049656A (en) * | 2022-08-15 | 2022-09-13 | 海门市刘氏铸造有限公司 | Method for identifying and classifying defects in silicon steel rolling process |
| CN115115642A (en) * | 2022-08-30 | 2022-09-27 | 启东万惠机械制造有限公司 | Strip steel scab defect detection method based on image processing |
| CN115131356A (en) * | 2022-09-01 | 2022-09-30 | 南通市恒瑞精密机械制造有限公司 | Steel plate defect classification method based on richness |
Non-Patent Citations (4)
| Title |
|---|
| QIWU LUO ET AL: "Automated Visual Defect Classification for Flat Steel Surface: A Survey", 《IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT》 * |
| WU-BIN LI ET AL: "A local annular contrast based real-time inspection algorithm for steel bar surface defects", 《APPLIED SURFACE SCIENCE》 * |
| 冯超 等: "基于空间全角度光源基于空间全角度光源的钢球表面微缺陷检测的钢球表面微缺陷检测", 《激光与光电子学进展》 * |
| 李维创 等: "工业金属板带材表面缺陷自动视觉检测研究进展", 《电子测量与仪器学报》 * |
Cited By (38)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115601369A (en) * | 2022-12-16 | 2023-01-13 | 国网山东省电力公司东营供电公司（Cn） | Quality evaluation method for power transformation equipment support for power transmission and transformation engineering |
| CN115797353A (en) * | 2023-02-08 | 2023-03-14 | 山东乾钢金属科技有限公司 | Intelligent detection system and method for quality of cold-rolled strip steel |
| CN115775250A (en) * | 2023-02-13 | 2023-03-10 | 惠州威尔高电子有限公司 | Golden finger circuit board defect rapid detection system based on digital image analysis |
| CN115841491A (en) * | 2023-02-24 | 2023-03-24 | 杭州电子科技大学 | Quality detection method of porous metal material |
| CN115984836B (en) * | 2023-03-20 | 2023-06-30 | 山东杨嘉汽车制造有限公司 | Tank opening identification positioning method for railway tank truck |
| CN115984836A (en) * | 2023-03-20 | 2023-04-18 | 山东杨嘉汽车制造有限公司 | Tank opening identification and positioning method for railway tank wagon |
| CN116030058B (en) * | 2023-03-29 | 2023-06-06 | 无锡斯达新能源科技股份有限公司 | Quality evaluation method for surface roughness of polishing pad |
| 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 |
| CN116091499A (en) * | 2023-04-07 | 2023-05-09 | 山东中胜涂料有限公司 | Abnormal paint production identification system |
| CN116188468A (en) * | 2023-04-27 | 2023-05-30 | 济宁市长荣电器科技有限公司 | HDMI cable transmission letter sorting intelligent control system |
| CN116258713A (en) * | 2023-05-11 | 2023-06-13 | 青岛穗禾信达金属制品有限公司 | Welding processing detection method for metal cabinet |
| CN116309577A (en) * | 2023-05-19 | 2023-06-23 | 山东晨光胶带有限公司 | Intelligent detection method and system for high-strength conveyor belt materials |
| CN116309577B (en) * | 2023-05-19 | 2023-08-04 | 山东晨光胶带有限公司 | Intelligent detection method and system for high-strength conveyor belt materials |
| CN116503408A (en) * | 2023-06-28 | 2023-07-28 | 曲阜远大集团工程有限公司 | Scanning technology-based steel structure surface defect detection method |
| CN116503408B (en) * | 2023-06-28 | 2023-08-25 | 曲阜远大集团工程有限公司 | Scanning technology-based steel structure surface defect detection method |
| CN116645372B (en) * | 2023-07-27 | 2023-10-10 | 汉克威(山东)智能制造有限公司 | Intelligent detection method and system for appearance image of brake chamber |
| CN116645372A (en) * | 2023-07-27 | 2023-08-25 | 汉克威(山东)智能制造有限公司 | Intelligent detection method and system for appearance image of brake chamber |
| CN116758084B (en) * | 2023-08-21 | 2023-10-13 | 金恒山电气无锡有限公司 | Intelligent detection method for welding defects of sheet metal parts based on image data |
| CN116758084A (en) * | 2023-08-21 | 2023-09-15 | 金恒山电气无锡有限公司 | Intelligent detection method for welding defects of sheet metal parts based on image data |
| CN116883688A (en) * | 2023-09-08 | 2023-10-13 | 泗水县锦川花生食品有限公司 | Intelligent picking method for fried peanuts |
| CN116883688B (en) * | 2023-09-08 | 2023-12-01 | 泗水县锦川花生食品有限公司 | Intelligent picking method for fried peanuts |
| CN116958127A (en) * | 2023-09-18 | 2023-10-27 | 山东九思新材料科技有限责任公司 | Silicon wafer quality assessment method based on image features |
| CN116958127B (en) * | 2023-09-18 | 2023-12-08 | 山东九思新材料科技有限责任公司 | Silicon wafer quality assessment method based on image features |
| CN117036341A (en) * | 2023-10-07 | 2023-11-10 | 青岛奥维特智能科技有限公司 | Pavement crack detection method based on image processing |
| CN117152160A (en) * | 2023-11-01 | 2023-12-01 | 湖南辰东科技有限公司 | Airport lamp detection method and system based on image processing |
| CN117152160B (en) * | 2023-11-01 | 2024-01-26 | 湖南辰东科技有限公司 | Airport lamp detection method and system based on image processing |
| CN117237344A (en) * | 2023-11-13 | 2023-12-15 | 深圳市帝一通讯有限公司 | Machine vision-based method for detecting appearance quality of electronic atomizer chip |
| CN117237344B (en) * | 2023-11-13 | 2024-01-19 | 深圳市帝一通讯有限公司 | Machine vision-based method for detecting appearance quality of electronic atomizer chip |
| CN117422705A (en) * | 2023-11-24 | 2024-01-19 | 太康精密（中山）有限公司 | Connector terminal quality detection method based on image vision |
| CN117422705B (en) * | 2023-11-24 | 2024-04-05 | 太康精密（中山）有限公司 | Connector terminal quality detection method based on image vision |
| CN117408890A (en) * | 2023-12-14 | 2024-01-16 | 武汉泽塔云科技股份有限公司 | Video image transmission quality enhancement method and system |
| CN117408890B (en) * | 2023-12-14 | 2024-03-08 | 武汉泽塔云科技股份有限公司 | Video image transmission quality enhancement method and system |
| CN117495863A (en) * | 2024-01-03 | 2024-02-02 | 深圳宝铭微电子有限公司 | Vision-assisted triode packaging quality detection method |
| CN117495863B (en) * | 2024-01-03 | 2024-04-02 | 深圳宝铭微电子有限公司 | Vision-assisted triode packaging quality detection method |
| CN117689655A (en) * | 2024-01-31 | 2024-03-12 | 东莞市恒兴隆实业有限公司 | Metal button surface defect detection method based on computer vision |
| CN117689655B (en) * | 2024-01-31 | 2024-04-19 | 东莞市恒兴隆实业有限公司 | Metal button surface defect detection method based on computer vision |
| CN117764986A (en) * | 2024-02-22 | 2024-03-26 | 宝鸡世邦钛制品有限公司 | titanium plate surface defect detection method based on image processing |
| CN117764986B (en) * | 2024-02-22 | 2024-05-07 | 宝鸡世邦钛制品有限公司 | Titanium plate surface defect detection method based on image processing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115375676B (en) | 2023-01-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115375676B (en) | Stainless steel product quality detection method based on image recognition | |
| CN111310558B (en) | Intelligent pavement disease extraction method based on deep learning and image processing method | |
| CN108629775B (en) | Thermal state high-speed wire rod surface image processing method | |
| CN109682839B (en) | Online detection method for surface defects of metal arc-shaped workpiece | |
| CN111179225B (en) | Test paper surface texture defect detection method based on gray gradient clustering | |
| CN102156996B (en) | Image edge detection method | |
| CN115222741A (en) | Cable surface defect detection method | |
| Aminzadeh et al. | Automatic thresholding for defect detection by background histogram mode extents | |
| US8520892B2 (en) | Method and apparatus for detecting objects | |
| CN115222733A (en) | Metal component surface defect detection method based on image recognition | |
| CN107490582B (en) | Assembly line workpiece detection system | |
| Wakaf et al. | Defect detection based on extreme edge of defective region histogram | |
| CN110717888A (en) | Automatic identification method for intravascular Optical Coherence Tomography (OCT) vessel wall inner contour | |
| CN116777916B (en) | Defect detection method based on metal shell of pump machine | |
| CN110070523B (en) | Foreign matter detection method for bottle bottom | |
| CN115063430B (en) | Electric pipeline crack detection method based on image processing | |
| Vorobel et al. | Segmentation of rust defects on painted steel surfaces by intelligent image analysis | |
| CN109658391B (en) | Circle radius measuring method based on contour merging and convex hull fitting | |
| CN115131359B (en) | Method for detecting pitting defects on surface of metal workpiece | |
| CN116883408B (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
| CN115115642A (en) | Strip steel scab defect detection method based on image processing | |
| CN115100191A (en) | Metal casting defect identification method based on industrial detection | |
| CN114913177A (en) | Automobile part defect detection method based on Hough circle | |
| CN114549441A (en) | Sucker defect detection method based on image processing | |
| CN115880280A (en) | Detection method for quality of steel structure weld joint |
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 |