WO2023179122A1 - 缺陷检测方法、装置、电子设备及可读存储介质 - Google Patents

缺陷检测方法、装置、电子设备及可读存储介质 Download PDF

Info

Publication number
WO2023179122A1
WO2023179122A1 PCT/CN2022/140036 CN2022140036W WO2023179122A1 WO 2023179122 A1 WO2023179122 A1 WO 2023179122A1 CN 2022140036 W CN2022140036 W CN 2022140036W WO 2023179122 A1 WO2023179122 A1 WO 2023179122A1
Authority
WO
WIPO (PCT)
Prior art keywords
area
image
defect
detected
contour
Prior art date
Application number
PCT/CN2022/140036
Other languages
English (en)
French (fr)
Inventor
李明亮
曾苏珊
杜兵
冯英俊
Original Assignee
广东利元亨智能装备股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 广东利元亨智能装备股份有限公司 filed Critical 广东利元亨智能装备股份有限公司
Publication of WO2023179122A1 publication Critical patent/WO2023179122A1/zh

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
    • 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
    • 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

Definitions

  • This application relates to the field of quality detection, specifically, to a defect detection method, device, electronic equipment and readable storage medium.
  • die-cutting materials With the development of intelligence and electronics, the application and development of die-cutting materials have also developed rapidly. Based on the rapid development of die-cutting modules and the increasingly important role of die-cutting materials in various fields, the detection of die-cutting materials has become very important, and the accuracy requirements for module material detection are also getting higher and higher. However, the current detection of die-cut materials is mainly based on manual detection, which is greatly affected by subjective factors.
  • the purpose of embodiments of the present application is to provide a defect detection method, device, electronic equipment and readable storage medium.
  • the image to be processed can be processed to detect the die-cut material.
  • embodiments of the present application provide a defect detection method, which includes: extracting the contour of the object to be detected on the image to be processed to obtain a defect map of the object to be detected; performing defect contour extraction on the defect map to obtain the object to be detected The defect area of the object is compared with the first area threshold. When the defect area is greater than the first area threshold, it is determined that the object to be detected has a defect.
  • the contours of the object to be detected and the defect can be obtained, and then the defect map can be obtained. Then, based on the defect map, the contour of the defect is further extracted and the contour area of the defect is obtained. The contour area of the defect is compared with the first area threshold. When the contour area of the defect is greater than the first area threshold, the object to be detected is determined. Flawed. Since the contour of the object to be detected is determined by comparing it with the set area threshold, on the one hand, larger defects can be effectively detected, and on the other hand, smaller defects can be ignored, achieving accurate and effective defect identification. Moreover, the entire process of defect identification is processed based on the image of the object to be detected, so detection automation is achieved, detection efficiency is improved, and the work intensity of the staff is reduced.
  • extracting the outline of the object to be detected on the image to be processed to obtain a defect map includes: extracting the image to be processed Binary processing to obtain a binary image; obtain the effective contour of the object to be detected in the binary image, and the effective contour of the object to be detected is the contour of the object to be detected that satisfies the preset conditions; from the Extract the target image from the effective contour; perform convolution calculation on the target image according to the set algorithm to obtain the defect map of the object to be detected.
  • a binary image that can reflect the overall and local characteristics of the image is obtained, without involving multiple extreme values of pixels, making the processing simple.
  • conditions are set for the contour, and the contours that meet the conditions are selected as effective contours, which eliminates some interfering contours and ensures the accuracy of contour acquisition.
  • the defect map obtained by convolution calculation of the target image is also more accurate, which improves the accuracy of obtaining defects.
  • the embodiment of the present application provides the second possible implementation manner of the first aspect, wherein: obtaining the effective contour of the object to be detected in the binary image,
  • the method includes: extracting a contour set of the object to be detected in the binary image, where the contour set of the object to be detected includes one or more contours of the object to be detected; and eliminating the contour set of the object to be detected that does not meet the preset conditions.
  • the contour of the object to be detected is obtained to obtain the effective contour of the object to be detected.
  • the contour area is smaller than the preset condition.
  • the conditional contour is regarded as an interference contour (that is, the contour of the non-detected object body).
  • embodiments of the present application provide a third possible implementation of the first aspect, wherein the intercepting the target image according to the effective contour of the object to be detected includes: Fit the effective contour of the object to be detected to a first target graphic; map the coordinates of the first target graphic with the coordinates of the image to be processed to obtain a second target graphic; perform Screenshot processing to obtain the target image.
  • the area and position points of the effective contour fitting are more closely aligned with the actual position of the object to be detected, ensuring that the target image is more accurate. It fits the actual situation and improves the accuracy of the target image.
  • embodiments of the present application provide a fourth possible implementation manner of the first aspect, wherein the obtaining the effective contour of the object to be detected in the binary image,
  • the method includes: performing contour extraction on the binary image to obtain the overall contour of the object to be detected; obtaining a coarse area image of the image to be processed according to the overall contour of the object to be detected; and obtaining an image of the object to be detected in the coarse area image of the image to be processed. Effective outline.
  • embodiments of the present application provide a fifth possible implementation manner of the first aspect, wherein the binary image includes a first target area binary image, so
  • the overall contour of the object to be detected includes the overall contour of the first target area
  • the coarse area image of the image to be processed includes the coarse area image of the first target area
  • the coarse area of the image to be processed is obtained according to the overall outline of the object to be detected.
  • the image includes: performing contour extraction on the binarized image of the first target area to obtain the overall contour of the first target area; fitting the overall contour of the first target area into a third target graphic; and performing the contour extraction on the first target area according to the set rules.
  • the three-target graphics are expanded to obtain a coarse-region binarized image of the first target region; and the coarse-region image of the first target region is intercepted according to the coarse-region binarized image of the first target region.
  • embodiments of the present application provide a sixth possible implementation manner of the first aspect, wherein the binary image includes a second target area binarized image, so
  • the overall contour of the object to be detected includes the overall contour of the second target area
  • the coarse area image of the image to be processed includes the coarse area image of the second target area
  • the coarse area of the image to be processed is obtained according to the overall outline of the object to be detected.
  • the image includes: performing XOR processing on the coarse area binarized image of the first target area and the overall image of the object to be detected to obtain a second target area coarse area image.
  • embodiments of the present application provide a seventh possible implementation manner of the first aspect, wherein the defect area includes a defective area located in the first target area of the image to be processed.
  • One or more first defect areas and one or more second defect areas located in the second target area of the image to be processed comparing the defect area with an area threshold, when the defect area is larger than the A first area threshold value, determining that the object to be detected has a defect includes: comparing one or more first defect areas with the first area threshold; comparing one or more second defect areas with the First area threshold comparison; if the first defect area is greater than the area threshold, or the second defect area is greater than the first area threshold, it indicates that the object to be detected is defective.
  • the defect areas in different areas of the object to be inspected are compared with the area thresholds. When any defect area is greater than the area threshold, it can be determined that the object to be inspected has a defect. Through comparison in all aspects, comprehensive defect detection is achieved and the accuracy of defect detection is improved.
  • embodiments of the present application also provide a defect detection device, including: a first contour extraction module: used to extract the contour of the object to be detected on the image to be processed, and obtain a defect map of the object to be detected; a second contour extraction module: Used to extract defect contours from the defect map to obtain the defect area of the object to be detected; Comparison module: used to compare the defect area with a first area threshold, when the defect area is greater than the first area threshold , it is determined that the object to be inspected is defective.
  • embodiments of the present application further provide an electronic device, including: a processor and a memory.
  • the memory stores machine-readable instructions executable by the processor.
  • the electronic device is running, the machine-readable instructions are stored in the memory.
  • the instructions are executed by the processor, the steps of the method in the above-mentioned first aspect or any possible implementation of the first aspect are performed.
  • embodiments of the present application further provide a computer-readable storage medium.
  • the computer-readable storage medium stores a computer program.
  • the computer program executes the above-mentioned first aspect, or any aspect of the first aspect. Steps of a defect detection method in a possible implementation.
  • Figure 1 is a block diagram of an electronic device provided by an embodiment of the present application.
  • Figure 2 is a flow chart of a defect detection method provided by an embodiment of the present application.
  • Figure 3 is a schematic diagram of the outline of an image to be processed provided by an embodiment of the present application.
  • Figure 4 is a flow chart of step 201 in the defect detection method provided by the embodiment of the present application.
  • Figure 5 is a schematic diagram of the partial composition of the image outline to be processed provided by the embodiment of the present application.
  • Figure 6 is a schematic diagram of the functional modules of the defect detection device provided by the embodiment of the present application.
  • die-cutting materials play a crucial role in various industries, the quality of die-cutting materials affects the quality of the entire equipment. Based on this, the detection of die-cut materials has also become very important. However, the current detection of die-cut materials still relies on manual detection. Human detection will be affected by subjective emotions, and people's energy is limited. Long-term work may cause lack of energy, and then the detection efficiency will be reduced, and the accuracy of detection cannot be guaranteed. and efficiency.
  • the inventor of the present application discovered during the process of detecting die-cut materials: If you want to detect whether there are defects in the die-cut materials, you can collect images of the film-cut materials, and determine the defect contour and die-cut materials based on the images of the die-cut materials. The body contour is extracted and the contour area is used to determine whether there are defects in the die-cut material.
  • the inventor of the present application proposes a defect detection method that can detect the object to be detected by processing the image of the object to be detected, replacing manual detection, making the detection more objective, accurate, and more efficient.
  • the detection methods disclosed in the embodiments of this application can be, but are not limited to, used for the detection of die-cut materials, chips, electronic screens, and glass materials.
  • the detection device of the present application can be used to realize defect detection of various objects to be detected.
  • the electronic device 100 may include a memory 111, a processor 113, and an input-output unit 115.
  • the electronic device 100 may include a memory 111, a processor 113, and an input-output unit 115.
  • FIG. 1 is only illustrative and does not limit the structure of the electronic device 100 .
  • electronic device 100 may also include more or fewer components than shown in FIG. 1 , or have a different configuration than shown in FIG. 1 .
  • the above-mentioned components of the memory 111, the processor 113 and the input/output unit 115 are directly or indirectly electrically connected to each other to realize data transmission or interaction.
  • these components may be electrically connected to each other through one or more communication buses or signal lines.
  • the above-mentioned processor 113 is used to execute executable modules stored in the memory.
  • the memory 111 can be, but is not limited to, random access memory (Random Access Memory, referred to as RAM), read-only memory (Read Only Memory, referred to as ROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM) ), Erasable Programmable Read-Only Memory (EPROM for short), Electrically Erasable Programmable Read-Only Memory (EEPROM for short), etc.
  • RAM Random Access Memory
  • ROM Read Only Memory
  • PROM programmable read-only memory
  • EPROM Erasable Programmable Read-Only Memory
  • EEPROM Electrically Erasable Programmable Read-Only Memory
  • the above-mentioned memory 111 can be used to store information such as the image to be processed, the first area threshold, the second area threshold, etc.
  • the above-mentioned memory 111 can also be used to temporarily store intermediate information such as target images, binarized images, and coarse area images.
  • the above-mentioned processor 113 may be an integrated circuit chip with signal processing capabilities.
  • the above-mentioned processor 113 can be a general-purpose processor, including a central processing unit (Central Processing Unit, referred to as CPU), a network processor (Network Processor, referred to as NP), etc.; it can also be a digital signal processor (Digital Signal Processor, referred to as DSP) ), Application Specific Integrated Circuit (ASIC for short), Field Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
  • CPU Central Processing Unit
  • NP Network Processor
  • DSP Digital Signal Processor
  • ASIC Application Specific Integrated Circuit
  • FPGA Field Programmable Gate Array
  • a general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.
  • the above-mentioned processor 113 is used to complete a series of actions such as contour extraction of the image to be processed, binarization of the image to be processed, filtering of contours, comparison of contour area and area threshold, fitting of effective contours, and mapping of target graphics.
  • the above-mentioned input and output unit 115 is used to provide input data to the user.
  • the input and output unit 115 may be, but is not limited to, a mouse, a keyboard, an interactive interface, etc.
  • the above-mentioned input and output unit 115 may be used to input data such as the first area threshold and the second area threshold. This data can be set according to different objects to be detected. When the object to be detected changes, the first area threshold and the preset condition can be modified accordingly through the above-mentioned input and output unit 115.
  • the electronic device 100 in this embodiment can be used to perform each step in each method provided by the embodiment of this application.
  • the implementation process of the defect detection method is described in detail below through several embodiments.
  • Figure 2 is a flow chart of a defect detection method provided by an embodiment of the present application. The specific process shown in Figure 2 will be elaborated below.
  • Step 201 Extract the contour of the object to be detected from the image to be processed to obtain a defect map of the object to be detected.
  • the above-mentioned image to be processed is a collected image of the object to be detected.
  • the image to be processed can be obtained through a collection device.
  • the first target area contour, the second target contour area, the defect contour, the irrelevant contour area, etc. of the device to be detected can be obtained.
  • the irrelevant contour area is a non-body area of the object to be detected.
  • it may be the outline of a platform on which the object to be detected is placed, or the outline of impurities around the object to be detected, etc.
  • the dotted circles in the figure can be regarded as irrelevant contours
  • the solid squares are the body contours of the object to be detected
  • the filled circles are the defect contours.
  • the image of the object to be detected can also be scaled to obtain the image to be processed.
  • the scaling process may use a bilinear interpolation algorithm, the scaling process may use a nearest neighbor interpolation algorithm, etc. By scaling the image of the object to be detected, the image can be processed quickly using less resources.
  • Step 202 Perform defect contour extraction on the defect map to obtain the defect area of the object to be detected.
  • the defect map may include defect contours and irrelevant contours, or the defect map may only include defect contours. If the defect map includes defect contours and irrelevant contours, the defect contours and irrelevant contours need to be extracted. If the defect map only includes defect contours, the defect contours need to be extracted.
  • the defect area can be calculated based on the length and width of the defect outline, and the defect area can also be calculated based on the radius of the defect outline.
  • Step 203 Compare the defect area with the first area threshold. When the defect area is greater than the first area threshold, it is determined that the object to be inspected has a defect.
  • the first area threshold is the maximum area value of allowed defects, and the first area threshold can be adjusted accordingly according to different objects to be detected.
  • step 201 includes: steps 2011-2013.
  • Step 2011 Binarize the image to be processed to obtain a binarized image.
  • the binarization process of the image is to set the gray value of the points on the image to 0 or 255, that is, the entire image presents an obvious black and white effect, that is, the grayscale image of multiple brightness levels is obtained through appropriate threshold selection
  • After the binarized image is obtained by binarizing the image, it is beneficial to further process the image.
  • the properties of the image are only related to the position of the point with a pixel value of 0 or 255, and no longer involve the multi-level values of the pixel. Makes processing easy.
  • each pixel point of the image to be processed is divided to obtain a binary image.
  • Step 2012 Obtain the effective contour of the object to be detected in the binary image.
  • the effective contour of the object to be detected is the contour that satisfies the preset conditions among the contours of the object to be detected.
  • the effective contour may include the body contour of the object to be detected and the defect contour.
  • the effective contour may also include only the body contour of the object to be detected.
  • the preset condition is: a contour with a contour area greater than 5 is a valid contour
  • contour A and contour E can be eliminated to obtain effective contours such as contour B, contour C, and contour D.
  • Step 2013 Extract the target image from the effective contour.
  • the target image may be an overall image of the processed object to be detected, and the target image may be one or more partial images of the processed object to be detected.
  • the target image can be extracted directly from the effective contour; the effective contour can also be fitted, and the image can be extracted from the fitted effective contour; the effective contour can also be mapped, and the mapped effective contour can be extracted. Extract images from contours.
  • Step 2014 Perform convolution calculation on the target image according to the set algorithm to obtain a defect map of the object to be inspected.
  • the convolution calculation on the target image according to the set algorithm can be processed based on Tensorflow, PyTorch, etc.
  • the defect map may be an overall defect map of the object to be inspected, or the defect map may be a partial defect map of the object to be inspected.
  • the object to be detected is a battery
  • the battery is the entire object to be detected
  • the tabs and the cell body are parts of the object to be detected.
  • the defect map may include an overall defect map of the battery
  • the defect map may include a tab defect map
  • the defect map may include a cell body defect map
  • the defect map may also include a tab defect map and a cell body defect map.
  • the object to be detected is a chip
  • the chip is the entire object to be detected
  • the main board and pins are parts of the object to be detected.
  • the defect map may include an overall chip defect map
  • the defect map may include a motherboard defect map
  • the defect map may include a pin defect map
  • the defect map may also include a motherboard defect map and a pin defect map.
  • step 2012 includes: extracting the contour set of the object to be detected in the binary image, eliminating the contours of the object to be detected that do not meet the preset conditions in the contour set of the object to be detected, and obtaining the object to be detected. effective outline.
  • the profile set of the object to be detected includes one or more profiles of the object to be detected.
  • the contour set of the object to be detected may include the body contour of the object to be detected, the partial contour of the object to be detected, the contour of internal defects of the object to be detected, or other interference contours, etc.
  • the preset condition may be that the contour area is greater than the second area threshold, the contour area is less than the second area threshold, or the contour area does not belong to the second area threshold range, and the second area threshold can be used to eliminate interference contours,
  • the second area threshold can also be used to eliminate defective contours, and the second area threshold can be used to eliminate interference contours and defective contours at the same time.
  • the second area threshold is used to eliminate interference contours, and the preset condition can be set to a contour area greater than 4, and the dotted circle outline can be eliminated according to the preset condition.
  • the second area threshold is used to eliminate interference contours and defective contours at the same time.
  • the preset condition can be set to a contour area greater than 8, and the dotted circle outline and the filled circle outline can be eliminated according to the preset condition.
  • the preset condition may also be whether the contour position is at a target position, and the target position may be the position of the body contour of the object to be detected.
  • the contour position is within the target position range, it means that the contour position meets the preset conditions.
  • the contour position is not within the target position range, it means that the contour position does not meet the preset conditions.
  • the solid square outline is the position of the body outline of the object to be detected, then it can be determined that the dotted circle outline does not meet the preset conditions, and the dotted circle outline can be eliminated.
  • step 2012 includes: fitting the effective contour of the object to be detected to a first target graphic; mapping the coordinates of the first target graphic with the coordinates of the image to be processed to obtain a second target graphic ; Perform screenshot processing on the second target graphic to obtain the target image.
  • the first target graphic may be a rectangle, the first target graphic may be a circle, the first target graphic may also be a triangle, etc.
  • the second target graphic may be a rectangle, the second target graphic may be a circle, the second target graphic may also be a triangle, etc.
  • the first target graph fitting can be done with contours via Opencv's boundingRect or minAreaRect constructors.
  • mapping the coordinates of the first target graphic to the coordinates of the image to be processed can scale the coordinates. Mapping the first target graphics coordinates with the image coordinates to be processed can also be achieved by setting the corresponding model for processing.
  • step 2012 includes: performing contour extraction on the binary image to obtain the overall contour of the object to be detected; obtaining the coarse area image of the image to be processed according to the overall contour of the object to be detected; obtaining the image of the coarse area of the image to be processed. The effective contour of the object to be detected in the coarse area image.
  • the object to be detected can be regarded as composed of multiple partial components.
  • the binarized image includes a first target area binarized image, a second target area binarized image, a third target area binarized image, etc.
  • the overall contour of the object to be detected includes the overall contour of the first target area, the overall contour of the second target area, the overall contour of the third target area, etc.
  • the coarse area image of the image to be processed includes a coarse area image of the first target area, a coarse area image of the second target area, a coarse area image of the third target area, and so on.
  • the defect area includes one or more first defect areas in the first target area and one or more second defect areas in the second target area.
  • the object to be detected can be regarded as consisting of a first target area 41 and a second target area 42 .
  • the first defect area includes 411, 412 and 423
  • the second defect area includes 420, 421, 422 and 423.
  • obtaining the effective contour of the object to be detected in the coarse area image of the image to be processed may include: performing threshold segmentation on the coarse area image of the image to be processed to obtain a binarized image of the coarse area of the image to be processed. Contour extraction is performed on the binarized image of the rough area to obtain a contour set of the object to be detected. The contours of the object to be detected that do not meet the preset conditions are eliminated from the contour set of the object to be detected, and an effective contour of the object to be detected is obtained.
  • obtaining the coarse area image of the image to be processed based on the overall contour of the object to be detected includes: performing contour extraction on the binarized image of the first target area to obtain the overall contour of the first target area. Fit the overall contour of the first target area to the third target graphic.
  • the third target graphic is expanded according to the set rules to obtain a coarse area binarized image of the first target area; a coarse area image of the first target area is intercepted according to the coarse area binarized image of the first target area.
  • the first target area may be the tab.
  • Obtaining the coarse area image of the image to be processed based on the overall contour of the object to be detected may include: performing contour extraction on the binarized image of the pole to obtain the overall contour of the pole. Fit the overall contour of the pole lug to the third target graphic.
  • the third target graphic is expanded according to the set rules to obtain a coarse area binarized image of the pole; the coarse area image of the pole is intercepted according to the coarse area binarized image of the pole.
  • the first target area may be a pin.
  • Obtaining the coarse area image of the image to be processed based on the overall contour of the object to be detected may include: performing contour extraction on the binary image of the pin to obtain the overall contour of the pin. Fit the overall outline of the pin to the third target shape.
  • the third target graphic is expanded according to the set rules to obtain a coarse area binarized image of the pin; the coarse area image of the pin is intercepted according to the coarse area binarized image of the pin.
  • the third target graphic may be a rectangle, the third target graphic may be a circle, the third target graphic may also be a triangle, etc.
  • the setting rule can be set to expand in a specified direction
  • the specified direction can be set to expand toward the Y coordinate axis
  • the specified direction can be set to expand toward the X coordinate axis.
  • obtaining the coarse area image of the image to be processed according to the overall contour of the object to be detected includes: performing XOR processing on the coarse area binarized image of the first target area and the overall image of the object to be detected, so as to Obtain the coarse area image of the second target area.
  • performing XOR processing on the coarse area binarized image of the first target area and the overall image of the object to be detected to obtain the coarse area image of the second target area may include: eliminating the first target area from the overall image of the object to be detected.
  • the coarse area binarization image to obtain the second target area coarse area image may include: eliminating the first target area from the overall image of the object to be detected.
  • the second target area may be the battery cell body.
  • Performing XOR processing on the coarse area binarized image of the first target area and the overall image of the object to be detected to obtain the coarse area image of the second target area may include: excluding the coarse area binarized image of the tabs from the overall battery image, To obtain the rough area image of the cell body.
  • the second target area may be the main board.
  • Performing XOR processing on the coarse area binarized image of the first target area and the overall image of the object to be detected to obtain the coarse area image of the second target area may include: removing the coarse area binarized image of the pin from the overall image of the chip, To get the rough area image of the motherboard.
  • obtaining a coarse area image of the image to be processed according to the overall contour of the object to be detected includes: binarizing the coarse area images of the first target area and the third target area Perform XOR processing with the entire image of the object to be detected to obtain the coarse area image of the second target area.
  • obtaining a coarse area image of the image to be processed according to the overall contour of the object to be detected includes: combining the first target area, the third target area and the third target area.
  • the coarse area binarized images of the four target areas are XORed with the overall image of the object to be detected to obtain the coarse area image of the second target area.
  • comparing the defect area with an area threshold, and determining that the object to be inspected has a defect when the defect area is greater than the first area threshold includes: comparing one or more first defect areas with the first area threshold Compare. The one or more second defect areas are compared to the first area threshold. If there is a first defect area greater than the first area threshold, or a second defect area greater than the first area threshold, it means that the object to be detected has a defect.
  • first defect areas and four second defect areas are shown.
  • the three first defect areas and the four second defect areas are compared with the first area threshold respectively. If at least one of the first defect area or the second defect area is larger than the first area threshold, it can be determined that the area to be detected is There are defects in the object.
  • the areas of the first defect areas 411, 412, and 413 in Figure 5 are 8 square millimeters, 6 square millimeters, and 3 square millimeters respectively, and the second defect areas 420, 421, 422, and 423 are respectively The areas are 4 square millimeters, 7 square millimeters, 10 square millimeters, and 13 square millimeters respectively. If the first area threshold at this time is 15 square millimeters, since all defect areas in the first defect area and the second defect area are smaller than the first area threshold, it can be determined that there is no defect in the object to be detected.
  • the areas of the first defect areas 411, 412, and 413 in Figure 5 are 8 mm2, 6 mm2, and 3 mm2 respectively, and the areas of the second defect areas 420, 421, 422, and 423 are 4 mm2, 7 mm2, respectively.
  • the object to be detected also includes a third target area and a fourth target area, then if the first defect area is greater than the area threshold, or the second defect area is greater than the first area threshold, or the third defect area is greater than the When the first area threshold value or the fourth defect area is greater than the first area threshold value, it indicates that the object to be detected has a defect.
  • an image to be processed is obtained by performing a series of processes on the image of the object to be detected, and the image to be processed is subjected to processing such as contour extraction, contour screening, and defect identification to detect defects in the object to be detected. Processing and inspection based on this image can more objectively identify defects in the object to be inspected.
  • the processing method in this application is to perform a series of processing and recognition on the image. These processes are all implemented based on automated equipment such as computers, which greatly reduces the work intensity of the staff and improves the detection efficiency.
  • the embodiment of the present application also provides a defect detection device corresponding to the defect detection method. Since the principle of solving the problem of the device in the embodiment of the present application is similar to that of the aforementioned defect detection method embodiment, in this embodiment
  • the principle of solving the problem of the device in the embodiment of the present application is similar to that of the aforementioned defect detection method embodiment, in this embodiment
  • For the implementation of the device please refer to the description in the embodiments of the above method, and repeated details will not be described again.
  • FIG. 6 is a schematic diagram of the functional modules of the defect detection device provided by the embodiment of the present application.
  • Each module in the defect detection device in this embodiment is used to perform each step in the above method embodiment.
  • the defect detection device includes a first contour extraction module 301, a second contour extraction module 302, and a comparison module 303; wherein,
  • the first contour extraction module 301 is used to extract the contour of the object to be detected from the image to be processed, and obtain a defect map of the object to be detected.
  • the second contour extraction module 302 is used to extract defect contours from the defect map to obtain the defect area of the object to be detected.
  • the comparison module 303 is used to compare the defect area with the first area threshold. When the defect area is greater than the first area threshold, it is determined that the object to be detected has a defect.
  • the second contour extraction module 302 is also used to: perform binary processing on the image to be processed to obtain a binary image; and obtain the effective contour of the object to be detected in the binary image, which is
  • the effective contour of the detection object is the contour of the object to be detected that meets the preset conditions; the target image is extracted from the effective contour; the target image is convolved according to the set algorithm to obtain the defect map of the object to be detected.
  • the second contour extraction module 302 is specifically configured to: extract a contour set of the object to be detected in the binary image, where the contour set of the object to be detected includes one or more contours of the object to be detected; The contours of the object to be detected that do not meet the preset conditions are eliminated from the contour set of the object to be detected, and an effective contour of the object to be detected is obtained.
  • the second contour extraction module 302 is specifically configured to: fit the effective contour of the object to be detected to the first target graphic; map the coordinates of the first target graphic with the coordinates of the image to be processed. , obtain the second target graphic; perform screenshot processing on the second target graphic to obtain the target image.
  • the second contour extraction module 302 is specifically configured to: perform contour extraction on a binary image to obtain the overall contour of the object to be detected; and obtain a coarse area image of the image to be processed based on the overall contour of the object to be detected; Obtain the effective contour of the object to be detected in the coarse area image of the image to be processed.
  • the second contour extraction module 302 is specifically configured to: perform contour extraction on the binary image of the first target area to obtain the overall contour of the first target area; and fit the overall contour of the first target area to The third target graphic; expand the third target graphic according to the set rules to obtain the coarse area binarized image of the first target area; intercept the coarse area of the first target area according to the coarse area binarized image of the first target area image.
  • the second contour extraction module 302 is specifically configured to perform XOR processing on the coarse area binarized image of the first target area and the overall image of the object to be detected, to obtain the second target area coarse area. image.
  • the comparison module 303 is also used to: compare the first defect area with the first area threshold; compare the second defect area with the first area threshold; if the first defect area is greater than the area threshold, or When the second defect area is greater than the first area threshold, it indicates that the object to be detected is defective.
  • embodiments of the present application also provide a computer-readable storage medium, which stores a computer program.
  • the computer program is run by a processor, the steps of the defect detection method described in the above-mentioned method embodiments are executed. .
  • the computer program product of the defect detection method provided by the embodiment of the present application includes a computer-readable storage medium storing program code.
  • the instructions included in the program code can be used to execute the steps of the defect detection method described in the above method embodiment. , please refer to the above method embodiments for details, and will not be described again here.
  • each block in the flowchart or block diagrams may represent a module, segment, or portion of code that contains one or more components for implementing the specified logical function(s). Executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures.
  • each block of the block diagram and/or flowchart illustration, and combinations of blocks in the block diagram and/or flowchart illustration can be implemented by special purpose hardware-based systems that perform the specified functions or acts. , or can be implemented using a combination of specialized hardware and computer instructions.
  • each functional module in each embodiment of the present application can be integrated together to form an independent part, each module can exist alone, or two or more modules can be integrated to form an independent part.
  • the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium.
  • the technical solution of the present application is essentially or the part that contributes to the existing technology or the part of the technical solution can be embodied in the form of a software product.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program code.
  • ROM read-only memory
  • RAM random access memory
  • magnetic disk or optical disk and other media that can store program code.
  • relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that these entities or operations are mutually exclusive. any such actual relationship or sequence exists between them.
  • the terms “comprises,” “comprises,” or any other variations thereof are intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus that includes a list of elements includes not only those elements, but also those not expressly listed other elements, or elements inherent to the process, method, article or equipment.
  • an element defined by the statement "comprising" does not exclude the presence of additional identical elements in a process, method, article, or device that includes the stated element.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)

Abstract

一种缺陷检测方法、装置、电子设备及可读存储介质，包括：对待处理图像进行待检测物轮廓提取，得到待检测物的缺陷图（201）；对所述缺陷图进行缺陷轮廓提取，以获得待检测物的缺陷面积（202）；将所述缺陷面积与第一面积阈值比较，当所述缺陷面积大于所述第一面积阈值，确定所述待检测物存在缺陷（203）。该方法通过对待检测物的图像进行轮廓提取，可以获取待检测物本体轮廓和缺陷轮廓，再进一步对该缺陷轮廓进行轮廓提取，获得缺陷的轮廓面积，将缺陷的轮廓面积与设定好的面积阈值进行对比后，便可以确定出待检测物存在缺陷。以此实现对目标物品的缺陷检测，减少了人为检测的工作强度，同时与人为检测相比不受主观情绪的影响，还增强了检测的准确性。

Description

缺陷检测方法、装置、电子设备及可读存储介质 技术领域
本申请涉及质量检测领域，具体而言，涉及一种缺陷检测方法、装置、电子设备及可读存储介质。
背景技术
随着智能化与电子化发展，模切材料应用和发展也得到了快速发展。基于模切模块的快速发展以及模切材料在各个领域逐渐占领重要作用，模切材料的检测变得十分重要，模块材料检测的准确性要求也越来越高。但是，目前模切材料的检测主要还是基于人工检测，受主观因素的影响较大。
发明内容
有鉴于此，本申请实施例的目的在于提供一种缺陷检测方法、装置、电子设备及可读存储介质。能够通过待处理图像进行处理以实现对模切材料的检测。
第一方面，本申请实施例提供了一种缺陷检测方法，包括：对待处理图像进行待检测物轮廓提取，得到待检测物的缺陷图；对所述缺陷图进行缺陷轮廓提取，以获得待检测物的缺陷面积；将所述缺陷面积与第一面积阈值比较，当所述缺陷面积大于所述第一面积阈值，确定所述待检测物存在缺陷。
在上述实现过程中，通过对待处理图像进行轮廓提取，可以获得待检测物本体和缺陷的轮廓，进而得到缺陷图。再在缺陷图的基础上进一步提取缺陷的轮廓并得到缺陷的轮廓面积，将缺陷的轮廓面积与第一面积阈值进行比较，当缺陷的轮廓面积大于第一面积阈值时，确定出该待检测物存在缺陷。由于在对待检测物的轮廓确定是和设定的面积阈值比较的，一方面可以有效的检测出较大缺陷，另一方面能够忽略较小的缺陷，实现了精准、有效的缺陷识别。且本缺陷识别的整个过程是基于待检测物的图像进行处理的，所以实现了检测自动化，提高了检测效率，降低了工作人员的工作强度。
结合第一方面，本申请实施例提供了第一方面的第一种可能的实施方式，其中：所述对待处理图像进行待检测物轮廓提取，得到缺陷图，包括：将所述待处理图像进行二值化处理，获得二值化图像；获取所述二值化图像中待检测物的有效轮廓，所述待检测物的有效轮廓为 待检测物的轮廓中满足预设条件的轮廓；从所述有效轮廓中提取目标图像；根据设定算法对所述目标图像进行卷积计算，以得到待检测物的缺陷图。
在上述实现过程中，通过将待处理图像进行二值化处理，得到可以反应图像整体和局部特征的二值化图像，不再涉及像素的多极值，使处理变得简单。同时，对轮廓设置条件，选择满足条件的轮廓作为有效轮廓，排除了一些干扰性轮廓，保证了轮廓获取的准确性。进而通过该有效轮廓提取的目标图像，对目标图像进行卷积计算得到的缺陷图也更加精确，提高了获取缺陷的准确率。
结合第一方面的第一种可能的实施方式，本申请实施例提供了第一方面的第二种可能的实施方式，其中：所述获取所述二值化图像中待检测物的有效轮廓，包括：提取所述二值化图像中待检测物的轮廓集，所述待检测物的轮廓集中包括待检测物的一个或多个轮廓；剔除所述待检测物的轮廓集中不满足预设条件的待检测物的轮廓，获得待检测物的有效轮廓。
在上述实现过程中，在实际轮廓提取过程中，可能会提取到一些非待检测物本体的一些轮廓，通过将图像中的各个轮廓面积与设定的预设条件进行比较，轮廓面积小于该预设条件的轮廓视为干扰轮廓(也就是非待检测物本体的轮廓)。通过将干扰轮廓剔除，便可以得到实际待检测物的有效轮廓，这样可以更加有针对性的进行进一步处理，简化了处理对象，提高了工作效率。
结合第一方面的第二种可能的实施方式，本申请实施例提供了第一方面的第三种可能的实施方式，其中，所述根据所述待检测物的有效轮廓截取目标图像，包括：将所述待检测物的有效轮廓拟合为第一目标图形；将所述第一目标图形的坐标以所述待处理图像坐标进行映射，得到第二目标图形；对所述第二目标图形进行截图处理，得到目标图像。
在上述实现过程中，通过对有效轮廓进行拟合，并以待处理图像的坐标进行影射，使得该有效轮廓拟合的面积和位置点更加贴合实际的待检测物的位置，保证目标图像更加贴合实际情况，提高了目标图像的准确率。
结合第一方面的第三种可能的实施方式，本申请实施例提供了第一方面的第四种可能的实施方式，其中，所述获取所述二值化图像中待检测物的有效轮廓，包括：对所述二值化图像进行轮廓提取，获取待检测物整体轮廓；根据所述待检测物整体轮廓获取待处理图像的粗区域图像；获取待处理图像的粗区域图像中待检测物的有效轮廓。
结合第一方面的第四种可能的实施方式，本申请实施例提供了第一方面的第五种可能的实施方式，其中，所述二值化图像包括第一目标区域二值化图像，所述待检测物整体轮廓包括第一目标区域整体轮廓，所述待处理图像的粗区域图像包括第一目标区域的粗区域图像， 所述根据所述待检测物整体轮廓获取待处理图像的粗区域图像，包括：对第一目标区域二值化图像进行轮廓提取，获取第一目标区域整体轮廓；将所述第一目标区域整体轮廓拟合为第三目标图形；按照设定规则对所述第三目标图形进行扩展，得到第一目标区域的粗区域二值化图像；根据所述第一目标区域的粗区域二值化图像截取第一目标区域的粗区域图像。
结合第一方面的第五种可能的实施方式，本申请实施例提供了第一方面的第六种可能的实施方式，其中，所述二值化图像包括第二目标区域二值化图像，所述待检测物整体轮廓包括第二目标区域整体轮廓，所述待处理图像的粗区域图像包括第二目标区域的粗区域图像，所述根据所述待检测物整体轮廓获取待处理图像的粗区域图像，包括：将所述第一目标区域的粗区域二值化图像与待检测物整体图像进行异或处理，以得到第二目标区域粗区域图像。
结合第一方面的第六种可能的实施方式，本申请实施例提供了第一方面的第七种可能的实施方式，其中，所述缺陷面积包括位于所述待处理图像的第一目标区域的一个或多个第一缺陷面积和位于所述待处理图像的第二目标区域的一个或多个第二缺陷面积，所述将所述缺陷面积与面积阈值比较，当所述缺陷面积大于所述第一面积阈值，确定所述待检测物存在缺陷，包括：将一个或多个所述第一缺陷面积与所述第一面积阈值比较；将一个或多个所述第二缺陷面积与所述第一面积阈值比较；若存在所述第一缺陷面积大于所述面积阈值，或存在所述第二缺陷面积大于所述第一面积阈值时，则表征所述待检测物存在缺陷。
在上述实现过程中，通过待检测物不同区域的缺陷面积与面积阈值分别进行比较，当其中任何一个缺陷面积大于面积阈值时，则可以确定该待检测物存在缺陷。通过各个方面对比，实现了全面的缺陷检测，提高了缺陷检测的准确率。
第二方面，本申请实施例还提供一种缺陷检测装置，包括：第一轮廓提取模块：用于对待处理图像进行待检测物轮廓提取，得到待检测物的缺陷图；第二轮廓提取模块：用于对所述缺陷图进行缺陷轮廓提取，以获得待检测物的缺陷面积；比较模块：用于将所述缺陷面积与第一面积阈值比较，当所述缺陷面积大于所述第一面积阈值，确定所述待检测物存在缺陷。
第三方面，本申请实施例还提供一种电子设备，包括：处理器、存储器，所述存储器存储有所述处理器可执行的机器可读指令，当电子设备运行时，所述机器可读指令被所述处理器执行时执行上述第一方面，或第一方面的任一种可能的实施方式中的方法的步骤。
第四方面，本申请实施例还提供一种计算机可读存储介质，该计算机可读存储介质上存储有计算机程序，该计算机程序被处理器运行时执行上述第一方面，或第一方面的任一种可能的实施方式中缺陷检测方法的步骤。
为使本申请的上述目的、特征和优点能更明显易懂，下文特举实施例，并配合所附附图， 作详细说明如下。
附图说明
为了更清楚地说明本申请实施例的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，应当理解，以下附图仅示出了本申请的某些实施例，因此不应被看作是对范围的限定，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他相关的附图。
图1为本申请实施例提供的电子设备的方框示意图；
图2为本申请实施例提供的缺陷检测方法的流程图；
图3为本申请实施例提供的待处理图像轮廓示意图；
图4为本申请实施例提供的缺陷检测方法中步骤201的流程图；
图5为本申请实施例提供的待处理图像轮廓的局部组成示意图；
图6为本申请实施例提供的缺陷检测装置的功能模块示意图。
具体实施方式
下面将结合本申请实施例中附图，对本申请实施例中的技术方案进行描述。
应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步定义和解释。同时，在本申请的描述中，术语“第一”、“第二”等仅用于区分描述，而不能理解为指示或暗示相对重要性。
目前，随着智能化、工业化的快速发展，电子设备、电器设备、精密仪器、电子通讯等行业也随即进入迅速发展阶段。而模切材料广泛的应用于这些行业，并逐渐在各个行业占领着重要作用，其市场也得到快速发展。
由于模切材料在各个行业都起着举足轻重的作用，模切材料的质量影响着整个设备的质量。基于此，对模切材料的检测也变得十分重要。但是，目前模切材料的检测还是依靠人为检测，人为检测会受主观情绪的影响，且人的精力有限，长时间的工作可能会造成精力不足，进而检测效率会降低，无法保证检测的准确性和高效性。
本申请发明人在对模切材料检测过程中发现：想要检测出模切材料是否存在缺陷，可以将通过对膜切材料进行图像采集，基于该模切材料的图像进行缺陷轮廓和模切材料本体轮廓提取，通过轮廓面积确定该模切材料是否存在缺陷。有鉴于此，本申请发明人提出一种缺陷检测方法，可以通过对待检测物的图像进行处理以实现对待检测物的检测，以替代人为检测， 使得检测更加客观、准确，同时更加高效率。
本申请实施例公开的检测方法可以但不限用于模切材料的检测、芯片的检测、电子屏幕的检测、玻璃材料的检测。通过对阈值及检测逻辑的设置可以利用本申请的检测装置实现多种待检测物的缺陷检测。
为便于对本实施例进行理解，首先对执行本申请实施例所公开的缺陷检测方法的电子设备进行详细介绍。
如图1所示，是电子设备的方框示意图。电子设备100可以包括存储器111、处理器113、输入输出单元115。本领域普通技术人员可以理解，图1所示的结构仅为示意，其并不对电子设备100的结构造成限定。例如，电子设备100还可包括比图1中所示更多或者更少的组件，或者具有与图1所示不同的配置。
上述的存储器111、处理器113及输入输出单元115各元件相互之间直接或间接地电性连接，以实现数据的传输或交互。例如，这些元件相互之间可通过一条或多条通讯总线或信号线实现电性连接。上述的处理器113用于执行存储器中存储的可执行模块。
其中，存储器111可以是，但不限于，随机存取存储器(Random Access Memory，简称RAM)，只读存储器(Read Only Memory，简称ROM)，可编程只读存储器(Programmable Read-Only Memory，简称PROM)，可擦除只读存储器(Erasable Programmable Read-Only Memory，简称EPROM)，电可擦除只读存储器(Electric Erasable Programmable Read-Only Memory，简称EEPROM)等。其中，存储器111用于存储程序，所述处理器113在接收到执行指令后，执行所述程序，本申请实施例任一实施例揭示的过程定义的电子设备100所执行的方法可以应用于处理器113中，或者由处理器113实现。
上述存储器111可以用于存储待处理图像、第一面积阈值、第二面积阈值等信息。上述存储器111还可以用于对目标图像、二值化图像、粗区域图像等中间信息的暂时存储。
上述的处理器113可能是一种集成电路芯片，具有信号的处理能力。上述的处理器113可以是通用处理器，包括中央处理器(Central Processing Unit，简称CPU)、网络处理器(Network Processor，简称NP)等；还可以是数字信号处理器(Digital Signal Processor，简称DSP)、专用集成电路(Application Specific Integrated Circuit，简称ASIC)、现场可编程门阵列(FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
上述处理器113用于完成对待处理图像的轮廓提取、待处理图像的二值化处理、轮廓的 筛选、轮廓面积与面积阈值的比较、有效轮廓的拟合、目标图形的映射等一系列动作。
上述的输入输出单元115用于提供给用户输入数据。所述输入输出单元115可以是，但不限于，鼠标、键盘和交互界面等。
上述的输入输出单元115可以用于输入第一面积阈值、第二面积阈值等数据。该数据可以根据待检测物的不同进行设置，当待检测物更改时，第一面积阈值与预设条件可以通过上述输入输出单元115进行相应的修改。
本实施例中的电子设备100可以用于执行本申请实施例提供的各个方法中的各个步骤。下面通过几个实施例详细描述缺陷检测方法的实现过程。
请参阅图2，是本申请实施例提供的缺陷检测方法的流程图。下面将对图2所示的具体流程进行详细阐述。
步骤201，对待处理图像进行待检测物轮廓提取，得到待检测物的缺陷图。
可选地，上述待处理图像为采集到的待检测物的图像。该待处理图像可以通过采集设备获得。
可选地，对待处理图像进行轮廓提取后可以获得待检测装置的第一目标区域轮廓、第二目标轮廓区域、缺陷轮廓、无关轮廓区域等。其中，无关轮廓区域为非待检测物本体区域，例如，可以是放置待检测物平台的轮廓、可以是待检测物周围的杂质的轮廓等。
示例性地，如图3所示，图中虚线圆圈可以视为无关轮廓，实线正方形为待检测物本体轮廓，填充的圆圈为缺陷轮廓。
可选地，在对待处理图像进行待检测物轮廓提取之前还可以对待检测物的图像进行缩放处理，得到待处理图像。该缩放处理可以采用双线性插值算法，该缩放处理可以采用最邻近插值算法等。通过对待检测物的图像进行缩放处理能够使用较少的资源对图像进行快速处理。
步骤202，对缺陷图进行缺陷轮廓提取，以获得待检测物的缺陷面积。
可选地，该缺陷图中可以包括缺陷轮廓与无关轮廓，该缺陷图中也可以只包括缺陷轮廓。若该缺陷图中包括缺陷轮廓与无关轮廓，则需要对该缺陷轮廓与无关轮廓进行提取。若该缺陷图中只包括缺陷轮廓，则需要对该缺陷轮廓进行提取。
可选地，可以根据缺陷轮廓的长宽计算该缺陷的面积，还可以根据该缺陷轮廓的半径计算缺陷面积。
步骤203，将缺陷面积与第一面积阈值比较，当缺陷面积大于第一面积阈值，确定该待检测物存在缺陷。
可选地，第一面积阈值为允许缺陷的最大面积值，该第一面积阈值可以根据不同的待检 测物进行相应的调整。
可选地，在对缺陷图进行缺陷轮廓提取后，还可以直接获取缺陷的长度和宽度，将该长度与长度阈值进行比较，将宽度与宽度阈值进行比较，当长度或宽度其中有一个值大于对应的阈值，则可以确定该待检测物存在缺陷。
在一种可能的实现方式中，如图4所示，步骤201包括：步骤2011-2013。
步骤2011，将待处理图像进行二值化处理，获得二值化图像。
图像的二值化处理就是将图像上的点的灰度值设置为0或255，也就是将整个图像呈现出明显的黑白效果，即将多个亮度等级的灰度图像通过适当的阈值选取而获得仍然可以反应图像整体和局部特征的二值化图像。通过对图像的二值化处理得到二值化图像后，有利于在对图像进一步进行处理时，图像的性质只和像素值0或255的点的位置有关，不再涉及像素的多级值，使得处理变得简单。
示例性地，假设当前阈值为188，若当前点值大于188时，则当前点值取255，若当前点值小于188，则当前点值取0。依照此方法将待处理图像的各个像素点进行划分，便得到二值化图像。
步骤2012，获取该二值化图像中待检测物的有效轮廓。
其中，待检测物的有效轮廓为待检测物的轮廓中满足预设条件的轮廓。
可选地，有效轮廓可以包括待检测物本体轮廓和缺陷轮廓。有效轮廓也可以仅包括待检测物本体轮廓。通过设置预设条件，将无效轮廓剔除，便可以得到有效轮廓。
示例性地，若预设条件为：轮廓面积大于5的轮廓为有效轮廓，此时二值化图像中存在A、B、C、D、E五个轮廓，将该五个轮廓的轮廓面积分别于预设条件5进行比较，获得比较结果为：轮廓A与轮廓E面积小于5，轮廓B、轮廓C及轮廓D廓面积大于5，则可以确定出轮廓A与轮廓E为无效轮廓，轮廓B、轮廓C及轮廓D为有效轮廓。进一步的，可以将轮廓A与轮廓E剔除，获得轮廓B、轮廓C及轮廓D等有效轮廓。
步骤2013，从该有效轮廓中提取目标图像。
可选地，目标图像可以是处理后的待检测物的整体图像，目标图像可以是处理后的待检测物的一个或多个局部图像。
可选地，可以直接从该有效轮廓中提取目标图像；也可以将该有效轮廓进行拟合，在拟合后的有效轮廓中提取图像；还可以对该有效轮廓进行映射，在映射后的有效轮廓中提取图像。
步骤2014，根据设定算法对目标图像进行卷积计算，以得到待检测物的缺陷图。
可选地，根据设定算法对目标图像进行卷积计算可以基于Tensorflow、PyTorch等进行处理。
可选地，该缺陷图可以是待检测物的整体缺陷图，该缺陷图还可以是待检测物的局部缺陷图。
示例性地，若待检测物是电池，则电池为待检测物的整体，极耳和电芯主体为待检测物的局部。则该缺陷图可以包括电池整体缺陷图，该缺陷图可以包括极耳缺陷图，该缺陷图可以包括电芯主体缺陷图，该缺陷图还可以包括极耳缺陷图和电芯主体缺陷图。
示例性地，若待检测物是芯片，则芯片为待检测物的整体，主板和引脚为待检测物的局部。则该缺陷图可以包括芯片整体缺陷图，该缺陷图可以包括主板缺陷图，该缺陷图可以包括引脚缺陷图，该缺陷图还可以包括主板缺陷图和引脚缺陷图。
在一种可能的实现方式中，步骤2012，包括：提取二值化图像中待检测物的轮廓集，剔除待检测物的轮廓集中不满足预设条件的待检测物的轮廓，获得待检测物的有效轮廓。
可选地，待检测物的轮廓集中包括待检测物的一个或多个轮廓。例如，待检测物的轮廓集中可以包括待检测物本体轮廓、待检测物局部轮廓、待检测物内部缺陷轮廓、或其他干扰轮廓等。
可选地，该预设条件可以是轮廓面积大于第二面积阈值、轮廓面积小于第二面积阈值、或轮廓面积不属于第二面积阈值范围，该第二面积阈值可以用于剔除干扰轮廓，该第二面积阈值还可以用于剔除缺陷轮廓，该第二面积阈值可以用于同时剔除干扰轮廓和缺陷轮廓。
示例性地，如图3所示，若图中虚线圆圈面积为3，实线正方形面积为10，填充的圆圈面积为5和7。此时第二面积阈值用于剔除干扰轮廓，则该预设条件可以设置为轮廓面积大于4，则可以根据该预设条件剔除虚线圆圈轮廓。
示例性地，如图3所示，若图中虚线圆圈面积为3，实线正方形面积为10，填充的圆圈面积为5和7。此时第二面积阈值用于同时剔除干扰轮廓和缺陷轮廓，则该预设条件可以设置为轮廓面积大于8，则可以根据该预设条件剔除虚线圆圈轮廓和填充的圆圈轮廓。
可选地，该预设条件还可以是轮廓位置是否在目标位置，该目标位置可以是待检测物本体轮廓的位置。当轮廓位置处于目标位置范围内，则表明轮廓位置满足预设条件。当轮廓位置不处于目标位置范围内，则表明轮廓位置不满足预设条件。
示例性地，如图3所示，实线正方形轮廓为待检测物本体轮廓的位置，则可以确定虚线圆圈轮廓不满足预设条件，则可以将虚线圆圈轮廓剔除。
在一种可能的实现方式中，步骤2012，包括：将待检测物的有效轮廓拟合为第一目标图 形；将该第一目标图形的坐标以待处理图像坐标进行映射，得到第二目标图形；对第二目标图形进行截图处理，得到目标图像。
可选地，该第一目标图形可以是矩形，该第一目标图形可以是圆形，该第一目标图形还可以是三角形等。相应地，该第二目标图形可以是矩形，该第二目标图形可以是圆形，该第二目标图形还可以是三角形等。
可选地，可以通过Opencv的boundingRect或minAreaRect构造函数用轮廓进行第一目标图形拟合。
可选地，将第一目标图形坐标以待处理图像坐标进行映射可以将坐标进行比例放大。将第一目标图形坐标以待处理图像坐标进行映射也可以通过设置相应的模型进行处理实现
示例性地，该映射可以通过以下过程实现，比例＝原图尺寸÷缩放图尺寸，第二目标图形的第一边＝第一目标图形的第一边×比例，第二目标图形的第二边＝第一目标图形的第二边×比例，第二目标图形的第三边＝第一目标图形的第三边×比例。
在一种可能的实现方式中，步骤2012，包括：对二值化图像进行轮廓提取，获取待检测物整体轮廓；根据待检测物整体轮廓获取待处理图像的粗区域图像；获取待处理图像的粗区域图像中待检测物的有效轮廓。
可选地，待检测物可以视为多个局部部件组成。则该二值化图像包括第一目标区域二值化图像、第二目标区域二值化图像、第三目标区域二值化图像等。待检测物整体轮廓包括第一目标区域整体轮廓、第二目标区域整体轮廓、第三目标区域整体轮廓等。待处理图像的粗区域图像包括第一目标区域的粗区域图像、第二目标区域的粗区域图像、第三目标区域的粗区域图像等。该缺陷面积包括第一目标区域的一个或多个第一缺陷面积和第二目标区域的一个或多个第二缺陷面积。
示例性地，如图5所示，待检测物可以视为有第一目标区域41与第二目标区域42组成。则第一缺陷面积包括411、412及423，第二缺陷面积包括420、421、422及423。
可选地，获取待处理图像的粗区域图像中待检测物的有效轮廓可以包括：对该待处理图像的粗区域图像进行阈值分割，得到待处理图像的粗区域的二值化图像。对该粗区域的二值化图像进行轮廓提取，得到待检测物的轮廓集。剔除待检测物的轮廓集中不满足预设条件的待检测物的轮廓，获得待检测物的有效轮廓。
在一种可能的实现方式中，根据待检测物整体轮廓获取待处理图像的粗区域图像，包括：对第一目标区域二值化图像进行轮廓提取，获取第一目标区域整体轮廓。将第一目标区域整体轮廓拟合为第三目标图形。按照设定规则对该第三目标图形进行扩展，得到第一目标区域 的粗区域二值化图像；根据该第一目标区域的粗区域二值化图像截取第一目标区域的粗区域图像。
示例性地，若待检测物为电池，则该以第一目标区域可以为极耳。根据待检测物整体轮廓获取待处理图像的粗区域图像可以包括：对极耳二值化图像进行轮廓提取，获取极耳整体轮廓。将极耳整体轮廓拟合为第三目标图形。按照设定规则对该第三目标图形进行扩展，得到极耳的粗区域二值化图像；根据该极耳的粗区域二值化图像截取极耳的粗区域图像。
示例性地，若待检测物为芯片，则该以第一目标区域可以为引脚。根据待检测物整体轮廓获取待处理图像的粗区域图像可以包括：对引脚二值化图像进行轮廓提取，获取引脚整体轮廓。将引脚整体轮廓拟合为第三目标图形。按照设定规则对该第三目标图形进行扩展，得到引脚的粗区域二值化图像；根据该引脚的粗区域二值化图像截取引脚的粗区域图像。
可选地，该第三目标图形可以是矩形，该第三目标图形可以是圆形，该第三目标图形还可以是三角形等。
可选地，该设定规则可以设置为按照指定方向进行扩展，该指定方向可以设置为向坐标Y轴进行扩展，该指定方向可以设置为向坐标X轴进行扩展。
在一种可能的实现方式中，根据待检测物整体轮廓获取待处理图像的粗区域图像，包括：将第一目标区域的粗区域二值化图像与待检测物整体图像进行异或处理，以得到第二目标区域粗区域图像。
可选地，第一目标区域的粗区域二值化图像与待检测物整体图像进行异或处理，以得到第二目标区域粗区域图像可以包括：在待检测物整体图像中剔除第一目标区域的粗区域二值化图像，以得到第二目标区域粗区域图像。
示例性地，若待检测物为电池，则该以第二目标区域可以为电芯本体。第一目标区域的粗区域二值化图像与待检测物整体图像进行异或处理，以得到第二目标区域粗区域图像可以包括：在电池整体图像中剔除极耳的粗区域二值化图像，以得到电芯本体粗区域图像。
示例性地，若待检测物为芯片，则该以第二目标区域可以为主板。第一目标区域的粗区域二值化图像与待检测物整体图像进行异或处理，以得到第二目标区域粗区域图像可以包括：在芯片整体图像中剔除引脚的粗区域二值化图像，以得到主板粗区域图像。
可选地，若待检测物还包括第三目标区域，则根据待检测物整体轮廓获取待处理图像的粗区域图像，包括：将第一目标区域和第三目标区域的粗区域二值化图像与待检测物整体图像进行异或处理，以得到第二目标区域粗区域图像。
可选地，若待检测物还包括第三目标区域与第四目标区域，则根据待检测物整体轮廓获 取待处理图像的粗区域图像，包括：将第一目标区域、第三目标区域和第四目标区域的粗区域二值化图像与待检测物整体图像进行异或处理，以得到第二目标区域粗区域图像。
在一种可能的实现方式中，将缺陷面积与面积阈值比较，当缺陷面积大于第一面积阈值，确定该待检测物存在缺陷，包括：将一个或多个第一缺陷面积与第一面积阈值比较。将一个或多个第二缺陷面积与第一面积阈值比较。若存在第一缺陷面积大于该第一面积阈值，或存在第二缺陷面积大于该第一面积阈值时，则表征该待检测物存在缺陷。
示例性地，如图5所示，图中示出三个第一缺陷面积和四个第二缺陷面积。将三个第一缺陷面积和四个第二缺陷面积分别与第一面积阈值比较，若第一缺陷面积或第二缺陷面积中存在至少一个大于该第一面积阈值时，则可以判定该待检测物中存在缺陷。
进一步的，为了更好的理解，假设图5中的第一缺陷面积411、412、413的面积分别为8平方毫米、6平方毫米、3平方毫米，第二缺陷面积420、421、422、423的面积分别为4平方毫米、7平方毫米、10平方毫米、13平方毫米。若此时的第一面积阈值为15平方毫米，由于第一缺陷面积与第二缺陷面积中的所有缺陷面积均小于第一面积阈值，则可以判定该待检测物中不存在缺陷。
假设图5中的第一缺陷面积411、412、413的面积分别为8平方毫米、6平方毫米、3平方毫米，第二缺陷面积420、421、422、423的面积分别为4平方毫米、7平方毫米、10平方毫米、13平方毫米。若此时的第一面积阈值为10平方毫米，由于第二缺陷面积中423大于第一面积阈值，则可以判定该待检测物中存在缺陷。
可选地，若待检测物还包括第三目标区域、第四目标区域，则若第一缺陷面积大于该面积阈值，或第二缺陷面积大于该第一面积阈值，或第三缺陷面积大于该第一面积阈值，或第四缺陷面积大于该第一面积阈值时，则表征该待检测物存在缺陷。
本申请实施例通过将待检测物的图像进行一系列处理后得到待处理图像，对该待处理图像进行轮廓提取、轮廓筛选、缺陷识别等处理，检测出待检测物中的缺陷。基于该图像进行处理和检测可以更加客观的甄别待检测物中的缺陷。同时，本申请中的处理方式是对图像进行一系列的处理和识别，这些过程都是基于计算机等自动化设备实现的，极大的减轻了工作人员的工作强度，提高了检测效率。
基于同一申请构思，本申请实施例中还提供了与缺陷检测方法对应的缺陷检测装置，由于本申请实施例中的装置解决问题的原理与前述的缺陷检测方法实施例相似，因此本实施例中的装置的实施可以参见上述方法的实施例中的描述，重复之处不再赘述。
请参阅图6，是本申请实施例提供的缺陷检测装置的功能模块示意图。本实施例中的缺 陷检测装置中的各个模块用于执行上述方法实施例中的各个步骤。缺陷检测装置包括第一轮廓提取模块301、第二轮廓提取模块302、比较模块303；其中，
第一轮廓提取模块301用于对待处理图像进行待检测物轮廓提取，得到待检测物的缺陷图。
第二轮廓提取模块302用于对缺陷图进行缺陷轮廓提取，以获得待检测物的缺陷面积。
比较模块303用于将缺陷面积与第一面积阈值比较，当缺陷面积大于第一面积阈值，确定待检测物存在缺陷。
一种可能的实施方式中，第二轮廓提取模块302，还用于：将待处理图像进行二值化处理，获得二值化图像；获取二值化图像中待检测物的有效轮廓，该待检测物的有效轮廓为待检测物的轮廓中满足预设条件的轮廓；从有效轮廓中提取目标图像；根据设定算法对目标图像进行卷积计算，以得到待检测物的缺陷图。
一种可能的实施方式中，第二轮廓提取模块302，具体用于：提取二值化图像中待检测物的轮廓集，该待检测物的轮廓集中包括待检测物的一个或多个轮廓；剔除待检测物的轮廓集中不满足预设条件的待检测物的轮廓，获得待检测物的有效轮廓。
一种可能的实施方式中，第二轮廓提取模块302，具体用于：将待检测物的有效轮廓拟合为第一目标图形；将第一目标图形的坐标以所述待处理图像坐标进行映射，得到第二目标图形；对第二目标图形进行截图处理，得到目标图像。
一种可能的实施方式中，第二轮廓提取模块302，具体用于：对二值化图像进行轮廓提取，获取待检测物整体轮廓；根据待检测物整体轮廓获取待处理图像的粗区域图像；获取待处理图像的粗区域图像中待检测物的有效轮廓。
一种可能的实施方式中，第二轮廓提取模块302，具体用于：对第一目标区域二值化图像进行轮廓提取，获取第一目标区域整体轮廓；将第一目标区域整体轮廓拟合为第三目标图形；按照设定规则对第三目标图形进行扩展，得到第一目标区域的粗区域二值化图像；根据第一目标区域的粗区域二值化图像截取第一目标区域的粗区域图像。
一种可能的实施方式中，第二轮廓提取模块302，具体用于：将第一目标区域的粗区域二值化图像与待检测物整体图像进行异或处理，以得到第二目标区域粗区域图像。
一种可能的实施方式中，比较模块303，还用于：将第一缺陷面积与第一面积阈值比较；将第二缺陷面积与第一面积阈值比较；若第一缺陷面积大于面积阈值，或第二缺陷面积大于第一面积阈值时，则表征待检测物存在缺陷。
此外，本申请实施例还提供一种计算机可读存储介质，该计算机可读存储介质上存储有 计算机程序，该计算机程序被处理器运行时执行上述方法实施例中所述的缺陷检测方法的步骤。
本申请实施例所提供的缺陷检测方法的计算机程序产品，包括存储了程序代码的计算机可读存储介质，所述程序代码包括的指令可用于执行上述方法实施例中所述的缺陷检测方法的步骤，具体可参见上述方法实施例，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的装置和方法，也可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的，例如，附图中的流程图和框图显示了根据本申请的多个实施例的装置、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或代码的一部分，所述模块、程序段或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现方式中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或动作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
另外，在本申请各个实施例中的各功能模块可以集成在一起形成一个独立的部分，也可以是各个模块单独存在，也可以两个或两个以上模块集成形成一个独立的部分。
所述功能如果以软件功能模块的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括……”限定的要素，并不排除在包括所述要素的过程、方法、 物品或者设备中还存在另外的相同要素。
以上所述仅为本申请的优选实施例而已，并不用于限制本申请，对于本领域的技术人员来说，本申请可以有各种更改和变化。凡在本申请的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步定义和解释。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以权利要求的保护范围为准。

Claims (11)

  1. 一种缺陷检测方法，其特征在于，包括：
    对待处理图像进行待检测物轮廓提取，得到待检测物的缺陷图；
    对所述缺陷图进行缺陷轮廓提取，以获得待检测物的缺陷面积；
    将所述缺陷面积与第一面积阈值比较，当所述缺陷面积大于所述第一面积阈值，确定所述待检测物存在缺陷。
  2. 根据权利要求1所述的方法，其特征在于，所述对待处理图像进行待检测物轮廓提取，得到缺陷图，包括：
    将所述待处理图像进行二值化处理，获得二值化图像；
    获取所述二值化图像中待检测物的有效轮廓，所述待检测物的有效轮廓为待检测物的轮廓中满足预设条件的轮廓；
    从所述有效轮廓中提取目标图像；
    根据设定算法对所述目标图像进行卷积计算，以得到待检测物的缺陷图。
  3. 根据权利要求2所述的方法，其特征在于，所述获取所述二值化图像中待检测物的有效轮廓，包括：
    提取所述二值化图像中待检测物的轮廓集，所述待检测物的轮廓集中包括待检测物的一个或多个轮廓；
    剔除所述待检测物的轮廓集中不满足预设条件的待检测物的轮廓，获得待检测物的有效轮廓。
  4. 根据权利要求2所述的方法，其特征在于，所述从所述有效轮廓中提取目标图像，包括：
    将所述待检测物的有效轮廓拟合为第一目标图形；
    将所述第一目标图形的坐标以所述待处理图像坐标进行映射，得到第二目标图形；
    对所述第二目标图形进行截图处理，得到目标图像。
  5. 根据权利要求2所述的方法，其特征在于，所述获取所述二值化图像中待检测物的有效轮廓，包括：
    对所述二值化图像进行轮廓提取，获取待检测物整体轮廓；
    根据所述待检测物整体轮廓获取待处理图像的粗区域图像；
    获取待处理图像的粗区域图像中待检测物的有效轮廓。
  6. 根据权利要求5所述的方法，其特征在于，所述二值化图像包括第一目标区域二值化 图像，所述待检测物整体轮廓包括第一目标区域整体轮廓，所述待处理图像的粗区域图像包括第一目标区域的粗区域图像，所述根据所述待检测物整体轮廓获取待处理图像的粗区域图像，包括：
    对第一目标区域二值化图像进行轮廓提取，获取第一目标区域整体轮廓；
    将所述第一目标区域整体轮廓拟合为第三目标图形；
    按照设定规则对所述第三目标图形进行扩展，得到第一目标区域的粗区域二值化图像；
    根据所述第一目标区域的粗区域二值化图像截取第一目标区域的粗区域图像。
  7. 根据权利要求6所述的方法，其特征在于，所述二值化图像包括第二目标区域二值化图像，所述待检测物整体轮廓包括第二目标区域整体轮廓，所述待处理图像的粗区域图像包括第二目标区域的粗区域图像，所述根据所述待检测物整体轮廓获取待处理图像的粗区域图像，包括：
    将所述第一目标区域的粗区域二值化图像与待检测物整体图像进行异或处理，以得到第二目标区域粗区域图像。
  8. 根据权利要求1所述的方法，其特征在于，所述缺陷面积包括位于所述待处理图像的第一目标区域的一个或多个第一缺陷面积和位于所述待处理图像的第二目标区域的一个或多个第二缺陷面积，所述将所述缺陷面积与第一面积阈值比较，当所述缺陷面积大于所述第一面积阈值，确定所述待检测物存在缺陷，包括：
    将一个或多个所述第一缺陷面积与所述第一面积阈值比较；
    将一个或多个所述第二缺陷面积与所述第一面积阈值比较；
    若存在所述第一缺陷面积大于所述第一面积阈值，或存在所述第二缺陷面积大于所述第一面积阈值时，则表征所述待检测物存在缺陷。
  9. 一种缺陷检测装置，其特征在于，包括：
    第一轮廓提取模块：用于对待处理图像进行待检测物轮廓提取，得到待检测物的缺陷图；
    第二轮廓提取模块：用于对所述缺陷图进行缺陷轮廓提取，以获得待检测物的缺陷面积；
    比较模块：用于将所述缺陷面积与第一面积阈值比较，当所述缺陷面积大于所述第一面积阈值，确定所述待检测物存在缺陷。
  10. 一种电子设备，其特征在于，包括：处理器、存储器，所述存储器存储有所述处理器可执行的机器可读指令，当电子设备运行时，所述机器可读指令被所述处理器执行时执行如权利要求1至8任一所述的方法的步骤。
  11. 一种计算机可读存储介质，其特征在于，该计算机可读存储介质上存储有计算机程 序，该计算机程序被处理器运行时执行如权利要求1至8任一所述的方法的步骤。
PCT/CN2022/140036 2022-03-23 2022-12-19 缺陷检测方法、装置、电子设备及可读存储介质 WO2023179122A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210291632.2 2022-03-23
CN202210291632.2A CN114627092A (zh) 2022-03-23 2022-03-23 缺陷检测方法、装置、电子设备及可读存储介质

Publications (1)

Publication Number Publication Date
WO2023179122A1 true WO2023179122A1 (zh) 2023-09-28

Family

ID=81903121

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/140036 WO2023179122A1 (zh) 2022-03-23 2022-12-19 缺陷检测方法、装置、电子设备及可读存储介质

Country Status (2)

Country Link
CN (1) CN114627092A (zh)
WO (1) WO2023179122A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117058155A (zh) * 2023-10-13 2023-11-14 西安空天机电智能制造有限公司 3dp金属打印铺粉缺陷检测方法、装置、设备及介质
CN117115157A (zh) * 2023-10-23 2023-11-24 湖南隆深氢能科技有限公司 基于pem电解槽的缺陷检测方法、系统、终端设备及介质
CN117173185A (zh) * 2023-11-03 2023-12-05 东北大学 轧制板材的区域检测方法及装置、存储介质、计算机设备
CN117474908A (zh) * 2023-12-26 2024-01-30 宁德时代新能源科技股份有限公司 标注方法、装置、设备及计算机可读存储介质
CN117723491A (zh) * 2024-02-07 2024-03-19 宁德时代新能源科技股份有限公司 电芯防爆阀的检测系统及检测方法

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114627092A (zh) * 2022-03-23 2022-06-14 广东利元亨智能装备股份有限公司 缺陷检测方法、装置、电子设备及可读存储介质
CN115829921B (zh) * 2022-09-16 2024-01-05 宁德时代新能源科技股份有限公司 检测电芯缺陷的方法、装置和计算机可读存储介质
CN115953373B (zh) * 2022-12-22 2023-12-15 青岛创新奇智科技集团股份有限公司 玻璃缺陷检测方法、装置、电子设备及存储介质
CN118115439A (zh) * 2023-06-16 2024-05-31 上海感图网络科技有限公司 金手指缺陷复核方法、装置、设备及存储介质
CN117611799A (zh) * 2023-11-28 2024-02-27 杭州深度视觉科技有限公司 基于图像识别的西林瓶缺陷检测方法及装置
CN117870564A (zh) * 2024-03-11 2024-04-12 宁德时代新能源科技股份有限公司 电芯Mylar膜的检测方法及系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030223631A1 (en) * 2002-06-04 2003-12-04 Dainippon Screen Mfg. Co., Ltd. Apparatus and method for defect detection and program thereof
CN108896278A (zh) * 2018-05-23 2018-11-27 精锐视觉智能科技（深圳）有限公司 一种滤光片丝印缺陷检测方法、装置及终端设备
CN112200805A (zh) * 2020-11-11 2021-01-08 北京平恒智能科技有限公司 一种工业品图像目标提取及缺陷判断方法
CN112700440A (zh) * 2021-01-18 2021-04-23 上海闻泰信息技术有限公司 物体缺陷检测方法、装置、计算机设备及存储介质
CN113436131A (zh) * 2020-03-04 2021-09-24 上海微创卜算子医疗科技有限公司 缺陷检测方法、装置、电子设备和存储介质
CN114627092A (zh) * 2022-03-23 2022-06-14 广东利元亨智能装备股份有限公司 缺陷检测方法、装置、电子设备及可读存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030223631A1 (en) * 2002-06-04 2003-12-04 Dainippon Screen Mfg. Co., Ltd. Apparatus and method for defect detection and program thereof
CN108896278A (zh) * 2018-05-23 2018-11-27 精锐视觉智能科技（深圳）有限公司 一种滤光片丝印缺陷检测方法、装置及终端设备
CN113436131A (zh) * 2020-03-04 2021-09-24 上海微创卜算子医疗科技有限公司 缺陷检测方法、装置、电子设备和存储介质
CN112200805A (zh) * 2020-11-11 2021-01-08 北京平恒智能科技有限公司 一种工业品图像目标提取及缺陷判断方法
CN112700440A (zh) * 2021-01-18 2021-04-23 上海闻泰信息技术有限公司 物体缺陷检测方法、装置、计算机设备及存储介质
CN114627092A (zh) * 2022-03-23 2022-06-14 广东利元亨智能装备股份有限公司 缺陷检测方法、装置、电子设备及可读存储介质

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117058155A (zh) * 2023-10-13 2023-11-14 西安空天机电智能制造有限公司 3dp金属打印铺粉缺陷检测方法、装置、设备及介质
CN117058155B (zh) * 2023-10-13 2024-03-12 西安空天机电智能制造有限公司 3dp金属打印铺粉缺陷检测方法、装置、设备及介质
CN117115157A (zh) * 2023-10-23 2023-11-24 湖南隆深氢能科技有限公司 基于pem电解槽的缺陷检测方法、系统、终端设备及介质
CN117115157B (zh) * 2023-10-23 2024-02-06 湖南隆深氢能科技有限公司 基于pem电解槽的缺陷检测方法、系统、终端设备及介质
CN117173185A (zh) * 2023-11-03 2023-12-05 东北大学 轧制板材的区域检测方法及装置、存储介质、计算机设备
CN117173185B (zh) * 2023-11-03 2024-01-19 东北大学 轧制板材的区域检测方法及装置、存储介质、计算机设备
CN117474908A (zh) * 2023-12-26 2024-01-30 宁德时代新能源科技股份有限公司 标注方法、装置、设备及计算机可读存储介质
CN117474908B (zh) * 2023-12-26 2024-05-28 宁德时代新能源科技股份有限公司 标注方法、装置、设备及计算机可读存储介质
CN117723491A (zh) * 2024-02-07 2024-03-19 宁德时代新能源科技股份有限公司 电芯防爆阀的检测系统及检测方法

Also Published As

Publication number Publication date
CN114627092A (zh) 2022-06-14

Similar Documents

Publication Publication Date Title
WO2023179122A1 (zh) 缺陷检测方法、装置、电子设备及可读存储介质
CN115147414B (zh) 一种双极型功率晶体管表面击穿缺陷检测方法
JP2018005640A (ja) 分類器生成装置、画像検査装置、及び、プログラム
JP2018005639A (ja) 画像分類装置、画像検査装置、及び、プログラム
Tonti et al. An automated approach to the segmentation of HEp-2 cells for the indirect immunofluorescence ANA test
TW201407154A (zh) 自動及手動缺陷分類之整合
JP2019087078A (ja) データ生成装置、データ生成方法及びデータ生成プログラム
CN107016394B (zh) 一种交叉纤维特征点匹配方法
CN111539238B (zh) 二维码图像修复方法、装置、计算机设备和存储介质
CN112037185B (zh) 染色体分裂相图像筛选方法、装置及终端设备
CN113469921B (zh) 图像缺陷修复方法、系统、计算机设备及存储介质
CN110570442A (zh) 一种复杂背景下轮廓检测方法、终端设备及存储介质
Deng et al. Defect detection of bearing surfaces based on machine vision technique
CN115294527B (zh) 一种基于计算机视觉的地铁隧道破损检测方法
CN113609984A (zh) 一种指针式仪表读数识别方法、装置及电子设备
CN111242899A (zh) 基于图像的瑕疵检测方法及计算机可读存储介质
CN115273115A (zh) 一种文档元素标注方法、装置、电子设备和存储介质
Liang et al. An extraction and classification algorithm for concrete cracks based on machine vision
CN114638294A (zh) 一种数据增强方法、装置、终端设备及存储介质
CN115471476A (zh) 一种部件缺陷检测方法、装置、设备及介质
CN116189220A (zh) 建筑构件检测识别方法、装置、计算机设备及存储介质
Tarabek Performance measurements of thinning algorithms
CN113537253B (zh) 一种红外图像目标检测方法、装置、计算设备及存储介质
CN111753723B (zh) 一种基于密度校准的指纹识别方法及装置
Meng et al. Defect detection of glassivation passivation parts wafer surface with random texture and different brightness

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22933171

Country of ref document: EP

Kind code of ref document: A1