CN109325930B - Boundary defect detection method, device and detection equipment - Google Patents
Boundary defect detection method, device and detection equipment Download PDFInfo
- Publication number
- CN109325930B CN109325930B CN201811062811.9A CN201811062811A CN109325930B CN 109325930 B CN109325930 B CN 109325930B CN 201811062811 A CN201811062811 A CN 201811062811A CN 109325930 B CN109325930 B CN 109325930B
- Authority
- CN
- China
- Prior art keywords
- boundary
- image
- detected
- line
- defect
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
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
-
- 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
- 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
Abstract
The invention provides a method, a device and equipment for detecting boundary defects, wherein the method comprises the steps of obtaining an image to be detected; extracting the boundary area of two adjacent image areas to obtain a boundary area image; linearly unfolding the boundary area image to form a linearly unfolded image; detecting boundary defects based on the linear expansion image; when the linear expansion image is detected to have boundary defects, the position of the boundary defects in the image to be detected is determined by using the boundary area image and the linear expansion image. The first preset distance and the second preset distance are respectively extended to the two adjacent image areas along the normal direction of each pixel point on the boundary line to obtain a first boundary connection line and a second boundary connection line, namely, all boundary defects of the image to be detected are divided into the boundary areas by setting a proper boundary detection range, and then after the boundary areas are linearly expanded, the defects crossing the boundaries of the areas can be completely detected, so that the detection accuracy is improved.
Description
Technical Field
The invention relates to the field of defect detection, in particular to a method, a device and equipment for detecting boundary defects.
Background
Because the surface defects of industrial products can bring adverse effects to the attractiveness, comfort, usability and the like of the products, the surface defects of the products can be detected by production enterprises before the products leave a factory so as to be timely discovered and controlled. Among them, the most common surface defects are boundary defects, which are classified into scratches, edge breakages, foreign substances, stains, and the like according to the types of defects.
The traditional defect detection mostly depends on manual detection, not only is the efficiency low, but also false detection or missing detection can be caused by eye fatigue or subjective reasons. Therefore, with the rapid development of digital image technology, machine vision-based defect detection methods are increasingly widely researched and applied in modern industries due to their high accuracy and high efficiency.
The detection method of the boundary defect commonly adopted in the prior art is a method based on template image comparison, and specifically comprises the following steps: firstly, an image without defects is obtained as a template image, then the relative position relation between the template image and a detection image is calculated through an image alignment algorithm, the difference between the detection image and the template image is calculated to obtain a difference image, and the defects are searched in the difference image.
However, in the process of researching the boundary defect detection method, the inventor finds that the existing defect detection method has poor robustness due to uncertainty of industrial environment, diversity of products and complexity of process, so that the accuracy of the existing defect detection method is low.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method, an apparatus, and a device for detecting a boundary defect, so as to solve the problem of low accuracy of the existing defect detection method.
Therefore, the embodiment of the invention provides the following technical scheme:
the first aspect of the present invention provides a method for detecting a boundary defect, including:
acquiring an image to be detected; the image to be detected comprises at least two image areas;
extracting boundary areas of two adjacent image areas to obtain boundary area images;
linearly unfolding the boundary area image to form a linearly unfolded image;
detecting the boundary defects based on the linear expansion image;
and when the linear expanded image is detected to have the boundary defect, determining the position of the boundary defect in the image to be detected by using the boundary area image and the linear expanded image.
According to the method for detecting the boundary defect, provided by the embodiment of the invention, the boundary areas of the two adjacent image areas are linearly expanded, namely, the irregular boundary area image is converted into the regular boundary area image, so that the robustness of the detection method can be improved, and the accuracy of the detection method is further improved; in addition, the defect detection of the boundary region can realize the detection of the defect close to or on the region boundary, and can also realize the complete detection of the defect across the region boundary, thereby having higher detection accuracy.
With reference to the first aspect, in a first implementation manner of the first aspect, the extracting a boundary region of two adjacent image regions to obtain a boundary region image includes:
acquiring a boundary line of two adjacent image areas and a normal and coordinates of each pixel point on the boundary line; wherein the coordinates are uniquely determined by a first reference coordinate system;
respectively shifting the boundary line to two adjacent image areas by a first preset distance and a second preset distance along the normal direction of each pixel point on the boundary line to obtain a first boundary connecting line and a second boundary connecting line;
and extracting a region between the first boundary connecting line and the second boundary connecting line to obtain the boundary region image.
According to the boundary defect detection method provided by the embodiment of the invention, the first preset distance and the second preset distance are respectively extended to the two adjacent image areas along the normal direction of each pixel point on the boundary line to obtain the first boundary connecting line and the second boundary connecting line, namely, all boundary defects of the image to be detected are divided into the boundary areas by setting a proper boundary detection range, and then after the boundary areas are linearly expanded, the defects crossing the area boundaries can be completely detected, so that the detection accuracy is improved.
With reference to the first embodiment of the first aspect, in a second embodiment of the first aspect, the linearly expanding the boundary area image to form a linearly expanded image includes:
sequentially mapping the boundary points to a second reference coordinate system along a first straight line by taking preset pixel points on the boundary line as starting points; the boundary points are pixel points at the intersection points of the normal lines and the boundary lines, and the normal lines and the boundary points are in one-to-one correspondence;
sequentially calculating the distance between each pixel point on the normal line and the corresponding boundary point;
sequentially mapping all pixel points on the boundary area image to the second reference coordinate system by using the distance so as to form the linear expansion image; and pixel points in the boundary region image correspond to pixel points in the linear expansion image one to one.
According to the boundary defect detection method provided by the embodiment of the invention, all pixel points on the boundary area image are mapped to the second reference coordinate system by utilizing the normal line, and the normal line is vertical to the tangent line of the corresponding pixel point, so that the accuracy is higher by adopting the normal line as the basis of linear expansion.
With reference to the second implementation manner of the first aspect, in a third implementation manner of the first aspect, the sequentially mapping all the pixel points on the boundary area image onto the second reference coordinate system by using the distance includes:
sequentially extracting pixel points on the normal;
inquiring a mapping boundary point corresponding to the normal; wherein the mapping boundary point is a mapping point of the boundary point corresponding to the normal on the second reference coordinate system;
mapping the extracted pixel points to the second reference coordinate system along a second straight line; the extracted pixel point mapping point and the extracted pixel point mapping boundary point are separated by the distance, the second straight line is perpendicular to the first straight line, and the second straight line corresponds to the mapping boundary point one to one.
With reference to the first embodiment of the first aspect, in a fourth embodiment of the first aspect, the performing the detection of the boundary defect based on the linearly expanded image includes:
filtering the linear expansion image to form a filtered image;
comparing the linear expansion image with the filtering image to obtain a difference image;
screening out areas meeting preset conditions in the difference image; wherein the screened area is the area of the detected boundary defect.
In the method for detecting a boundary defect provided in the embodiment of the present invention, in the difference image, since the pixels corresponding to the boundary defect have a large change with respect to the background, the corresponding difference value is large, and the region satisfying the condition screened by the preset condition (for example, the preset defect area, the preset length threshold, and the like) is the detected defect; the defect meeting the condition is determined by utilizing the obvious difference value, and the efficiency of the detection method can be improved.
With reference to the first embodiment of the first aspect, in a fifth embodiment of the first aspect, the determining the position of the boundary defect in the image to be detected by using the boundary area image and the straight line expansion image includes:
extracting coordinates of all pixel points in the area of the boundary defect on a second reference coordinate system;
and matching the extracted coordinates with coordinates on a first reference coordinate system to determine the position of the boundary defect in the image to be detected.
The method for detecting the boundary defect provided by the embodiment of the invention can quickly determine the position of the boundary defect in the image to be detected by using a coordinate matching method, and has higher detection efficiency and detection accuracy.
With reference to the first implementation manner of the first aspect, in a sixth implementation manner of the first aspect, the acquiring a boundary line between two adjacent image areas includes:
acquiring a defect-free template image; the non-defective template images correspond to the images to be detected one by one;
extracting the boundary line of two adjacent image areas of the non-defective template image;
aligning the non-defective image and the image to be detected;
mapping the extracted boundary line to the image to be detected; and extracting the boundary line, wherein the extracted boundary line is the boundary line of two adjacent image areas of the image to be detected.
According to the boundary defect detection method provided by the embodiment of the invention, the boundary line of the non-defective template image is mapped to the boundary line of the image to be detected, so that the problem that the subsequent detection accuracy is influenced due to inaccurate division of the boundary line caused by the boundary defect of the image to be detected can be avoided.
According to a second aspect, an embodiment of the present invention further provides an apparatus for detecting a boundary defect, including:
the acquisition module is used for acquiring an image to be detected; the image to be detected comprises at least two image areas;
the extraction module is used for extracting the boundary areas of two adjacent image areas to obtain a boundary area image;
the linear expansion module is used for linearly expanding the boundary area image to form a linearly expanded image;
the detection module is used for detecting the boundary defects based on the linear expansion image;
and the boundary defect determining module is used for determining the position of the boundary defect in the image to be detected by using the boundary area image and the linear expansion image when the linear expansion image is detected to have the boundary defect.
According to the boundary defect detection device provided by the embodiment of the invention, the boundary areas of two adjacent image areas are linearly expanded, namely, an irregular boundary area image is converted into a regular boundary area image, so that the robustness of the detection method can be improved, and the accuracy of the detection method is further improved; in addition, the defect detection of the boundary region can realize the detection of the defect close to or on the region boundary, and can also realize the complete detection of the defect across the region boundary, thereby having higher detection accuracy.
According to a third aspect, an embodiment of the present invention further provides a detection apparatus, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the method for detecting a boundary defect as described in the first aspect or any of the embodiments of the first aspect.
According to a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which computer instructions are stored, where the instructions are executed by a processor to perform the steps of the boundary defect detection method described in the first aspect or any one of the embodiments of the first aspect.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are 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 a method of detecting boundary defects according to an embodiment of the present invention;
FIG. 2 is a schematic illustration of an image to be detected according to an embodiment of the invention;
FIG. 3 is a schematic illustration of a border area image according to an embodiment of the invention;
FIG. 4 is a flow chart of a method of detecting boundary defects according to an embodiment of the present invention;
FIG. 5 is a schematic illustration of a border area image according to an embodiment of the invention;
FIG. 6 is a schematic diagram of a linearly unwrapped image in accordance with an embodiment of the present invention;
FIGS. 7 a-7 c are schematic diagrams of the formation of a difference image according to an embodiment of the invention;
FIG. 8 is a flowchart of a method for obtaining a boundary line between two adjacent image regions according to an embodiment of the present invention;
FIG. 9 is a flow chart of a method of detecting boundary defects according to an embodiment of the present invention;
FIG. 10 is a block diagram of an apparatus for detecting boundary defects according to an embodiment of the present invention;
FIG. 11 is a block diagram of an apparatus for detecting boundary defects according to an embodiment of the present invention;
fig. 12 is a schematic hardware structure diagram of a detection device according to an embodiment of the present invention;
reference numerals:
10-boundary line; 11-a first boundary line; 12-a second boundary line; 13-normal;
10' -mapping the boundary line; 11' -mapping the first boundary line; 12' -mapping the second boundary line; 13' -mapping the normal;
a1, a2, A3, a4, a 5-boundary points; a6-pixel point;
a1 ', A2 ', A3 ', A4 ', A5 ' -mapping boundary points; a6' -mapping pixel points;
b1, B2-defects;
b1 ', B2' -map defects.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In accordance with an embodiment of the present invention, there is provided an embodiment of a method for detecting boundary defects, it is noted that the steps illustrated in the flowchart of the drawings may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in an order different than that described herein.
In this embodiment, a method for detecting a boundary defect is provided, which can be used in a detection apparatus, and fig. 1 is a flowchart of boundary defect detection according to an embodiment of the present invention, as shown in fig. 1, the flowchart includes the following steps:
and S11, acquiring an image to be detected.
Wherein the image to be detected comprises at least two image areas.
The detection device may acquire the image to be detected through a variety of ways, for example, the image to be detected may be captured by a CCD or CMOS camera, or may be a result of preprocessing other images, and it is only necessary to ensure that the image to be detected acquired by the detection device includes at least two image areas.
The so-called image area, that is, the value of the pixel point in the area is clearly distinguished from the pixel point outside the area, for example, the image area in the acquired image to be detected is divided into: the image of the product to be detected and the background image of the product to be detected. For example, as shown in fig. 2, the image to be detected includes two image regions, i.e., a region 1 and a region 2, where the region 1 is an image of a product to be detected, the region 2 is a background image, and the region 1 and the region 2 are divided by a boundary line.
And S12, extracting the boundary area of two adjacent image areas to obtain a boundary area image.
After the detection device acquires the image to be detected, the boundary areas of two adjacent image areas are sequentially extracted to obtain a boundary area image. The boundary region is an image formed by two adjacent regions.
For example, as shown in fig. 3, the area filled in fig. 3 may be formed by shifting a distance into the area 1 and a distance into the area 2 with reference to the boundary line, that is, the boundary area image. In addition, the boundary region image may be extracted from the region 1 and from the region 2, and it is only necessary to ensure that the extracted portions from the two regions are adjacent to each other.
Alternatively, only a part of the boundary region may be extracted, for example, it is known in advance that there is a defect in a partial range of the boundary, and when the boundary region is extracted, only the boundary region having the defect needs to be extracted, and the entire boundary region does not need to be extracted.
S13, the boundary area image is linearly expanded to form a linearly expanded image.
The detection device expands the boundary area image along a straight line after extracting the boundary area image, and converts the irregular boundary area image into a regular straight line expanded image. Specifically, a certain pixel point in the image of the boundary area can be used as a starting point for expansion, then adjacent pixel points of the certain pixel point are sequentially extracted and placed along a straight line according to the position relationship between the two pixel points; or all pixel points on the boundary line are used as starting points for expansion, and then the normal line corresponding to the pixel points on the boundary line and all the pixel points on the normal line are used for sequentially realizing rearrangement of all the pixel points in the boundary area image so as to form a linear expansion image; or, the method can be realized in other modes, and only the detection equipment is required to convert the irregular boundary area image into a regular linear expansion image.
S14, boundary defects are detected based on the linear expansion image.
After the detection equipment forms the linear expansion image, the detection of the boundary defect can be carried out on the linear expansion image; for example, a non-defective image corresponding to the image to be detected can be adopted, a matching image of the non-defective image is extracted according to the steps, and the linear expansion image is compared with the matching image, so that the boundary defect on the linear expansion image can be detected; filtering the linear expansion image, and detecting the boundary defect by using a difference image of the linear expansion image and the filtered image; or detecting boundary defects in other ways, etc.
And S15, when the linear expansion image is detected to have boundary defects, determining the position of the boundary defects in the image to be detected by using the boundary area image and the linear expansion image.
When the detection equipment detects that the boundary defect exists in the linear expansion image, the boundary defect in the boundary area image is positioned in the linear expansion image according to the operation opposite to the operation of S13, so that the position of the boundary defect in the image to be detected can be determined, and the detection of the boundary defect of the image to be detected is realized.
According to the method for detecting the boundary defect, provided by the embodiment of the invention, the boundary areas of the two adjacent image areas are linearly expanded, namely, the irregular boundary area image is converted into the regular boundary area image, so that the robustness of the detection method can be improved, and the accuracy of the detection method is further improved; in addition, the defect detection of the boundary region can realize the detection of the defect close to or on the region boundary, and can also realize the complete detection of the defect across the region boundary, thereby having higher detection accuracy.
It should be noted that, in the embodiment of the present invention, the image to be detected includes not only two regions, but also 3 regions, or a plurality of regions, and the boundary defect detection method is adopted for each two adjacent regions to detect the boundary defects, that is, all the boundary defects in the image to be detected can be detected.
An embodiment of the present invention further provides a method for detecting a boundary defect, as shown in fig. 4, the method includes:
and S21, acquiring an image to be detected.
Wherein the image to be detected comprises at least two image areas. Please refer to S11 in fig. 1, which is not described herein again.
And S22, extracting the boundary area of two adjacent image areas to obtain a boundary area image.
The method specifically comprises the following steps of for extracting the boundary area of two adjacent image areas:
s221, a boundary line of two adjacent image areas and a normal line and coordinates of each pixel point on the boundary line are obtained.
The image to be detected is composed of a plurality of pixel points. The detection device extracts the boundary line of two adjacent image regions and the normal of each pixel point on the boundary line by using an edge detection operator (for example, a Canny edge detection operator or other edge detection operators can be adopted).
In addition, the coordinate of each pixel point is uniquely determined through the first reference coordinate system, and the coordinate is used for expressing the relative position relation between the pixel points in the image to be detected. Specifically, the origin of the first reference coordinate system may be a central point of the image to be detected; or, when the image to be detected is rectangular, the origin of the first reference coordinate system may be a pixel point at the upper left corner or the lower right corner of the image to be detected, or any pixel point in the image to be detected, and the like; only the first reference coordinate system is required to be used for uniquely determining the coordinates of each pixel point on the boundary line. The coordinate of each pixel point on the boundary line can be determined through the first reference coordinate system, and the coordinate of each pixel point on the image to be detected can also be determined.
S222, shifting the boundary line to two adjacent image regions by a first preset distance and a second preset distance along a normal direction of each pixel point on the boundary line, so as to obtain a first boundary connection line and a second boundary connection line.
Specifically, as shown in fig. 5, when the approximate position of the defect is known in advance, the edge detection operator is used to extract the boundary line 10 between two adjacent image regions, and the boundary line 10 is partially extended by a first preset distance L1 and a second preset distance L2 toward the two adjacent image regions, respectively. The specific distances of L1 and L2 may be specifically set according to actual conditions, and it is only necessary to ensure that L1 is greater than the farthest distances of defect B1 and defect B2 to boundary line 10 in region 1, and that L2 is greater than the farthest distance of defect B2 to boundary line 10 in region 2. When the L1 is larger than the farthest distance from the defect B1 to the boundary line 10 in the region 1, the defect close to the boundary region can be effectively detected after the linear expansion; when L2 is greater than the farthest distance of defect B2 to the boundary line 10 within zone 2, the defect crossing the boundary zone can be completely detected subsequently after the straight line deployment.
When the defect in the image to be detected is not known in advance, the boundary line 10 may be shifted to the area 1 by a first preset distance and to the area 2 by a second preset distance as a whole, as shown in fig. 3.
S223, extracting the area between the first boundary connecting line and the second boundary connecting line to obtain a boundary area image.
As shown in fig. 5, the closed region formed by the first boundary connecting lines, the second boundary connecting lines, and the normals corresponding to the boundary point a1 and the boundary point a5 is a boundary region image, and the detecting device extracts the boundary region image, so as to perform boundary defect detection by using the boundary region image.
S23, the boundary area image is linearly expanded to form a linearly expanded image.
Specifically, a linearly expanded image formed after linearly expanding the boundary area image is shown in fig. 6. Wherein, corresponding to fig. 5, the mapping boundary line 10' corresponds to the boundary line 10; mapping the first boundary line 11' to correspond to the first boundary line 11; mapping a second boundary line 12' to correspond to the second boundary line 12; mapping normal 13' to correspond with normal 13; mapping defects B1 'and B2' to defects B1 and B2; mapping boundary points A1 '-A5' to correspond to boundary points A1-A5; the mapped pixel A6' corresponds to pixel A6.
For the rest, please refer to S13 in the embodiment shown in fig. 1, which is not described herein again.
S24, boundary defects are detected based on the linear expansion image.
The detection device can detect the boundary defect by filtering and binarizing the linearly expanded image after the linearly expanded image is formed. The method specifically comprises the following steps:
s241, filter the linearly expanded image to form a filtered image.
Fig. 7a shows a linearly expanded image, which is filtered (e.g., block, mean, or gaussian, etc. filtering may be used), resulting in a filtered image as shown in fig. 7 b.
And S242, comparing the linear expansion image with the filtering image to obtain a difference image.
The detection means compares the line-expanded image with the filtered image to obtain a difference image as shown in fig. 7 c.
S243, screening out the areas which meet the preset conditions in the difference image.
As shown in fig. 7c, in the difference image, since the pixels corresponding to the defect have a large variation with respect to the background and the difference value is large, after the difference image is subjected to binarization segmentation, the regions satisfying the condition are screened out by using a preset condition, wherein the screened out regions are the regions of the detected boundary defects. The defect meeting the condition is determined by utilizing the obvious difference value, and the efficiency of the detection method can be improved.
Specifically, it may be implemented to set an area threshold and/or a length threshold of the defect, and when there is a region satisfying any one of the preset conditions in the difference image, the region satisfying the preset condition is a region of the detected boundary defect.
And S25, when the linear expansion image is detected to have boundary defects, determining the position of the boundary defects in the image to be detected by using the boundary area image and the linear expansion image. Please refer to S15 in fig. 1, which is not described herein again.
Compared with the embodiment shown in fig. 1, the method for detecting boundary defects provided in this embodiment extends the first preset distance and the second preset distance to the two adjacent image regions respectively along the normal direction of each pixel point on the boundary line to obtain the first boundary connecting line and the second boundary connecting line, that is, all boundary defects of the image to be detected are divided into the boundary regions by setting a proper boundary detection range, and then after the boundary regions are linearly expanded, the defects crossing the boundaries of the regions can be completely detected, thereby improving the detection accuracy.
As an optional implementation manner of this embodiment, the first boundary connecting line and the second boundary connecting line may be any straight lines, and are not limited to the deviation of the boundary line, and it is only required to ensure that the area between the first boundary connecting line and the second boundary connecting line can cover all defects in the image to be detected.
As an alternative embodiment of the present embodiment, as shown in fig. 8, in the step of acquiring the boundary line between two adjacent image areas in S221, the following steps are included for the image to be detected formed by the irregularly-shaped product:
s41, acquiring a defect-free template image.
Wherein, the non-defective template image corresponds to the image to be detected one by one.
S42, the boundary line between two adjacent image regions of the defect-free template image is extracted.
And S43, aligning the non-defective image and the image to be detected.
Generally, for mechanical or self reasons, the image to be detected formed by the product to be detected may have certain distortions such as rotation, scaling, offset and the like relative to a non-defective template image, and a template matching method based on a contour may be adopted for alignment so as to reduce the influence of the distortion.
And S44, mapping the extracted boundary line to the image to be detected.
The extracted boundary line is the boundary line of two adjacent image areas of the image to be detected.
The method utilizes the boundary line mapping of the non-defective template image as the boundary line of the image to be detected, and can avoid the inaccurate division of the boundary line caused by the boundary defect of the image to be detected, thereby influencing the subsequent detection accuracy.
As another alternative to this embodiment, for the image to be detected formed by a product with a regular shape, for example, the boundary of two adjacent image regions in the image to be detected is a straight boundary, a circular boundary, an elliptical boundary, or the like, the boundary line of the image to be detected may be directly extracted, and then the coordinates and the normal of the boundary point may be obtained by fitting the boundary region.
An embodiment of the present invention further provides a method for detecting a boundary defect, as shown in fig. 9, the method includes:
and S51, acquiring an image to be detected.
Wherein the image to be detected comprises at least two image areas. Please refer to S21 in fig. 4 for details, which are not described herein.
And S52, extracting the boundary area of two adjacent image areas to obtain a boundary area image. Please refer to S22 in fig. 4 for details, which are not described herein.
S53, the boundary area image is linearly expanded to form a linearly expanded image.
The boundary region image is shown in fig. 5, and the linearly expanded image formed by linearly expanding the boundary region image is shown in fig. 6. And the detection equipment linearly expands the boundary area image by using each pixel point on the boundary line and the normal line corresponding to the pixel point. Specifically, the method comprises the following steps:
and S531, sequentially mapping the boundary points to a second reference coordinate system along a first straight line by taking preset pixel points on the boundary line as starting points.
The boundary points are pixel points at the intersection of the normal and the boundary line, and the normal and the boundary points are in one-to-one correspondence.
Specifically, as shown in fig. 5, each of the pixels a1-a5 on the boundary line corresponds to a normal line, and the intersection point of the normal line and the boundary line is the boundary point (or the pixels a1-a 5). For example, starting from the boundary point a1 (i.e., the predetermined pixel), all boundary points on the boundary line are sequentially mapped to the second reference coordinate system along the first line, so as to form the mapped boundary points a1 '-a 5', as shown in fig. 6.
And S532, sequentially calculating the distance between each pixel point on the normal line and the corresponding boundary point.
As shown in fig. 5, the normal 13 corresponding to the boundary point a1 extracts coordinates of all pixels of the normal 13 on the line segment between the first boundary line 11 and the second boundary line 12, and calculates a distance between each pixel on the line segment and the boundary point a1 using the extracted coordinates and the coordinates of the boundary point a 1. And sequentially calculating the distances between the pixel points on the normal corresponding to the rest boundary points and the corresponding boundary points to obtain the relationship between the pixel points on the boundary area image and the corresponding boundary points.
And S533, sequentially mapping all pixel points on the boundary region image to a second reference coordinate system by using the distance to form a linear expansion image.
And pixel points in the boundary area image correspond to pixel points in the linear expansion image one by one. The method specifically comprises the following steps:
(1) and sequentially extracting pixel points on the normal line.
For example, the pixel point a6 on the normal to the boundary point a4 is extracted, and the distance between a4 and a6 has been calculated as L in S333.
(2) Inquiring mapping boundary points corresponding to the normal; the mapping boundary point is a mapping point of the boundary point corresponding to the normal on a second reference coordinate system;
the detection device uses the extracted pixel point a6 to inquire that the corresponding boundary point in the boundary region image is a4, and the mapping boundary point of a4 in the second reference coordinate system is a 4'. The origin of the second reference coordinate system can be a preset pixel point or other points; and only the second reference coordinate system is required to be used for uniquely determining the coordinates of each pixel point on the mapping boundary line. Subsequently, the coordinates of each pixel point on the mapping boundary line can be determined through the second reference coordinate system, and the coordinates of each pixel point on the linear expansion image can also be determined.
(3) Mapping the extracted pixel points to a second reference coordinate system along a second straight line; the mapping point of the extracted pixel point is separated from the mapping boundary point by a distance, the second straight line is perpendicular to the first straight line, and the second straight line corresponds to the mapping boundary point one to one.
For example, as shown in fig. 6, the mapping boundary point a4 ' forms a second straight line in a direction perpendicular to the first straight line, and the pixel point a6 is mapped onto the second straight line, where a distance between the mapping pixel point a6 ' and the mapping boundary point a4 ' is L.
And continuously repeating the steps until all the pixel points in the boundary area image are mapped into the second reference coordinate system.
And after the mapping of all the pixel points is completed, the detection equipment records the coordinates of all the pixel points in the formed linear expansion image in the second reference coordinate system for subsequently determining the position of the boundary defect in the image to be detected. The coordinates can be determined by the number of pixels between two pixels. For example, the coordinates of a1 'are (0,0), there are 100 pixel points between a 1' and a4 ', and then the coordinates of a 4' are (101, 0).
S54, boundary defects are detected based on the linear expansion image. Please refer to S24 in fig. 4 for details, which are not described herein.
And S55, when the linear expansion image is detected to have boundary defects, determining the position of the boundary defects in the image to be detected by using the boundary area image and the linear expansion image.
As shown in fig. 7c, when the detection device detects that the linear expanded image has a boundary defect, all pixel points at the defect position are marked, coordinates of the marked pixel points in the second reference coordinate system are extracted, and the extracted coordinates are matched with coordinates in the first reference coordinate system, so that the position of the boundary defect in the image to be detected can be determined. The method specifically comprises the following steps:
and S551, extracting coordinates of all pixel points in the area with the boundary defects on a second reference coordinate system.
And S552, matching the extracted coordinates with the coordinates on the first reference coordinate system to determine the position of the boundary defect in the image to be detected.
In the step S331, the coordinates of each pixel point in the boundary region image in the first reference coordinate system are obtained, and meanwhile, the pixel points in the boundary region image correspond to the pixel points in the linear expansion image one to one, so that the positions of the boundary defects can be determined in the image to be detected by using the coordinates.
Compared with the embodiment shown in fig. 4, the method for detecting the boundary defect provided by the embodiment can determine the position of the boundary defect in the image to be detected more quickly by using the coordinate matching method, and has higher detection efficiency and detection accuracy.
As an optional implementation manner of this embodiment, the steps S52 and S53 may be replaced by the following steps:
(1) and extracting boundary lines of two adjacent image areas, normal lines of each pixel point on the boundary lines and coordinates of each pixel point on the boundary lines on a first reference coordinate system.
(2) Based on the boundary line, the boundary area of the image to be detected is linearly expanded to form a linearly expanded image.
Wherein, the step (2) specifically comprises the following steps:
step (2.1): and sequentially mapping the boundary points to a second reference coordinate system along the first straight line by taking the preset pixel points on the boundary line as starting points.
Step (2.2): and respectively extracting all pixel points which are separated from the corresponding boundary points by a first preset distance and a second preset distance in the two adjacent image areas along the normal direction.
For example, as shown in fig. 5, the normal 13 corresponding to the boundary point a1 extends L1 to the region 2, extends L2 to the region 1, and extracts all the pixels of the normal 13 on the line segment length L1+ L2 and the distances between all the pixels and the boundary point a 1.
The boundary region is a closed region formed by extending the normal lines L1 to the region 2 and L2 to the region 1.
Please refer to S532 of the embodiment shown in fig. 9 in detail, which is not described herein again.
Step (2.3): and mapping all the extracted pixel points to a second reference coordinate system in sequence by using the distance between the pixel point on the normal line and the corresponding boundary point so as to form a linear expansion image. Please refer to S533 in fig. 9, which is not described herein.
In this embodiment, a boundary defect detection apparatus is further provided, and the apparatus is used to implement the foregoing embodiments and preferred embodiments, and the description of which has been already made is omitted. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
The present embodiment provides a boundary defect detecting apparatus, as shown in fig. 10, including:
an obtaining module 61, configured to obtain an image to be detected; wherein the image to be detected comprises at least two image areas.
And an extracting module 62, configured to extract a boundary area between two adjacent image areas to obtain a boundary area image.
And a linear expansion module 63 for linearly expanding the boundary area image to form a linearly expanded image.
And the detection module 64 is used for detecting the boundary defects based on the linear expansion image.
And a boundary defect determining module 65, configured to determine, when it is detected that the linear expanded image has a boundary defect, a position of the boundary defect in the image to be detected by using the boundary area image and the linear expanded image.
According to the boundary defect detection device provided by the embodiment of the invention, the boundary areas of two adjacent image areas are linearly expanded, namely, an irregular boundary area image is converted into a regular boundary area image, so that the robustness of the detection method can be improved, and the accuracy of the detection method is further improved; in addition, the defect detection of the boundary region can realize the detection of the defect close to or on the region boundary, and can also realize the complete detection of the defect across the region boundary, thereby having higher detection accuracy.
In some optional implementations of this embodiment, as shown in fig. 11, wherein the extracting module 62 includes:
the obtaining unit 621 is configured to obtain a boundary line between two adjacent image areas and a normal of each pixel point on the boundary line.
The shifting unit 622 is configured to shift the boundary line to two adjacent image regions by a first preset distance and a second preset distance respectively along a normal direction of each pixel point on the boundary line, so as to obtain a first boundary connection line and a second boundary connection line.
An extracting unit 623, configured to extract a region between the first boundary connecting line and the second boundary connecting line to obtain the boundary region image.
Further optionally, as shown in fig. 11, wherein the linear deployment module 63 comprises:
a first mapping unit 631, configured to map the boundary points sequentially onto a second reference coordinate system along a first straight line with preset pixel points on the boundary line as starting points; the boundary points are pixel points at the intersection points of the normal lines and the boundary lines, and the normal lines correspond to the boundary points one to one.
A calculating unit 632, configured to sequentially calculate a distance between each pixel point on the normal line and the corresponding boundary point.
A second mapping unit 633, configured to sequentially map all pixel points on the boundary region image onto the second reference coordinate system by using the distance, so as to form the linear expansion image; and pixel points in the boundary region image correspond to pixel points in the linear expansion image one to one.
The boundary defect detection apparatus in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC circuit, a processor and memory executing one or more software or fixed programs, and/or other devices that may provide the above-described functionality.
Further functional descriptions of the modules are the same as those of the corresponding embodiments, and are not repeated herein.
The embodiment of the invention also provides a detection device, which is provided with the boundary defect detection device shown in the figure 10 and the figure 11.
Referring to fig. 12, fig. 12 is a schematic structural diagram of a detection apparatus according to an alternative embodiment of the present invention, and as shown in fig. 12, the detection apparatus may include: at least one processor 71, such as a CPU (Central Processing Unit), at least one communication interface 73, memory 74, at least one communication bus 72. Wherein a communication bus 72 is used to enable the connection communication between these components. The communication interface 73 may include a Display (Display) and a Keyboard (Keyboard), and the optional communication interface 73 may also include a standard wired interface and a standard wireless interface. The Memory 74 may be a high-speed RAM Memory (volatile Random Access Memory) or a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. The memory 74 may alternatively be at least one memory device located remotely from the processor 71. Wherein the processor 71 may be combined with the apparatus described in fig. 10 and fig. 11, the memory 74 stores an application program, and the processor 71 calls the program code stored in the memory 74 for executing any of the above method steps.
The communication bus 72 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus 72 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 12, but this is not intended to represent only one bus or type of bus.
The memory 74 may include a volatile memory (RAM), such as a random-access memory (RAM); the memory may also include a non-volatile memory (english: non-volatile memory), such as a flash memory (english: flash memory), a hard disk (english: hard disk drive, abbreviated: HDD) or a solid-state drive (english: SSD); the memory 74 may also comprise a combination of memories of the kind described above.
The processor 71 may be a Central Processing Unit (CPU), a Network Processor (NP), or a combination of CPU and NP.
The processor 71 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a Programmable Logic Device (PLD), or a combination thereof. The PLD may be a Complex Programmable Logic Device (CPLD), a field-programmable gate array (FPGA), a General Array Logic (GAL), or any combination thereof.
Optionally, the memory 74 is also used for storing program instructions. The processor 71 may call program instructions to implement the boundary defect detection method as shown in the embodiments of fig. 1, fig. 4, fig. 8 and fig. 9 of the present application.
An embodiment of the present invention further provides a non-transitory computer storage medium, where a computer-executable instruction is stored in the computer storage medium, and the computer-executable instruction can execute the boundary defect detection method in any of the above method embodiments. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD), a Solid State Drive (SSD), or the like; the storage medium may also comprise a combination of memories of the kind described above.
Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope defined by the appended claims.
Claims (9)
1. A method for detecting a boundary defect, comprising:
acquiring an image to be detected; the image to be detected comprises at least two image areas, wherein the image areas comprise an image of a product to be detected and a background image of the product to be detected;
extracting boundary areas of two adjacent image areas to obtain boundary area images;
linearly unfolding the boundary area image to form a linearly unfolded image;
detecting the boundary defects based on the linear expansion image;
when the linear expansion image is detected to have the boundary defect, determining the position of the boundary defect in the image to be detected by using the boundary area image and the linear expansion image;
wherein, the extracting the boundary area of two adjacent image areas to obtain a boundary area image comprises:
acquiring a boundary line of two adjacent image areas and a normal and coordinates of each pixel point on the boundary line; wherein the coordinates are uniquely determined by a first reference coordinate system;
respectively shifting the boundary line to two adjacent image areas by a first preset distance and a second preset distance along the normal direction of each pixel point on the boundary line to obtain a first boundary connecting line and a second boundary connecting line;
and extracting a region between the first boundary connecting line and the second boundary connecting line to obtain the boundary region image.
2. The method of claim 1, wherein said linearly expanding the boundary region image to form a linearly expanded image comprises:
sequentially mapping the boundary points to a second reference coordinate system along a first straight line by taking preset pixel points on the boundary line as starting points; the boundary points are pixel points at the intersection points of the normal lines and the boundary lines, and the normal lines and the boundary points are in one-to-one correspondence;
sequentially calculating the distance between each pixel point on the normal line and the corresponding boundary point;
sequentially mapping all pixel points on the boundary area image to the second reference coordinate system by using the distance so as to form the linear expansion image; and pixel points in the boundary region image correspond to pixel points in the linear expansion image one to one.
3. The method according to claim 2, wherein said sequentially mapping all pixels on the boundary region image onto the second reference coordinate system by using the distance comprises:
sequentially extracting pixel points on the normal;
inquiring a mapping boundary point corresponding to the normal; wherein the mapping boundary point is a mapping point of the boundary point corresponding to the normal on the second reference coordinate system;
mapping the extracted pixel points to the second reference coordinate system along a second straight line; the extracted pixel point mapping point and the extracted pixel point mapping boundary point are separated by the distance, the second straight line is perpendicular to the first straight line, and the second straight line corresponds to the mapping boundary point one to one.
4. The method of claim 1, wherein the detecting the boundary defect based on the linearly expanded image comprises:
filtering the linear expansion image to form a filtered image;
comparing the linear expansion image with the filtering image to obtain a difference image;
screening out areas meeting preset conditions in the difference image; wherein the screened area is the area of the detected boundary defect.
5. The method according to claim 1, wherein the determining the position of the boundary defect in the image to be detected by using the boundary region image and the straight line expansion image comprises:
extracting coordinates of all pixel points in the area of the boundary defect on a second reference coordinate system;
and matching the extracted coordinates with coordinates on a first reference coordinate system to determine the position of the boundary defect in the image to be detected.
6. The method according to claim 1, wherein the obtaining the boundary line between two adjacent image areas comprises:
acquiring a defect-free template image; the non-defective template images correspond to the images to be detected one by one;
extracting the boundary line of two adjacent image areas of the non-defective template image;
aligning the non-defective template image and the image to be detected;
mapping the extracted boundary line to the image to be detected; and extracting the boundary line, wherein the extracted boundary line is the boundary line of two adjacent image areas of the image to be detected.
7. An apparatus for detecting a boundary defect, comprising:
the acquisition module is used for acquiring an image to be detected; the image to be detected comprises at least two image areas, wherein the image areas comprise an image of a product to be detected and a background image of the product to be detected;
the extraction module is used for extracting the boundary areas of two adjacent image areas to obtain a boundary area image;
the linear expansion module is used for linearly expanding the boundary area image to form a linearly expanded image;
the detection module is used for detecting the boundary defects based on the linear expansion image;
the boundary defect determining module is used for determining the position of the boundary defect in the image to be detected by utilizing the boundary area image and the linear expansion image when the linear expansion image is detected to have the boundary defect;
wherein, the extracting the boundary area of two adjacent image areas to obtain a boundary area image comprises:
acquiring a boundary line of two adjacent image areas and a normal and coordinates of each pixel point on the boundary line; wherein the coordinates are uniquely determined by a first reference coordinate system;
respectively shifting the boundary line to two adjacent image areas by a first preset distance and a second preset distance along the normal direction of each pixel point on the boundary line to obtain a first boundary connecting line and a second boundary connecting line;
and extracting a region between the first boundary connecting line and the second boundary connecting line to obtain the boundary region image.
8. A detection apparatus, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the method of detecting boundary defects of any of claims 1-6.
9. A computer-readable storage medium having stored thereon computer instructions, which when executed by a processor, perform the steps of the method for detecting boundary defects according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811062811.9A CN109325930B (en) | 2018-09-12 | 2018-09-12 | Boundary defect detection method, device and detection equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811062811.9A CN109325930B (en) | 2018-09-12 | 2018-09-12 | Boundary defect detection method, device and detection equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109325930A CN109325930A (en) | 2019-02-12 |
| CN109325930B true CN109325930B (en) | 2021-09-28 |
Family
ID=65265869
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811062811.9A Active CN109325930B (en) | 2018-09-12 | 2018-09-12 | Boundary defect detection method, device and detection equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109325930B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111951290B (en) * | 2019-05-16 | 2023-11-03 | 杭州睿琪软件有限公司 | Edge detection method and device for object in image |
| CN110717902B (en) * | 2019-09-29 | 2023-06-09 | 中山市瑞福达触控显示技术有限公司 | Processing method for display image edge |
| CN111353974B (en) * | 2020-02-20 | 2023-08-18 | 苏州凌云光工业智能技术有限公司 | Method and device for detecting image boundary defects |
| CN113538478A (en) * | 2020-04-15 | 2021-10-22 | 深圳市光鉴科技有限公司 | Image-based box boundary extraction method, system, equipment and storage medium |
| CN111598074B (en) * | 2020-05-21 | 2023-07-07 | 杭州睿琪软件有限公司 | Edge detection method and device, electronic equipment and storage medium |
| CN111798449B (en) * | 2020-09-09 | 2021-02-05 | 江苏恒力化纤股份有限公司 | Spinneret plate residual impurity detection method based on image technology |
| CN112686919B (en) * | 2020-12-29 | 2022-10-21 | 上海闻泰信息技术有限公司 | Object boundary line determining method and device, electronic equipment and storage medium |
| CN113034527B (en) * | 2021-03-30 | 2022-05-03 | 长江存储科技有限责任公司 | Boundary detection method and related product |
| CN113658133B (en) * | 2021-08-16 | 2022-06-21 | 江苏鑫丰源机电有限公司 | Gear surface defect detection method and system based on image processing |
| CN114820594B (en) * | 2022-06-21 | 2022-09-23 | 中科慧远视觉技术(北京)有限公司 | Method for detecting edge sealing defect of plate based on image, related equipment and storage medium |
| CN115880248B (en) * | 2022-12-13 | 2024-02-09 | 哈尔滨耐是智能科技有限公司 | Surface scratch defect identification method and visual detection equipment |
| CN115830043B (en) * | 2023-01-09 | 2023-05-26 | 北京阿丘机器人科技有限公司 | Boundary detection method, device, equipment and storage medium for wireless charging magnet |
| CN117437233B (en) * | 2023-12-21 | 2024-03-26 | 山东润通齿轮集团有限公司 | Gear defect detection method and system based on image processing |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102062739A (en) * | 2006-05-23 | 2011-05-18 | 麒麟工程技术系统公司 | Surface examining device |
| CN103745475A (en) * | 2014-01-22 | 2014-04-23 | 哈尔滨工业大学 | Detection and positioning method used for spherical pin element |
| CN105066892A (en) * | 2015-08-05 | 2015-11-18 | 哈尔滨工业大学 | BGA element detecting and positioning method based on linear clustering analysis |
| CN106600600A (en) * | 2016-12-26 | 2017-04-26 | 华南理工大学 | Wafer defect detection method based on characteristic matching |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102004015110A1 (en) * | 2004-03-27 | 2005-10-13 | Texmag Gmbh Vertriebsgesellschaft | Detection apparatus for detecting defects in joints of sheet pieces, has source(s) of electromagnetic non-unidirectional radiations, and one or more sensors that make two-dimensional detection of reflected or refracted radiation |
| CN102288620B (en) * | 2011-08-10 | 2013-04-03 | 天津大学 | Method and device for unfolding surface of steel ball on basis of multiple image sensors |
| JP5726045B2 (en) * | 2011-11-07 | 2015-05-27 | 株式会社神戸製鋼所 | Tire shape inspection method and tire shape inspection device |
| JP2013175684A (en) * | 2012-02-27 | 2013-09-05 | Canon Inc | Detector, imprint device and article manufacturing method |
| US8864283B1 (en) * | 2013-05-09 | 2014-10-21 | Xerox Corporation | System and method for visually detecting defective inkjets in an inkjet imaging apparatus |
| CN103499590B (en) * | 2013-10-17 | 2015-11-18 | 福州大学 | Ring-shaped work pieces end face defect detection and screening technique and system |
| CN104568983B (en) * | 2015-01-06 | 2017-03-15 | 浙江工业大学 | Pipeline Inner Defect Testing device and method based on active panoramic vision |
| CN105913415B (en) * | 2016-04-06 | 2018-11-30 | 博众精工科技股份有限公司 | A kind of image sub-pixel edge extracting method with extensive adaptability |
| CN106056546B (en) * | 2016-05-25 | 2019-02-22 | 广东工业大学 | A kind of image repair method and device based on Exemplar Matching |
| CN107328781A (en) * | 2017-05-23 | 2017-11-07 | 镇江苏仪德科技有限公司 | A kind of columnar product detection method of surface flaw and device based on machine vision |
| CN108459027A (en) * | 2018-03-21 | 2018-08-28 | 华北电力大学 | A kind of blade of wind-driven generator detection method of surface flaw based on image procossing |
-
2018
- 2018-09-12 CN CN201811062811.9A patent/CN109325930B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102062739A (en) * | 2006-05-23 | 2011-05-18 | 麒麟工程技术系统公司 | Surface examining device |
| CN103745475A (en) * | 2014-01-22 | 2014-04-23 | 哈尔滨工业大学 | Detection and positioning method used for spherical pin element |
| CN105066892A (en) * | 2015-08-05 | 2015-11-18 | 哈尔滨工业大学 | BGA element detecting and positioning method based on linear clustering analysis |
| CN106600600A (en) * | 2016-12-26 | 2017-04-26 | 华南理工大学 | Wafer defect detection method based on characteristic matching |
Non-Patent Citations (1)
| Title |
|---|
| 钢板表面缺陷图像检测与分类技术研究;于海燕 等;《中国优秀硕士学位论文全文数据库 信息科技辑》;20170715(第07期);第I138-655页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109325930A (en) | 2019-02-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109325930B (en) | Boundary defect detection method, device and detection equipment | |
| CN106934803B (en) | method and device for detecting surface defects of electronic device | |
| CN110476056B (en) | Visual detection method, detection equipment and robot | |
| CN109242853B (en) | PCB defect intelligent detection method based on image processing | |
| WO2019188040A1 (en) | Image processing device, image processing method, and image processing program | |
| CN111612781A (en) | Screen defect detection method and device and head-mounted display equipment | |
| CN105160652A (en) | Handset casing testing apparatus and method based on computer vision | |
| CN110321933B (en) | Fault identification method and device based on deep learning | |
| CN110596120A (en) | Glass boundary defect detection method, device, terminal and storage medium | |
| CN112577969B (en) | Defect detection method and defect detection system based on machine vision | |
| WO2015100777A1 (en) | Display panel defect detecting method and defect detecting device | |
| CN113077437B (en) | Workpiece quality detection method and system | |
| US20210272272A1 (en) | Inspection support apparatus, inspection support method, and inspection support program for concrete structure | |
| CN113160161B (en) | Method and device for detecting defects at edge of target | |
| CN112164050B (en) | Method and device for detecting surface defects of products on production line and storage medium | |
| CN109191516B (en) | Rotation correction method and device of structured light module and readable storage medium | |
| JP6177541B2 (en) | Character recognition device, character recognition method and program | |
| CN115018846A (en) | AI intelligent camera-based multi-target crack defect detection method and device | |
| US10074551B2 (en) | Position detection apparatus, position detection method, information processing program, and storage medium | |
| CN105205803A (en) | Display panel defect detection method | |
| JP6197340B2 (en) | Image processing apparatus, image processing method, and program | |
| JP4230880B2 (en) | Defect inspection method | |
| CN113785181A (en) | OLED screen point defect judgment method and device, storage medium and electronic equipment | |
| WO2014103617A1 (en) | Alignment device, defect inspection device, alignment method, and control program | |
| CN114677348A (en) | IC chip defect detection method and system based on vision and storage medium |
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 |