CN111583223B - Defect detection method, defect detection device, computer equipment and computer readable storage medium - Google Patents
Defect detection method, defect detection device, computer equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN111583223B CN111583223B CN202010377497.4A CN202010377497A CN111583223B CN 111583223 B CN111583223 B CN 111583223B CN 202010377497 A CN202010377497 A CN 202010377497A CN 111583223 B CN111583223 B CN 111583223B
- Authority
- CN
- China
- Prior art keywords
- detection result
- image
- product
- deep learning
- area
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 316
- 230000007547 defect Effects 0.000 title claims abstract description 135
- 238000013135 deep learning Methods 0.000 claims abstract description 150
- 230000002950 deficient Effects 0.000 claims abstract description 80
- 238000010191 image analysis Methods 0.000 claims abstract description 50
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000012545 processing Methods 0.000 claims description 44
- 238000004590 computer program Methods 0.000 claims description 35
- 238000004458 analytical method Methods 0.000 claims description 23
- 238000012549 training Methods 0.000 claims description 19
- 238000007781 pre-processing Methods 0.000 claims description 11
- 230000007797 corrosion Effects 0.000 claims description 8
- 238000005260 corrosion Methods 0.000 claims description 8
- 238000010586 diagram Methods 0.000 description 11
- 230000003287 optical effect Effects 0.000 description 5
- 230000000295 complement effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000003042 antagnostic effect Effects 0.000 description 3
- 239000003086 colorant Substances 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 230000003628 erosive effect Effects 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000012935 Averaging Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000003796 beauty Effects 0.000 description 1
- 230000002902 bimodal effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000010339 dilation Effects 0.000 description 1
- 235000019800 disodium phosphate Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 239000010979 ruby Substances 0.000 description 1
- 229910001750 ruby Inorganic materials 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0006—Industrial image inspection using a design-rule based approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- 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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
-
- 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 discloses a defect detection method, a defect detection device, computer equipment and a computer readable storage medium. The method comprises the following steps: acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information; analyzing the contour information by using an image analysis algorithm to obtain a first detection result of the product to be detected; analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected; and obtaining a final detection result of the product to be detected according to the first detection result and the second detection result. The scheme provided by the invention can combine the image analysis algorithm and the image deep learning algorithm, improves the detection rate of defects, solves the judgment dispute between good products and defective products, and reduces the misjudgment rate of products to be detected.
Description
Technical Field
The embodiment of the invention relates to the technical field of product detection, in particular to a defect detection method, a defect detection device, computer equipment and a computer readable storage medium.
Background
The surface defects are defects such as spots, pits, color differences, scratches, defects and the like on the surface of the product. In the industrial field, surface defects directly affect the properties of the product, such as beauty, performance and the like, so the surface quality of the product is important.
In order to ensure the surface quality of the product, surface defects of the product need to be detected. The detection method of the surface defects is mainly divided into an artificial detection method and a machine detection method. The manual detection method needs to consume a large amount of manpower, and the manual detection often cannot ensure that the detection standards of all people are consistent; the machine detection method is divided into two types of image analysis algorithm and image deep learning algorithm, and the two types of algorithms cannot avoid the occurrence of misjudgment, so that the detection rate of defects is not high.
Disclosure of Invention
The embodiment of the invention provides a defect detection method, a device, computer equipment and a computer readable storage medium, which can combine an image analysis algorithm and an image deep learning algorithm, improve the detection rate of defects, solve the judgment dispute between good products and defective products and reduce the misjudgment rate of products to be detected.
In an embodiment, the present invention provides a defect detection method, including:
Acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information;
analyzing the contour information by using an image analysis algorithm to obtain a first detection result of the product to be detected;
analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected;
and obtaining a final detection result of the product to be detected according to the first detection result and the second detection result.
Optionally, acquiring image information of the product to be detected includes:
acquiring picture information;
the picture information is preprocessed to obtain contour information, wherein the preprocessing comprises at least one of image graying processing, binarization processing, expansion processing and corrosion processing.
Optionally, analyzing the profile information by using an image analysis algorithm to obtain a first detection result of the product to be detected, including:
dividing the contour information into a defect area and a normal area by using an image analysis algorithm;
calculating the area of the defect area and the area of the normal area respectively;
if the ratio of the area of the defect area to the area of the normal area is greater than or equal to a preset threshold value, the first detection result is a defective product;
if the ratio of the area of the defect area to the area of the normal area is smaller than a preset threshold value, the first detection result is good.
Optionally, the image deep learning algorithm includes an image deep learning strict model and an image deep learning loose model, the image deep learning strict model having a lower tolerance to defects than the image deep learning loose model.
Optionally, when the first detection result is a defective product, analyzing the image information by using an image deep learning algorithm to obtain a second detection result of the product to be detected, including:
inputting the picture information into an image deep learning strict model to obtain a second detection result;
correspondingly, according to the first detection result and the second detection result, a final detection result of the product to be detected is obtained, which comprises the following steps:
if the first detection result is defective and the second detection result is defective, the final detection result is defective;
if the first detection result is defective and the second detection result is good, the final detection result is good.
Optionally, when the first detection result is good, analyzing the image information by using an image deep learning algorithm to obtain a second detection result of the product to be detected, including:
inputting the picture information into an image deep learning loose model to obtain a second detection result;
Correspondingly, according to the first detection result and the second detection result, a final detection result of the product to be detected is obtained, which comprises the following steps:
if the first detection result is good and the second detection result is good, the final detection result is good;
if the first detection result is good and the second detection result is defective, the final detection result is defective.
Optionally, the defect detection method further includes:
acquiring historical image information;
and training to obtain an image deep learning strict model and an image deep learning loose model according to the historical image information.
In an embodiment, the embodiment of the invention also provides a defect detection device, which comprises an image processing module, a first analysis module, a second analysis module and a result analysis module;
the image processing module is used for acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information;
the first analysis module is used for analyzing the contour information by utilizing an image analysis algorithm to obtain a first detection result of the product to be detected;
the second analysis module is used for analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected;
And the result analysis module is used for obtaining a final detection result of the product to be detected according to the first detection result and the second detection result.
In an embodiment, the embodiment of the present invention further provides a computer device, including: a processor for implementing the method of any of the above embodiments when executing a computer program.
In an embodiment, the present invention further provides a computer readable storage medium storing a computer program, where the computer program when executed by a processor implements the method of any of the above embodiments.
In an embodiment, the embodiment of the present invention further provides a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements the steps of the defect detection method provided in any embodiment of the present application when executing the computer program.
In an embodiment, the present invention further provides a computer readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the defect detection method provided in any embodiment of the present application.
The embodiment of the invention provides a defect detection method, a defect detection device, computer equipment and a computer readable storage medium. By the method, firstly, an image analysis algorithm is adopted to detect a product to be detected, and a first detection result is obtained; then detecting a product to be detected by adopting an image deep learning algorithm to obtain a second detection result; and finally, combining the first detection result and the second detection result, and comprehensively analyzing to obtain a final detection result of the product to be detected. The image analysis algorithm and the image deep learning algorithm are combined, and form a complementary and antagonistic relation, so that the detection rate of defects is improved, the judgment dispute between good products and defective products is solved to a certain extent, and the misjudgment rate of products to be detected is reduced.
Drawings
FIG. 1 is a flow chart of a defect detection method according to an embodiment;
FIG. 2 is a flow chart illustrating another defect detection method according to an embodiment;
FIG. 3 is a schematic diagram of picture information and outline information of a product to be inspected according to an embodiment;
FIG. 4 is a schematic diagram of picture information and outline information of another product to be inspected according to an embodiment;
FIG. 5 is a flow chart illustrating another defect detection method according to an embodiment;
FIG. 6 is a schematic diagram of a defect detecting apparatus according to an embodiment;
FIG. 7 is a schematic diagram of another defect detecting device according to an embodiment;
fig. 8 is a schematic structural diagram of a computer device according to an embodiment.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present invention are shown in the drawings.
In the field of industrial product defect detection, the detection method of surface defects is mainly divided into an artificial detection method and a machine detection method. Aiming at the machine detection method, the method is mainly divided into two major categories of an image analysis algorithm and an image deep learning algorithm, and the two categories of algorithms have advantages and disadvantages: the image analysis algorithm can accurately calculate the position and the area of the defect, and judge the defect degree more accurately, but the image analysis algorithm is effective only for relatively obvious defects, and has poor effect for some relatively special and unobvious defects (such as tiny foreign matters and the like) and possibly cannot be detected; the image deep learning algorithm has higher detection rate of defects, but the judgment of the defect degree (such as the size of the defect area) is not as simple and direct as the image analysis algorithm. Thus, the image analysis algorithm and the image deep learning algorithm are controversial in judging good products and defective products.
To this end, the present invention provides a defect detection method, apparatus, computer device, and computer-readable storage medium. Firstly, detecting a product to be detected by adopting an image analysis algorithm to obtain a first detection result; then detecting a product to be detected by adopting an image deep learning algorithm to obtain a second detection result; and finally, combining the first detection result and the second detection result, and comprehensively analyzing to obtain a final detection result of the product to be detected. The image analysis algorithm and the image deep learning algorithm are combined, and form a complementary and antagonistic relation, so that the detection rate of defects is improved, the judgment dispute between good products and defective products is solved to a certain extent, and the misjudgment rate of products to be detected is reduced.
The following embodiments of the present invention may be implemented individually or in combination with each other, and the embodiments of the present invention are not limited thereto. Various components are described in terms of "first," "second," etc. in embodiments of the invention, but these components should not be limited by these terms. These terms are only used to distinguish one element from another element.
Next, a defect detection method, apparatus, and technical effects thereof will be described.
Fig. 1 is a flow chart of a defect detection method according to an embodiment, and as shown in fig. 1, the method provided in the embodiment is suitable for a defect detection device, and the method includes the following steps.
S101, acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information.
Specifically, the method for acquiring the image information of the product to be detected may include the following two steps:
and step 1, acquiring picture information.
The picture information may be acquired by the defect detection device itself or may be acquired by the defect detection device from another device.
The defect detecting device may be any device having a defect detecting function, such as a terminal device or a computer. Alternatively, the defect detection device may comprise an image acquisition device (such as a camera or a video camera) for acquiring picture information of the product to be detected. Of course, when the defect detecting device does not include the image capturing apparatus, the defect detecting device may be electrically connected to the image capturing apparatus to receive the picture information of the product to be detected captured by the image capturing apparatus.
And 2, preprocessing the picture information to obtain contour information, wherein the preprocessing comprises at least one of image graying processing, binarization processing, expansion processing and corrosion processing.
The preprocessing of the picture information means that the picture information is processed by at least one processing method of image graying processing, binarization processing, expansion processing and corrosion processing to obtain the outline of the product to be detected.
The R (red) G (green) B (blue) color mode is to assign an intensity value in the range of 0-255 to the RGB components of each pixel in the image using the RGB model. RGB images can be mixed according to different proportions using only three colors, and 16777216 (256×256) colors are presented on the screen. In the RGB model, if r=g=b, the color represents a gray color, where the value of r=g=b is called a gray value, and thus, the gray image only needs one byte for storing the gray value (also called an intensity value, a brightness value) for each pixel, and the gray range is 0-255. For example, 0% of the gray scale RGB values are 255,255; the RGB value for 1% gray is 253,253,253; the 2% gray RGB value is 250,250,250.
Gray scale images are different from black and white images, and in the field of computer images, black and white images only have two colors, and gray scale images have a plurality of levels of color depth between black and white. The image graying processing provided in the present embodiment may be any one of a component method (taking the luminance of three components in a color image as the gray value of three gray images, one gray image may be selected according to application requirements), a maximum value method (taking the maximum value of the luminance of three components in a color image as the gray value of a gray image), an average method (averaging the luminance of three components in a color image to obtain one gray image), and a weighted average method (weighted averaging three components with different weights according to importance and other indexes).
Binarization (Thresholding) is a simple method of image segmentation. The binarization process may convert the gray scale image into a binary image. The pixel gradation larger than a certain critical gradation value is set as a gradation maximum value, and the pixel gradation smaller than this value is set as a gradation minimum value, thereby realizing binarization. Depending on the choice of the threshold, the binarization algorithm can be divided into a fixed threshold and an adaptive threshold. The more commonly used binarization methods are: bimodal, P-parametric, iterative, OTSU, etc.
The expansion process (expansion) and the Erosion process (Erosion) are morphological operations, which are a series of image processing operations based on shapes. Dilation and erosion processes are commonly used to eliminate noise, to split (isolate) individual image elements, to connect (join) adjacent image elements, to find distinct maxima or minima regions in an image.
S102, analyzing the contour information by using an image analysis algorithm to obtain a first detection result of the product to be detected.
The contour information is analyzed by using an image analysis algorithm, so that the contours and the positions of some obvious defects can be obtained, and the areas of the defects can be accurately calculated. By analyzing the outline, the position and the area of the defect, a first detection result of the product to be detected can be obtained, namely whether the product to be detected is a good product or a defective product can be judged by utilizing an image analysis algorithm.
And S103, analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected.
The image deep learning algorithm comprises an image deep learning strict model and an image deep learning loose model, wherein the tolerance of the image deep learning strict model to the defects is lower than that of the image deep learning loose model. And analyzing the picture information by selecting a proper model to obtain a second detection result of the product to be detected, namely judging whether the product to be detected is good or bad by utilizing an image deep learning algorithm.
Image deep learning rigorous model (High): the tolerance to defects is low and strict, so that the image deep learning strict model can be approximately considered to be judged as good, and the image deep learning strict model is necessarily considered to be the good; however, if the strict model for image deep learning is judged to be defective, the model is not necessarily defective. Namely, the judgment reliability of the strict model for image deep learning on good products is high.
Image deep learning loose model (Low): the tolerance to defects is higher and is looser, so that the image deep learning loose model can be approximately considered to be defective if the image deep learning loose model is judged to be defective; however, if the loose model for image deep learning is judged to be good, the loose model is not necessarily good. Namely, the reliability of judging defective products by the image deep learning loose model is high.
S104, obtaining a final detection result of the product to be detected according to the first detection result and the second detection result.
And combining the detection results of the steps S102 and S103, and combining an image analysis algorithm and an image deep learning algorithm to obtain a final detection result of the product to be detected. I.e. whether the product to be detected is a good product or a defective product finally.
Fig. 2 is a schematic flow chart of another defect detection method according to an embodiment, as shown in fig. 2, where the method provided in this embodiment is applicable to a defect detection device, and is different from the embodiment shown in fig. 1, in the embodiment shown in fig. 2, a complementary and opposite relationship between an image analysis algorithm and an image deep learning algorithm is explained in detail. The method comprises the following steps.
S201, acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information.
Specifically, the method for acquiring the image information of the product to be detected may include the following two steps:
and step 1, acquiring picture information.
And 2, preprocessing the picture information to obtain contour information, wherein the preprocessing comprises at least one of image graying processing, binarization processing, expansion processing and corrosion processing.
S202, dividing the contour information into a defect area and a normal area by using an image analysis algorithm.
Fig. 3 is a schematic diagram of image information and profile information of a product to be detected according to an embodiment, as shown in fig. 3, fig. 3 (a) is the image information of the product to be detected, and fig. 3 (b) is the profile information of the product to be detected in fig. 3 (a). The contour information is analyzed by using an image analysis algorithm, and the contour information can be divided into a defect area and a normal area. Referring to fig. 3 (b), a white area is a normal area (may also be referred to as a good area), a black area is a background area, and a gray area is a defective area.
S203, calculating the area of the defect area and the area of the normal area respectively.
By accurately quantifying the area of the defective area and the area of the normal area, the size and position of the defective area can be clarified.
S204, judging the relation between the ratio of the area of the defect area to the area of the normal area and the preset threshold value.
And judging the relation between the ratio of the area of the defect area to the area of the normal area and the preset threshold value, and obtaining a detection result of detecting the product to be detected by adopting an image analysis algorithm.
Namely: if the ratio of the area of the defect area to the area of the normal area is greater than or equal to a preset threshold value, the first detection result is a defective product; if the ratio of the area of the defect area to the area of the normal area is smaller than a preset threshold value, the first detection result is good.
The value of the preset threshold value can be set according to the requirement of an actual product.
However, the image analysis algorithm has poor detection effect on some defects (such as hair-like foreign matters, thin and long obvious defects, light leakage bright spots and the like) which are special and not obvious. Fig. 4 is a schematic diagram of picture information and profile information of another product to be detected according to an embodiment, fig. 4 (a) is picture information (with hair-like foreign matter) of the product to be detected, and fig. 4 (b) is profile information of the product to be detected in fig. 4 (a); fig. 4 (c) is picture information (with thin and long scratches) of the product to be inspected, and fig. 4 (d) is profile information of the product to be inspected in fig. 4 (c); fig. 4 (e) is picture information (with light leakage spot) of the product to be detected, and fig. 4 (f) is profile information of the product to be detected in fig. 4 (e). Referring to fig. 4 (b), 4 (d) and 4 (f), it can be seen that the profile information of the product to be inspected cannot find the defects. Therefore, it is necessary to verify the first detection result obtained by using the image analysis algorithm using the image deep learning algorithm.
S205, if the ratio of the area of the defect area to the area of the normal area is greater than or equal to a preset threshold value, the first detection result is a defective product.
S206, inputting the picture information into an image deep learning strict model to obtain a second detection result.
And selecting an image deep learning strict model for verification for the product to be detected, wherein the first detection result is defective. Specifically, the picture information is input into an image deep learning strict model, and a second detection result is obtained.
Image deep learning rigorous model (High): the tolerance to defects is low and strict, so that the image deep learning strict model can be approximately considered to be judged as good, and the image deep learning strict model is necessarily considered to be the good; however, if the strict model for image deep learning is judged to be defective, the model is not necessarily defective. Namely, the judgment reliability of the strict model for image deep learning on good products is high.
S207, if the first detection result is defective and the second detection result is defective, the final detection result is defective.
S208, if the first detection result is a defective product and the second detection result is a good product, the final detection result is a good product.
S209, if the ratio of the area of the defect area to the area of the normal area is smaller than a preset threshold value, the first detection result is good.
S210, inputting the picture information into an image deep learning loose model to obtain a second detection result.
And selecting the image deep learning loose model for verification for the product to be detected with the first detection result being good. Specifically, the picture information is input into an image deep learning loose model, and a second detection result is obtained.
Image deep learning loose model (Low): the tolerance to defects is higher and is looser, so that the image deep learning loose model can be approximately considered to be defective if the image deep learning loose model is judged to be defective; however, if the loose model for image deep learning is judged to be good, the loose model is not necessarily good. Namely, the reliability of judging defective products by the image deep learning loose model is high.
S211, if the first detection result is good and the second detection result is good, the final detection result is good.
S212, if the first detection result is good and the second detection result is defective, the final detection result is defective.
From the above steps, it can be seen that firstly, an image analysis algorithm is adopted to detect the product to be detected, so as to obtain a first detection result; then detecting a product to be detected by adopting an image deep learning algorithm to obtain a second detection result; and finally, combining the first detection result and the second detection result, and comprehensively analyzing to obtain a final detection result of the product to be detected. The image analysis algorithm and the image deep learning algorithm are mutually assisted to fully cover the defects, so that the detection rate of the defects is improved.
But at the same time, the image analysis algorithm and the image deep learning algorithm "challenge" each other. Namely, the image analysis algorithm judges that the image is good, the image deep learning algorithm judges that the image is bad as much as possible (namely, the image analysis algorithm judges that the image is good, and the image deep learning loose model is used for verification, so that the image deep learning loose model has high reliability of judging that the image is bad, only the image deep learning loose model is used for re-detecting once, and if the image deep learning loose model judges that the image is bad, and the image deep learning loose model judges that the image is bad; the image analysis algorithm judges that the image is defective, the image deep learning algorithm judges the image as being defective (namely, the image analysis algorithm judges the image as being defective and then verifies the image by using the image deep learning strict model, the image deep learning strict model has high reliability of judging the defective, only the image deep learning strict model is used for detecting the image again, and if the image deep learning strict model judges the image as being defective, the image deep learning algorithm judges the image as being defective, and the image deep learning algorithm judges the image as being defective.
Fig. 5 is a schematic flow chart of another defect detection method according to an embodiment, and as shown in fig. 5, the method according to the embodiment is applicable to a defect detection device, unlike the embodiment shown in fig. 1, in which training of a model is involved in the embodiment shown in fig. 5. The method comprises the following steps.
S301, acquiring historical image information.
The historical image information can be image information of a product to be detected which is detected in a historical mode, and can also be image information which is selected for model training.
S302, training to obtain an image deep learning strict model and an image deep learning loose model according to historical image information.
When training the image deep learning strict model and the image deep learning loose model, the specific training can be performed according to the type and the size of the defects aiming at the wide and strict requirements of detection.
Illustratively, in training the model, the same set of training products (assuming that product 1, product 2, and product 3 are included) differ in their corresponding training data for the image deep learning strict model and the image deep learning relaxed model. This is because the image deep learning strict model has a low tolerance to defects, while the image deep learning loose model has a high tolerance to defects, for example, the product 1 is a defective product under the detection of the image deep learning strict model, and is a good product under the detection of the image deep learning loose model; the product 2 is a defective product under the detection of the image deep learning strict model, and is a defective product under the detection of the image deep learning loose model; the product 3 is good under the detection of the image deep learning strict model, and is also good under the detection of the image deep learning loose model. Thus, training data of the image Deep Learning strict model and training data of the image Deep Learning loose model can be selectively set, and are respectively input into a Deep Learning (DL) model for training, so that the image Deep Learning strict model and the image Deep Learning loose model are obtained.
Deep Learning (DL) algorithms belong to the field of Machine Learning (ML), and are processes for Learning the internal rules and presentation layers of sample data, and information obtained in these Learning processes greatly helps to explain data such as text, images and sounds. The deep learning algorithm can be implemented through a deep learning model, and training of an image deep learning strict model and an image deep learning loose model can be performed by adopting a convolutional neural network (convolutional neural network) model, a Deep Belief Network (DBN) model and a stacked auto-encoder network (stacked auto-encoder network) model.
S303, acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information.
S304, analyzing the contour information by using an image analysis algorithm to obtain a first detection result of the product to be detected.
S305, analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected.
S306, obtaining a final detection result of the product to be detected according to the first detection result and the second detection result.
The embodiment of the invention provides a defect detection method, which comprises the following steps: acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information; analyzing the contour information by using an image analysis algorithm to obtain a first detection result of the product to be detected; analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected; and obtaining a final detection result of the product to be detected according to the first detection result and the second detection result. By the method, firstly, an image analysis algorithm is adopted to detect a product to be detected, and a first detection result is obtained; then detecting a product to be detected by adopting an image deep learning algorithm to obtain a second detection result; and finally, combining the first detection result and the second detection result, and comprehensively analyzing to obtain a final detection result of the product to be detected. The image analysis algorithm and the image deep learning algorithm are combined, and form a complementary and antagonistic relation, so that the detection rate of defects is improved, the judgment dispute between good products and defective products is solved to a certain extent, and the misjudgment rate of products to be detected is reduced.
Fig. 6 is a schematic structural diagram of a defect detecting device according to an embodiment, as shown in fig. 6, the defect detecting device includes an image processing module 10, a first analysis module 11, a second analysis module 12, and a result analysis module 13;
the image processing module 10 is used for acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information;
the first analysis module 11 is configured to analyze the profile information by using an image analysis algorithm, so as to obtain a first detection result of the product to be detected;
the second analysis module 12 is configured to analyze the picture information by using an image deep learning algorithm, so as to obtain a second detection result of the product to be detected;
and the result analysis module 13 is used for obtaining a final detection result of the product to be detected according to the first detection result and the second detection result.
In order to implement the defect detection method of the foregoing embodiment, the defect detection device provided in this embodiment has similar implementation principles and technical effects to those of the foregoing embodiment, and will not be described herein.
Optionally, the image processing module 10 is specifically configured to obtain picture information; the picture information is preprocessed to obtain contour information, wherein the preprocessing comprises at least one of image graying processing, binarization processing, expansion processing and corrosion processing.
Optionally, the first analysis module 11 is specifically configured to divide the profile information into a defect area and a normal area by using an image analysis algorithm; calculating the area of the defect area and the area of the normal area respectively; if the ratio of the area of the defect area to the area of the normal area is greater than or equal to a preset threshold value, the first detection result is a defective product; if the ratio of the area of the defect area to the area of the normal area is smaller than a preset threshold value, the first detection result is good.
Optionally, the image deep learning algorithm includes an image deep learning strict model and an image deep learning loose model, the image deep learning strict model having a lower tolerance to defects than the image deep learning loose model.
Optionally, when the first detection result is a defective product, the second analysis module 12 is specifically configured to input the picture information into the image deep learning strict model to obtain a second detection result;
correspondingly, the result analysis module 13 is specifically configured to, if the first detection result is a defective product and the second detection result is a defective product, determine that the final detection result is a defective product; if the first detection result is defective and the second detection result is good, the final detection result is good.
Optionally, when the first detection result is good, the second analysis module 12 is specifically configured to input the picture information into the image deep learning loose model to obtain a second detection result;
correspondingly, the result analysis module 13 is specifically configured to, if the first detection result is good and the second detection result is good, determine that the final detection result is good; if the first detection result is good and the second detection result is defective, the final detection result is defective.
Optionally, referring to fig. 6, fig. 7 is a schematic structural diagram of another defect detecting device according to an embodiment, as shown in fig. 7, further including: a training module 14;
a training module 14 for acquiring historical image information; and training to obtain an image deep learning strict model and an image deep learning loose model according to the historical image information.
The embodiment of the invention provides a computer device, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the defect detection method provided by any embodiment of the application when executing the computer program:
acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information;
Analyzing the contour information by using an image analysis algorithm to obtain a first detection result of the product to be detected;
analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected;
and obtaining a final detection result of the product to be detected according to the first detection result and the second detection result.
In an embodiment, the processor when executing the computer program further performs the steps of:
obtaining image information of a product to be detected, including:
acquiring picture information;
the picture information is preprocessed to obtain contour information, wherein the preprocessing comprises at least one of image graying processing, binarization processing, expansion processing and corrosion processing.
In an embodiment, the processor when executing the computer program further performs the steps of:
analyzing the contour information by using an image analysis algorithm to obtain a first detection result of the product to be detected, wherein the first detection result comprises:
dividing the contour information into a defect area and a normal area by using an image analysis algorithm;
calculating the area of the defect area and the area of the normal area respectively;
if the ratio of the area of the defect area to the area of the normal area is greater than or equal to a preset threshold value, the first detection result is a defective product;
If the ratio of the area of the defect area to the area of the normal area is smaller than a preset threshold value, the first detection result is good.
In an embodiment, the processor when executing the computer program further performs the steps of:
the image deep learning algorithm comprises an image deep learning strict model and an image deep learning loose model, wherein the tolerance of the image deep learning strict model to the defects is lower than that of the image deep learning loose model.
In an embodiment, the processor when executing the computer program further performs the steps of:
when the first detection result is defective, analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected, wherein the second detection result comprises the following steps:
inputting the picture information into an image deep learning strict model to obtain a second detection result;
correspondingly, according to the first detection result and the second detection result, a final detection result of the product to be detected is obtained, which comprises the following steps:
if the first detection result is defective and the second detection result is defective, the final detection result is defective;
if the first detection result is defective and the second detection result is good, the final detection result is good.
In an embodiment, the processor when executing the computer program further performs the steps of:
when the first detection result is good, analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected, wherein the second detection result comprises the following steps:
inputting the picture information into an image deep learning loose model to obtain a second detection result;
correspondingly, according to the first detection result and the second detection result, a final detection result of the product to be detected is obtained, which comprises the following steps:
if the first detection result is good and the second detection result is good, the final detection result is good;
if the first detection result is good and the second detection result is defective, the final detection result is defective.
In an embodiment, the processor when executing the computer program further performs the steps of:
the defect detection method further comprises the following steps:
acquiring historical image information;
and training to obtain an image deep learning strict model and an image deep learning loose model according to the historical image information.
FIG. 8 is a schematic structural diagram of a computer device according to an embodiment, as shown in FIG. 8, the computer device includes a processor 30, a memory 31 and a communication interface 32; the number of processors 30 in the computer device may be one or more, one processor 30 being taken as an example in fig. 8; the processor 30, memory 31, communication interface 32 in the computer device may be connected by a bus or other means, for example in fig. 8. Bus means one or more of several types of bus structures including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures.
The memory 31 is provided as a computer readable storage medium, and may be configured to store a software program, a computer executable program, and modules, such as program instructions/modules corresponding to the methods in the embodiments of the present invention. The processor 30 executes at least one functional application of the computer device and data processing, i.e. the steps of implementing the above-described defect detection method, by running software programs, instructions and modules stored in the memory 31.
The memory 31 may include a storage program area that may store an operating system, at least one application program required for functions, and a storage data area; the storage data area may store data created according to the use of the computer device, etc. In addition, the memory 31 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some examples, memory 31 may include memory located remotely from processor 30, which may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The communication interface 32 may be configured to receive and transmit data.
In an embodiment, the embodiment of the present invention further provides a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements the steps of the defect detection method provided in any embodiment of the present application when executing the computer program.
In an embodiment, the present invention further provides a computer readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the defect detection method provided in any embodiment of the present application.
The embodiment of the invention also provides a computer readable storage medium, and the computer readable storage medium stores a computer program, and the computer program realizes the steps of the audio processing method provided by any embodiment of the application when being executed by a processor, and the method comprises the following steps:
acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information;
analyzing the contour information by using an image analysis algorithm to obtain a first detection result of the product to be detected;
analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected;
And obtaining a final detection result of the product to be detected according to the first detection result and the second detection result.
In an embodiment, the computer program when executed by the processor further performs the steps of:
obtaining image information of a product to be detected, including:
acquiring picture information;
the picture information is preprocessed to obtain contour information, wherein the preprocessing comprises at least one of image graying processing, binarization processing, expansion processing and corrosion processing.
In an embodiment, the computer program when executed by the processor further performs the steps of:
analyzing the contour information by using an image analysis algorithm to obtain a first detection result of the product to be detected, wherein the first detection result comprises:
dividing the contour information into a defect area and a normal area by using an image analysis algorithm;
calculating the area of the defect area and the area of the normal area respectively;
if the ratio of the area of the defect area to the area of the normal area is greater than or equal to a preset threshold value, the first detection result is a defective product;
if the ratio of the area of the defect area to the area of the normal area is smaller than a preset threshold value, the first detection result is good.
In an embodiment, the computer program when executed by the processor further performs the steps of:
The image deep learning algorithm comprises an image deep learning strict model and an image deep learning loose model, wherein the tolerance of the image deep learning strict model to the defects is lower than that of the image deep learning loose model.
In an embodiment, the computer program when executed by the processor further performs the steps of:
when the first detection result is defective, analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected, wherein the second detection result comprises the following steps:
inputting the picture information into an image deep learning strict model to obtain a second detection result;
correspondingly, according to the first detection result and the second detection result, a final detection result of the product to be detected is obtained, which comprises the following steps:
if the first detection result is defective and the second detection result is defective, the final detection result is defective;
if the first detection result is defective and the second detection result is good, the final detection result is good.
In an embodiment, the computer program when executed by the processor further performs the steps of:
when the first detection result is good, analyzing the picture information by utilizing an image deep learning algorithm to obtain a second detection result of the product to be detected, wherein the second detection result comprises the following steps:
Inputting the picture information into an image deep learning loose model to obtain a second detection result;
correspondingly, according to the first detection result and the second detection result, a final detection result of the product to be detected is obtained, which comprises the following steps:
if the first detection result is good and the second detection result is good, the final detection result is good;
if the first detection result is good and the second detection result is defective, the final detection result is defective.
In an embodiment, the computer program when executed by the processor further performs the steps of:
the defect detection method further comprises the following steps:
acquiring historical image information;
and training to obtain an image deep learning strict model and an image deep learning loose model according to the historical image information.
The computer storage media of embodiments of the invention may take the form of any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium can be, for example but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. Computer-readable storage media include (a non-exhaustive list): an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access Memory (Random Access Memory, RAM), a Read-Only Memory (ROM), an erasable programmable Read-Only Memory (electrically erasable, programmable Read-Only Memory, EPROM), a flash Memory, an optical fiber, a portable compact disc Read-Only Memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The computer readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with computer readable program code embodied in the data signal. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, radio Frequency (RF), etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++, ruby, go and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (Local Area Network, LAN) or a wide area network (Wide Area Network, WAN), or may be connected to an external computer, for example, through the internet using an internet service provider.
It will be appreciated by those skilled in the art that the term user terminal encompasses any suitable type of wireless user equipment, such as a mobile telephone, a portable data processing device, a portable web browser or a car mobile station.
In general, the various embodiments of the invention may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. For example, some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device, although the invention is not limited thereto.
Embodiments of the invention may be implemented by a data processor of a mobile device executing computer program instructions, e.g. in a processor entity, either in hardware, or in a combination of software and hardware. The computer program instructions may be assembly instructions, instruction set architecture (Instruction Set Architecture, ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages.
The block diagrams of any of the logic flows in the figures of this invention may represent program steps, or may represent interconnected logic circuits, modules, and functions, or may represent a combination of program steps and logic circuits, modules, and functions. The computer program may be stored on a memory. The memory may be of any type suitable to the local technical environment and may be implemented using any suitable data storage technology, such as, but not limited to, read Only Memory (ROM), random Access Memory (RAM), optical storage devices and systems (digital versatile disk DVD or CD optical disk), etc. The computer readable medium may include a non-transitory storage medium. The data processor may be of any type suitable to the local technical environment, such as, but not limited to, general purpose computers, special purpose computers, microprocessors, digital signal processors (Digital Signal Processing, DSPs), application specific integrated circuits (Application Specific Integrated Circuit, ASICs), programmable logic devices (Field-Programmable Gate Array, FGPA), and processors based on a multi-core processor architecture.
Claims (10)
1. A defect detection method, characterized in that the defect detection method comprises:
acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information;
analyzing the contour information by using an image analysis algorithm, and determining the area of a defect area in the contour information to obtain a first detection result of the product to be detected, wherein the first detection result is related to the area of the defect area;
analyzing the picture information by using a model based on an image deep learning algorithm to obtain a second detection result of the product to be detected;
and obtaining a final detection result of the product to be detected according to the first detection result and the second detection result.
2. The defect detection method according to claim 1, wherein the acquiring image information of the product to be detected includes:
acquiring the picture information;
and preprocessing the picture information to obtain the profile information, wherein the preprocessing comprises at least one of image graying processing, binarization processing, expansion processing and corrosion processing.
3. The defect detection method according to claim 1, wherein the analyzing the profile information by using an image analysis algorithm to obtain the first detection result of the product to be detected comprises:
Dividing the contour information into a defect area and a normal area by using an image analysis algorithm;
calculating the area of the defect area and the area of the normal area respectively;
if the ratio of the area of the defect area to the area of the normal area is greater than or equal to a preset threshold value, the first detection result is a defective product;
if the ratio of the area of the defect area to the area of the normal area is smaller than the preset threshold, the first detection result is good.
4. The defect detection method of claim 1, wherein the image deep learning algorithm-based models include an image deep learning strict model and an image deep learning loose model, the image deep learning strict model having a lower tolerance to defects than the image deep learning loose model.
5. The defect detection method according to claim 4, wherein when the first detection result is a defective product, the analyzing the picture information by using an image deep learning algorithm to obtain a second detection result of the product to be detected includes:
inputting the picture information into the image deep learning strict model to obtain the second detection result;
Correspondingly, the obtaining the final detection result of the product to be detected according to the first detection result and the second detection result includes:
if the first detection result is a defective product and the second detection result is a defective product, the final detection result is a defective product;
and if the first detection result is a defective product and the second detection result is a good product, the final detection result is a good product.
6. The method for detecting defects according to claim 4, wherein when the first detection result is good, the analyzing the picture information by using an image deep learning algorithm to obtain a second detection result of the product to be detected comprises:
inputting the picture information into the image deep learning loose model to obtain the second detection result;
correspondingly, the obtaining the final detection result of the product to be detected according to the first detection result and the second detection result includes:
if the first detection result is good and the second detection result is good, the final detection result is good;
and if the first detection result is good and the second detection result is defective, the final detection result is defective.
7. The defect detection method of claim 4, wherein the defect detection method further comprises:
acquiring historical image information;
and training to obtain the image deep learning strict model and the image deep learning loose model according to the historical image information.
8. The defect detection device is characterized by comprising an image processing module, a first analysis module, a second analysis module and a result analysis module;
the image processing module is used for acquiring image information of a product to be detected, wherein the image information comprises picture information and contour information;
the first analysis module is used for analyzing the contour information by utilizing an image analysis algorithm, determining the area of a defect area in the contour information, and obtaining a first detection result of the product to be detected, wherein the first detection result is related to the area of the defect area;
the second analysis module is used for analyzing the picture information by using a model based on an image deep learning algorithm to obtain a second detection result of the product to be detected;
the result analysis module is used for obtaining a final detection result of the product to be detected according to the first detection result and the second detection result.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the defect detection method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor realizes the steps of the defect detection method according to any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010377497.4A CN111583223B (en) | 2020-05-07 | 2020-05-07 | Defect detection method, defect detection device, computer equipment and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010377497.4A CN111583223B (en) | 2020-05-07 | 2020-05-07 | Defect detection method, defect detection device, computer equipment and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111583223A CN111583223A (en) | 2020-08-25 |
| CN111583223B true CN111583223B (en) | 2023-12-19 |
Family
ID=72112559
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010377497.4A Active CN111583223B (en) | 2020-05-07 | 2020-05-07 | Defect detection method, defect detection device, computer equipment and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111583223B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112505049B (en) * | 2020-10-14 | 2021-08-03 | 上海互觉科技有限公司 | Mask inhibition-based method and system for detecting surface defects of precision components |
| CN112669271A (en) * | 2020-12-22 | 2021-04-16 | 深圳酷派技术有限公司 | Object surface defect detection method, related device and computer storage medium |
| CN112611757A (en) * | 2020-12-23 | 2021-04-06 | 深兰人工智能芯片研究院（江苏）有限公司 | AI auxiliary detection method, device, electronic equipment and computer readable storage medium |
| CN112785556A (en) * | 2020-12-31 | 2021-05-11 | 深兰人工智能芯片研究院（江苏）有限公司 | Reinspection method, reinspection device, electronic equipment and computer-readable storage medium |
| CN113066079A (en) * | 2021-04-19 | 2021-07-02 | 北京滴普科技有限公司 | Method, system and storage medium for automatically detecting wood defects |
| CN113379728A (en) * | 2021-07-02 | 2021-09-10 | 上海电气集团股份有限公司 | Method, system, equipment and readable storage medium for detecting defects on surface of rail |
| CN113706465B (en) * | 2021-07-22 | 2022-11-15 | 杭州深想科技有限公司 | Pen defect detection method based on deep learning, computing equipment and storage medium |
| CN114841915A (en) * | 2022-03-14 | 2022-08-02 | 阿里巴巴（中国）有限公司 | Tile flaw detection method and system based on artificial intelligence and storage medium |
| CN114998192B (en) * | 2022-04-19 | 2023-05-30 | 深圳格芯集成电路装备有限公司 | Defect detection method, device, equipment and storage medium based on deep learning |
| CN114881996A (en) * | 2022-05-25 | 2022-08-09 | 京东方科技集团股份有限公司 | Defect detection method and device |
| CN114782451B (en) * | 2022-06-23 | 2022-09-16 | 季华实验室 | Workpiece defect detection method and device, electronic equipment and readable storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019233166A1 (en) * | 2018-06-04 | 2019-12-12 | 杭州海康威视数字技术股份有限公司 | Surface defect detection method and apparatus, and electronic device |
| CN110992329A (en) * | 2019-11-28 | 2020-04-10 | 上海微创医疗器械（集团）有限公司 | Product surface defect detection method, electronic device and readable storage medium |
| CN111044522A (en) * | 2019-12-14 | 2020-04-21 | 中国科学院深圳先进技术研究院 | Defect detection method and device and terminal equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10395362B2 (en) * | 2017-04-07 | 2019-08-27 | Kla-Tencor Corp. | Contour based defect detection |
-
2020
- 2020-05-07 CN CN202010377497.4A patent/CN111583223B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019233166A1 (en) * | 2018-06-04 | 2019-12-12 | 杭州海康威视数字技术股份有限公司 | Surface defect detection method and apparatus, and electronic device |
| CN110992329A (en) * | 2019-11-28 | 2020-04-10 | 上海微创医疗器械（集团）有限公司 | Product surface defect detection method, electronic device and readable storage medium |
| CN111044522A (en) * | 2019-12-14 | 2020-04-21 | 中国科学院深圳先进技术研究院 | Defect detection method and device and terminal equipment |
Non-Patent Citations (1)
| Title |
|---|
| 苗永菲 ; 游洋 ; 李赵松 ; 黎红军 ; 宋康 ; 侯朝云 ; .基于机器视觉的玻璃缺陷检测技术.电子设计工程.2020,(08),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111583223A (en) | 2020-08-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111583223B (en) | Defect detection method, defect detection device, computer equipment and computer readable storage medium | |
| CN109142366B (en) | Spherical shell defect detection method and device and computer readable storage medium | |
| CN108280822B (en) | Screen scratch detection method and device | |
| US10740912B2 (en) | Detection of humans in images using depth information | |
| CN107273838B (en) | Processing method and device for snapshot pictures of traffic signal lamps | |
| US20140286527A1 (en) | Systems and methods for accelerated face detection | |
| CN112949767B (en) | Sample image increment, image detection model training and image detection method | |
| CN112116542B (en) | Image contrast enhancement method, device, electronic equipment and storage medium | |
| JP2014119363A (en) | Automatic unevenness detecting device and automatic unevenness detecting method for flat-panel display | |
| CN114862817A (en) | Circuit board golden finger area defect detection method, system, device and medium | |
| CN111626967A (en) | Image enhancement method, image enhancement device, computer device and readable storage medium | |
| CN113785181A (en) | OLED screen point defect judgment method and device, storage medium and electronic equipment | |
| CN113039577A (en) | Product testing method and device, computer readable storage medium and electronic equipment | |
| CN114298985B (en) | Defect detection method, device, equipment and storage medium | |
| CN112312001A (en) | Image detection method, device, equipment and computer storage medium | |
| CN110310341B (en) | Method, device, equipment and storage medium for generating default parameters in color algorithm | |
| CN112883762A (en) | Living body detection method, device, system and storage medium | |
| CN113628192B (en) | Image blur detection method, apparatus, device, storage medium, and program product | |
| CN114219760A (en) | Reading identification method and device of instrument and electronic equipment | |
| CN116993654A (en) | Camera module defect detection method, device, equipment, storage medium and product | |
| CN115131340A (en) | Power plant pulverized coal leakage identification method, device and equipment and scale storage medium | |
| CN108447107B (en) | Method and apparatus for generating video | |
| CN114937003A (en) | Multi-type defect detection system and method for glass panel | |
| CN110211085B (en) | Image fusion quality evaluation method and system | |
| CN111062984A (en) | Method, device and equipment for measuring area of video image region 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 |