CN114937003A - Multi-type defect detection system and method for glass panel - Google Patents
Multi-type defect detection system and method for glass panel Download PDFInfo
- Publication number
- CN114937003A CN114937003A CN202210380919.2A CN202210380919A CN114937003A CN 114937003 A CN114937003 A CN 114937003A CN 202210380919 A CN202210380919 A CN 202210380919A CN 114937003 A CN114937003 A CN 114937003A
- Authority
- CN
- China
- Prior art keywords
- image
- detection
- defect
- glass panel
- fitting
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
-
- 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
- G06T2207/30121—CRT, LCD or plasma display
-
- 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
- Y02P40/00—Technologies relating to the processing of minerals
- Y02P40/50—Glass production, e.g. reusing waste heat during processing or shaping
- Y02P40/57—Improving the yield, e-g- reduction of reject rates
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a multi-type defect detection system and method for a glass panel, and relates to the technical field of machine vision.A picture acquisition module acquires an original picture of the glass panel, a scene selection module selects a detection scene, and a detection gallery module judges whether the detection scene is passed; the drawing and inspecting partition module determines an image of the region of interest, and the preprocessing module performs preprocessing operation on the image of the region of interest; the threshold segmentation module calculates a threshold of the preprocessed image, and the threshold is utilized to carry out binarization operation on the preprocessed image; the measurement fitting module performs defect shape fitting on the binary image, calculates characteristic parameters of the defect shape, and determines the defect type of the original image of the glass panel in the current scene; and the judging and summarizing module is used for traversing all detection scenes, summarizing the defect detection results under multiple scenes and outputting the defect detection results. The invention realizes the detection of the multi-type defects of the glass panel under the multi-detection scene, and has short detection time, high detection efficiency and high reuse rate of the detection algorithm.
Description
Technical Field
The invention relates to the technical field of machine vision, in particular to a multi-type defect detection system and method for a glass panel.
Background
The liquid crystal display is a flat ultrathin display device composed of a certain number of color or black and white pixels, and is placed in front of a light source or a reflecting surface, an electric field is applied to the liquid crystal to change the molecular arrangement of the liquid crystal, and the liquid crystal display is matched with a polarizing plate to have the function of preventing light from passing through, namely, the light can smoothly pass through when the electric field is not applied, and the light is blocked when the electric field is applied. If the color filter is matched, the light transmittance of a certain color can be changed by changing the applied voltage, or the light transmittance can be changed by changing the voltage at two ends of the liquid crystal. With the mass production of lcd liquid crystal panels, the production quality of lcd liquid crystal panels needs to be detected more and more. The lcd panel defects are mainly classified into point defects, line defects, and mura-like defects. Wherein, the point defects are divided into bright points and dark points; the line defects are divided into vertical, horizontal and oblique line defects; mura-type defects are subdivided into multiple types of defects. Therefore, in the lcd panel inspection, the defects need to be inspected accordingly. The existing detection scheme is to generate a detection algorithm logic for each type of defects under different detection scenes. The multi-scene detection is set to highlight a certain type of defects by setting different external environments for the same workpiece, so that all abnormal defects generated in the production process are covered. For example, a set of specific algorithms needs to be developed for point defects in a white screen scene, and other detection operators need to be developed for other defects. The defect detection of the lcd panel has the characteristics of multiple defect types and multiple detection scenes, so that the reusability of the algorithm is greatly reduced by the conventional detection mode, the development period is prolonged, and the development cost is increased.
The prior art discloses a method and a system for detecting surface defects of a small-size glass panel, wherein the method comprises the following steps: firstly, extracting the edge outline and the sound hole outline of a panel, and extracting the gray level image of an internal area on the basis of the edge outline and the sound hole outline; further obtaining an internal region characteristic diagram, and calculating a gray average value of the characteristic diagram; and setting a threshold value based on the gray level mean value of the feature map for segmentation, performing BLOB analysis on the segmented result, judging according to the features such as the selected area, the aspect ratio and the like, eliminating pseudo defects such as dust and the like, and detecting the defects. The application can only detect the defects of the glass panel in a single scene, the detection efficiency is low, and the reuse rate of the detection algorithm is low.
Disclosure of Invention
In order to overcome the defect that the prior art can only detect single defects of the glass panel in a single scene, the invention provides a multi-type defect detection system and method for the glass panel, which can detect the multi-type defects of the glass panel in multiple scenes and improve the detection efficiency and the detection algorithm reuse rate.
In order to solve the technical problems, the technical scheme of the invention is as follows:
the invention provides a multi-type defect detection system for a glass panel, which comprises:
the image acquisition module is used for acquiring an original image of the glass panel;
the scene selection module is used for selecting a detection scene from the existing detection scene library;
the detection gallery module is used for judging whether an intermediate process image of the original image of the glass panel in the current detection scene exists in the detection gallery; if yes, returning to the scene selection module to select the next scene; otherwise, sending the original image of the glass panel to a drawing and inspecting partition module;
the drawing and inspecting partition module is used for drawing an interested area on the original image of the glass panel to obtain an interested area image;
the preprocessing module is used for preprocessing the image of the region of interest to obtain a preprocessed image;
the threshold segmentation module is used for calculating an image threshold of the preprocessed image and segmenting the preprocessed image by using the image threshold to obtain a binary image;
the measurement fitting module is used for fitting the defect shape of the binary image, calculating characteristic parameters of the fitted defect shape, and comparing the characteristic parameters with a preset parameter threshold value to obtain the defect type of the glass panel in the current detection scene;
the judging and summarizing module is used for judging whether all detection scenes are traversed or not; if not, the region-of-interest image, the preprocessed image and the binary image which are obtained in the current detection scene are used as intermediate process images, sent to a detection image library module for storage, returned to a scene selection module, selected to be the next detection scene, and the process is repeated; if so, summarizing the defect types of the glass panel under each detection scene obtained by the measurement fitting module and outputting.
Preferably, in the drawing and inspecting partition module, a specific method for obtaining the image of the region of interest includes:
on the original image of the glass panel, a detection area is obtained by automatically selecting a rectangle or manually selecting any shape; and adding a mask with an arbitrary shape on the detection area as a shielding area to form an interested area, and extracting the interested area to obtain an interested area image.
Because the glass panels are mostly rectangular, the rectangular detection area can be automatically selected according to the distribution positions of the glass panels in the whole visual field, or the rectangular detection area, the circular detection area, the polygonal detection area and the like can be manually selected according to the requirement; and a mask in any shape is added to the detection area to serve as a shielding area, so that unnecessary interference information is avoided, the range is narrowed for subsequent detection, and the overall detection speed is improved.
Preferably, the preprocessing operations include filtering operations, morphological processing, color space conversion and image enhancement operations;
the filtering operation comprises spatial filtering and frequency domain filtering;
the morphological treatment comprises expansion, corrosion, opening operation, closing operation and top cap transformation;
the color space conversion comprises the interconversion between an RGB space and a gray scale space, the interconversion between an HSV space and a gray scale space and the interconversion between the RGB space and the HSV space;
the image enhancement operation comprises a histogram equalization operation, an image normalization operation and an image operation;
and the preprocessing module selects at least one of the preprocessing operations according to the current detection scene, and processes the image of the region of interest to obtain a preprocessed image.
The purpose of the filtering operation is to eliminate noise interference on the original image of the glass panel, such as gaussian noise, salt and pepper noise, regular lines and the like; the purpose of morphological processing is to eliminate the effect of illumination non-uniformity; the color space conversion is to adapt to scenes needing other color spaces; the purpose of the image enhancement operation is to stretch the contrast of the original image, highlighting the defect of presenting a weak contrast. The at least one preprocessing operation is adopted to process the original image of the glass panel, so that the subsequent detection difficulty can be greatly reduced.
Preferably, the threshold segmentation module calculates an image threshold of the preprocessed image by using a global threshold method or a local threshold method, and segments the preprocessed image by using the image threshold to obtain a binarized image.
The global threshold method comprises an OTSU self-adaptive threshold method and a fixed threshold method; the local threshold method is to adopt the idea of sliding windows to calculate the mean value or median of each window as a threshold; for the preprocessed image with the gray histogram having the unimodal and bimodal characteristics, the OTSU self-adaptive threshold method is applied, so that not only can an ideal binary image be obtained, but also the preprocessed image can be segmented by calculating a proper threshold, and the robustness of the algorithm is greatly improved; for the preprocessed image with uneven illumination, if the preprocessed image is a gray-scale image and has poor consistency, a local threshold value method is applied; for pre-processed images that are color images, the tone scale based on RGB or HSV three channels is used as a threshold.
Preferably, the measurement fitting module performs defect shape fitting on the binary image according to a BLOB analysis algorithm, wherein the defect shape fitting comprises standard circle fitting, ellipse fitting, straight line fitting, bulk fitting, minimum circumscribed polygon fitting, maximum inscribed polygon fitting, minimum circumscribed rectangle fitting and maximum circumscribed rectangle fitting; calculating characteristic parameters of the fitted defect shape, including contrast, area, perimeter and aspect ratio;
comparing the characteristic parameters with a preset parameter threshold, wherein if the characteristic parameters are larger than the parameter threshold, the glass panel in the current detection scene has a defect type corresponding to the defect shape; otherwise, the defect type does not correspond to the defect type.
The invention also provides a multi-type defect detection method for the glass panel, which comprises the following steps:
s1: acquiring an original image of a glass panel;
s2: selecting a pre-detection scene from an existing detection scene library;
s3: judging whether an intermediate process image of the original image of the glass panel in the current detection scene exists in the detection image library; if yes, returning to the step S2 to select the next scene; otherwise, go to step S4;
s4: drawing an interested area on an original image of the glass panel to obtain an interested area image;
s5: preprocessing the image of the region of interest to obtain a preprocessed image;
s6: calculating an image threshold value of the preprocessed image, and segmenting the preprocessed image by using the image threshold value to obtain a binary image;
s7: performing defect shape fitting on the binary image, calculating characteristic parameters of the fitted defect shape, and comparing the characteristic parameters with a preset parameter threshold value to obtain the defect type of the glass panel in the current detection scene;
s8: judging whether all detection scenes are traversed; if not, taking the region-of-interest image, the preprocessed image and the binary image obtained in the current detection scene as intermediate process images, storing the intermediate process images into a detection gallery, returning to the step S2, selecting the next detection scene, and repeating the steps S3-S7; if yes, summarizing the defect types of the glass panel in each detection scene and outputting the defect types.
Preferably, in step S4, the specific method for obtaining the image of the region of interest is:
on the original image of the glass panel, a detection area is obtained by automatically selecting a rectangle or manually selecting any shape; and adding a mask with an arbitrary shape on the detection area as a shielding area to form an interested area, and extracting the interested area to obtain an interested area image.
Preferably, in the step S5, the preprocessing operation includes a filtering operation, a morphological processing, a color space conversion, and an image enhancement operation;
the filtering operation comprises spatial filtering and frequency domain filtering;
the morphological processing comprises expansion, corrosion, opening operation, closing operation and top-hat transformation;
the color space conversion comprises the interconversion between an RGB space and a gray scale space, the interconversion between an HSV space and the gray scale space and the interconversion between the RGB space and the HSV space;
the image enhancement operation comprises a histogram equalization operation, an image normalization operation and an image operation;
and the preprocessing module selects at least one of the preprocessing operations according to the current detection scene, and processes the image of the region of interest to obtain a preprocessed image.
Preferably, in step S6, an image threshold of the preprocessed image is calculated by using a global threshold method or a local threshold method, and the preprocessed image is segmented by using the image threshold, so as to obtain a binarized image.
Preferably, in step S7, fitting the defect shape of the binarized image according to a BLOB analysis algorithm, including standard circle fitting, ellipse fitting, straight line fitting, bulk fitting, minimum circumscribed polygon fitting, maximum inscribed polygon fitting, minimum circumscribed rectangle fitting, and maximum circumscribed rectangle fitting; calculating characteristic parameters of the fitted defect shape, including contrast, area, perimeter and aspect ratio;
comparing the characteristic parameters with a preset parameter threshold, wherein if the characteristic parameters are larger than the parameter threshold, the glass panel in the current detection scene has a defect type corresponding to the defect shape; otherwise, the defect type does not exist.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
the image acquisition module firstly acquires an original image of the glass panel, the scene selection module selects a detection scene, and the detection image library module judges whether the original image has already passed through the detection scene, so that the detection flow is saved; the drawing and inspecting partition module is used for determining an image of an interested area of the glass panel, reducing the detection range for subsequent detection and improving the overall detection speed; in addition, the interesting region graph can be reused in other detection scenes, so that the detection efficiency is improved; the preprocessing module is used for preprocessing the images of the region of interest, removing noise interference, enhancing contrast and reducing the subsequent detection difficulty; the threshold segmentation module calculates the threshold of the preprocessed image, and performs binarization operation on the preprocessed image by using the threshold to realize the effect of threshold segmentation; the measurement fitting module is used for fitting the defect shape of the binary image, calculating characteristic parameters of the defect shape and determining the defect type of the original image of the glass panel in the current scene; and the judging and summarizing module is used for traversing all detection scenes and summarizing the defect detection results in multiple scenes. The invention realizes the detection of the multi-type defects of the glass panel under the multi-detection scene, and has short detection time, high detection efficiency and high reuse rate of the detection algorithm.
Drawings
Fig. 1 is a schematic structural diagram of a multi-type defect detection system for a glass panel according to embodiment 1.
Fig. 2 is a schematic diagram of a detection architecture of the modules described in embodiment 2.
Fig. 3 is a flowchart of a multi-type defect detection method for a glass panel according to embodiment 3.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
for the purpose of better illustrating the embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product;
it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
Example 1
The present embodiment provides a multi-type defect detection system for a glass panel, as shown in fig. 1, including:
the image acquisition module is used for acquiring an original image of the glass panel;
the scene selection module is used for selecting a detection scene from the existing detection scene library;
the detection gallery module is used for judging whether an intermediate process image of the original image of the glass panel in the current detection scene exists in the detection gallery; if yes, returning to the scene selection module to select the next scene; otherwise, sending the original image of the glass panel to a drawing and inspecting partition module;
the drawing and inspecting partition module is used for drawing an interested area on the original image of the glass panel to obtain an interested area image;
the preprocessing module is used for preprocessing the image of the region of interest to obtain a preprocessed image;
the threshold segmentation module is used for calculating an image threshold of the preprocessed image and segmenting the preprocessed image by using the image threshold to obtain a binary image;
the measurement fitting module is used for fitting the defect shape of the binary image, calculating characteristic parameters of the fitted defect shape, and comparing the characteristic parameters with a preset parameter threshold value to obtain the defect type of the glass panel in the current detection scene;
the judging and summarizing module is used for judging whether all detection scenes are traversed or not; if not, the region-of-interest image, the preprocessed image and the binary image which are obtained in the current detection scene are used as intermediate process images, sent to a detection image library module for storage, returned to a scene selection module, selected to be the next detection scene, and the process is repeated; if yes, summarizing the defect types of the glass panel under each detection scene obtained by the measurement fitting module and outputting the defect types.
In a specific implementation process, the image acquisition module in the embodiment firstly acquires an original image of the glass panel, the scene selection module selects a detection scene, and the detection gallery module judges whether the original image has already passed through the detection scene, so that a detection flow is saved; the drawing and inspecting partition module is used for determining an image of an interested area of the glass panel, reducing the detection range for subsequent detection and improving the overall detection speed; in addition, the interesting region graph can be reused in other detection scenes, so that the detection efficiency is improved; the preprocessing module is used for preprocessing the images of the region of interest, removing noise interference, enhancing contrast and reducing subsequent detection difficulty; the threshold segmentation module calculates the threshold of the preprocessed image, and performs binarization operation on the preprocessed image by using the threshold to realize the effect of threshold segmentation; the measurement fitting module is used for fitting the defect shape of the binary image, calculating characteristic parameters of the defect shape and determining the defect type of the original image of the glass panel in the current scene; and the judging and summarizing module is used for traversing all detection scenes and summarizing the defect detection results in multiple scenes.
Example 2
Taking a mobile phone screen LCD glass panel as an example, the present embodiment provides a multi-type defect detection system for a glass panel, including:
the image acquisition module is used for acquiring an original image of the glass panel;
the scene selection module is used for selecting a detection scene from the existing detection scene library;
the method comprises the following steps of storing dust screen scenes, gray screen scenes, white screen scenes and red screen scenes which are common detection scenes of an LCD glass panel of a mobile phone screen into a detection scene library;
the detection gallery module is used for judging whether an intermediate process image of the original image of the glass panel in the current detection scene exists in the detection gallery; if yes, returning to the scene selection module to select the next scene; otherwise, sending the original image of the glass panel to a drawing and checking partition module;
FIG. 2 is a schematic diagram of a detection architecture of each module;
the drawing and inspecting partition module is used for drawing an interested area on the original image of the glass panel to obtain an interested area image;
on the original image of the glass panel, a detection area is obtained by automatically selecting a rectangle or manually selecting any shape; and adding a mask with an arbitrary shape on the detection area as a shielding area to form an interested area, and extracting the interested area to obtain an interested area image.
The preprocessing module is used for preprocessing the image of the region of interest to obtain a preprocessed image;
the preprocessing operation comprises a filtering operation, a morphological processing operation, a color space conversion operation and an image enhancement operation;
the filtering operation comprises spatial filtering and frequency domain filtering;
the morphological processing comprises expansion, corrosion, opening operation, closing operation and top-hat transformation;
the color space conversion comprises the interconversion between an RGB space and a gray scale space, the interconversion between an HSV space and a gray scale space and the interconversion between the RGB space and the HSV space;
the image enhancement operation comprises a histogram equalization operation, an image normalization operation and an image operation;
and the preprocessing module selects at least one from the preprocessing operations according to the current detection scene, and processes the images of the interested area to obtain preprocessed images.
The threshold segmentation module is used for calculating an image threshold of the preprocessed image by using a global threshold method or a local threshold method, and segmenting the preprocessed image by using the image threshold to obtain a binary image;
the measurement fitting module is used for fitting the defect shape of the binary image according to a BLOB analysis algorithm, calculating characteristic parameters of the fitted defect shape, and comparing the characteristic parameters with a preset parameter threshold value to obtain the defect type of the glass panel under the current detection scene;
the common defect types of the LCD glass panel of the mobile phone screen are three, namely point defects, bulk defects and line defects; characteristic parameters of the point defect include contrast, area and perimeter; characteristic parameters of the bulk defect include perimeter, contrast and area; characteristic parameters of the line defect include aspect ratio, contrast, area and perimeter; obtaining characteristic parameters of all point, line and cluster defects on the binary image through fitting processing, comparing the characteristic parameters with a preset parameter threshold, and if the characteristic parameters are larger than the parameter threshold, determining that the glass panel has a defect type corresponding to the defect shape in the current detection scene; otherwise, the defect type does not correspond to the defect type.
The judging and summarizing module is used for judging whether all detection scenes are traversed or not; if not, the region-of-interest image, the preprocessed image and the binary image which are obtained in the current detection scene are used as intermediate process images, sent to a detection image library module for storage, returned to a scene selection module, selected to be the next detection scene, and the process is repeated; if so, summarizing the defect types of the glass panel under each detection scene obtained by the measurement fitting module and outputting.
Example 3
The embodiment provides a method for detecting multiple types of defects of a glass panel, as shown in fig. 3, including:
s1: acquiring an original image of a glass panel;
s2: selecting a detection scene from an existing detection scene library;
s3: judging whether an intermediate process image of the original image of the glass panel in the current detection scene exists in the detection image library; if yes, returning to the step S2 to select the next scene; otherwise, go to step S4;
s4: drawing an interested area on an original image of the glass panel to obtain an interested area image;
s5: carrying out preprocessing operation on the image of the region of interest to obtain a preprocessed image;
s6: calculating an image threshold value of the preprocessed image, and segmenting the preprocessed image by using the image threshold value to obtain a binary image;
s7: performing defect shape fitting on the binary image, calculating characteristic parameters of the fitted defect shape, and comparing the characteristic parameters with a preset parameter threshold value to obtain the defect type of the glass panel in the current detection scene;
s8: judging whether all detection scenes are traversed or not; if not, taking the region-of-interest image, the preprocessed image and the binarized image obtained in the current detection scene as an intermediate process image, storing the intermediate process image into a detection image library, returning to the step S2, selecting the next detection scene, and repeating the steps S3-S7; if yes, summarizing the defect types of the glass panel under each detection scene and outputting.
In step S4, the specific method for obtaining the image of the region of interest includes:
on the original image of the glass panel, a detection area is obtained by automatically selecting a rectangle or manually selecting any shape; and adding a mask with an arbitrary shape on the detection area as a shielding area to form an interested area, and extracting the interested area to obtain an interested area image.
In step S5, the preprocessing operations include filtering operations, morphological processing, color space conversion, and image enhancement operations;
the filtering operation comprises spatial filtering and frequency domain filtering;
the morphological processing comprises expansion, corrosion, opening operation, closing operation and top-hat transformation;
the color space conversion comprises the interconversion between an RGB space and a gray scale space, the interconversion between an HSV space and a gray scale space and the interconversion between the RGB space and the HSV space;
the image enhancement operation comprises a histogram equalization operation, an image normalization operation and an image operation;
and the preprocessing module selects at least one of the preprocessing operations according to the current detection scene, and processes the image of the region of interest to obtain a preprocessed image.
In step S6, an image threshold of the preprocessed image is calculated by using a global threshold method or a local threshold method, and the preprocessed image is segmented by using the image threshold, so as to obtain a binarized image.
In the step S7, performing defect shape fitting on the binarized image according to a BLOB analysis algorithm, including standard circle fitting, ellipse fitting, straight line fitting, bulk fitting, minimum circumscribed polygon fitting, maximum inscribed polygon fitting, minimum circumscribed rectangle fitting, and maximum circumscribed rectangle fitting; calculating characteristic parameters of the fitted defect shape, including contrast, area, perimeter and aspect ratio;
comparing the characteristic parameters with a preset parameter threshold, and if the characteristic parameters are greater than the parameter threshold, determining that the glass panel has a defect type corresponding to the defect shape in the current detection scene; otherwise does not have a corresponding defect type
The same or similar reference numerals correspond to the same or similar parts;
the terms describing positional relationships in the drawings are for illustrative purposes only and are not to be construed as limiting the patent;
it should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. This need not be, nor should it be exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (10)
1. A multi-type defect detection system for a glass panel, comprising:
the image acquisition module is used for acquiring an original image of the glass panel;
the scene selection module is used for selecting a detection scene from the existing detection scene library;
the detection gallery module is used for judging whether an intermediate process image of the original image of the glass panel in the current detection scene exists in the detection gallery; if yes, returning to the scene selection module to select the next scene; otherwise, sending the original image of the glass panel to a drawing and inspecting partition module;
the drawing and inspecting partition module is used for drawing an interested area on the original image of the glass panel to obtain an interested area image;
the preprocessing module is used for preprocessing the image of the region of interest to obtain a preprocessed image;
the threshold segmentation module is used for calculating an image threshold of the preprocessed image and segmenting the preprocessed image by using the image threshold to obtain a binary image;
the measurement fitting module is used for fitting the defect shape of the binary image, calculating characteristic parameters of the fitted defect shape, and comparing the characteristic parameters with a preset parameter threshold value to obtain the defect type of the glass panel in the current detection scene;
the judging and summarizing module is used for judging whether all detection scenes are traversed or not; if not, the region-of-interest image, the preprocessed image and the binary image which are obtained in the current detection scene are used as intermediate process images, sent to a detection image library module for storage, returned to a scene selection module, selected to be the next detection scene, and the process is repeated; if so, summarizing the defect types of the glass panel under each detection scene obtained by the measurement fitting module and outputting.
2. The system of claim 1, wherein the specific method of obtaining the image of the region of interest in the inspection sub-area module is:
on the original image of the glass panel, a detection area is obtained by automatically selecting a rectangle or manually selecting any shape; and adding a mask with an arbitrary shape on the detection area as a shielding area to form an interested area, and extracting the interested area to obtain an interested area image.
3. The multi-type defect detection system for glass panels of claim 1, wherein said preprocessing operations comprise filtering operations, morphological processing, color space conversion and image enhancement operations;
the filtering operation comprises spatial filtering and frequency domain filtering;
the morphological processing comprises expansion, corrosion, opening operation, closing operation and top-hat transformation;
the color space conversion comprises the interconversion between an RGB space and a gray scale space, the interconversion between an HSV space and a gray scale space and the interconversion between the RGB space and the HSV space;
the image enhancement operation comprises a histogram equalization operation, an image normalization operation and an image operation;
and the preprocessing module selects at least one from the preprocessing operations according to the current detection scene, and processes the images of the interested area to obtain preprocessed images.
4. The system of claim 1, wherein the threshold segmentation module computes an image threshold of the pre-processed image using a global threshold method or a local threshold method, and segments the pre-processed image using the image threshold to obtain a binarized image.
5. The multi-type defect detection system for glass panels as claimed in claim 1, wherein said measurement fitting module performs defect shape fitting on the binarized image according to BLOB analysis algorithm including standard circle fitting, ellipse fitting, straight line fitting, BLOB fitting, minimum circumscribed polygon fitting, maximum inscribed polygon fitting, minimum circumscribed rectangle fitting, maximum circumscribed rectangle fitting; calculating characteristic parameters of the fitted defect shape, including contrast, area, perimeter and aspect ratio;
comparing the characteristic parameters with a preset parameter threshold, wherein if the characteristic parameters are larger than the parameter threshold, the glass panel in the current detection scene has a defect type corresponding to the defect shape; otherwise, the defect type does not correspond to the defect type.
6. A method for detecting multiple types of defects of a glass panel is characterized by comprising the following steps:
s1: acquiring an original image of a glass panel;
s2: selecting a detection scene from an existing detection scene library;
s3: judging whether an intermediate process image of the original image of the glass panel in the current detection scene exists in the detection image library; if yes, returning to the step S2 to select the next scene; otherwise, go to step S4;
s4: drawing an interested area on an original image of the glass panel to obtain an interested area image;
s5: preprocessing the image of the region of interest to obtain a preprocessed image;
s6: calculating an image threshold value of the preprocessed image, and segmenting the preprocessed image by using the image threshold value to obtain a binary image;
s7: defect shape fitting is carried out on the binary image, characteristic parameters of the fitted defect shape are calculated and compared with a preset parameter threshold value, and the defect type of the glass panel in the current detection scene is obtained;
s8: judging whether all detection scenes are traversed or not; if not, taking the region-of-interest image, the preprocessed image and the binarized image obtained in the current detection scene as an intermediate process image, storing the intermediate process image into a detection image library, returning to the step S2, selecting the next detection scene, and repeating the steps S3-S7; if yes, summarizing the defect types of the glass panel under each detection scene and outputting.
7. The system of claim 6, wherein the specific method for obtaining the image of the region of interest in step S4 is as follows:
on the original image of the glass panel, a detection area is obtained by automatically selecting a rectangle or manually selecting any shape; and adding a mask with an arbitrary shape on the detection area as a shielding area to form an interested area, and extracting the interested area to obtain an interested area image.
8. The multi-type defect detection system for glass panels as claimed in claim 6, wherein in said step S5, the preprocessing operations include filtering operations, morphological processing, color space conversion and image enhancement operations;
the filtering operation comprises spatial filtering and frequency domain filtering;
the morphological processing comprises expansion, corrosion, opening operation, closing operation and top-hat transformation;
the color space conversion comprises the interconversion between an RGB space and a gray scale space, the interconversion between an HSV space and a gray scale space and the interconversion between the RGB space and the HSV space;
the image enhancement operation comprises a histogram equalization operation, an image normalization operation and an image operation;
and the preprocessing module selects at least one of the preprocessing operations according to the current detection scene, and processes the image of the region of interest to obtain a preprocessed image.
9. The system for multi-type defect detection on glass panels as claimed in claim 6, wherein in step S6, the image threshold of the pre-processed image is calculated by using global threshold method or local threshold method, and the pre-processed image is segmented by using the image threshold to obtain the binary image.
10. The system for multi-type defect detection of glass panels as claimed in claim 6, wherein in said step S7, defect shape fitting is performed on the binarized image according to BLOB analysis algorithm, including standard circle fitting, ellipse fitting, straight line fitting, bulk fitting, minimum circumscribed polygon fitting, maximum inscribed polygon fitting, minimum circumscribed rectangle fitting, maximum circumscribed rectangle fitting; calculating characteristic parameters of the fitted defect shape, including contrast, area, perimeter and aspect ratio;
comparing the characteristic parameters with a preset parameter threshold, and if the characteristic parameters are greater than the parameter threshold, determining that the glass panel has a defect type corresponding to the defect shape in the current detection scene; otherwise, the defect type does not correspond to the defect type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210380919.2A CN114937003A (en) | 2022-04-12 | 2022-04-12 | Multi-type defect detection system and method for glass panel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210380919.2A CN114937003A (en) | 2022-04-12 | 2022-04-12 | Multi-type defect detection system and method for glass panel |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114937003A true CN114937003A (en) | 2022-08-23 |
Family
ID=82862868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210380919.2A Pending CN114937003A (en) | 2022-04-12 | 2022-04-12 | Multi-type defect detection system and method for glass panel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114937003A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116309563A (en) * | 2023-05-17 | 2023-06-23 | 成都数之联科技股份有限公司 | Method, device, medium, equipment and program product for detecting defect of panel edge |
| CN118334033A (en) * | 2024-06-14 | 2024-07-12 | 成都图灵威视科技有限公司 | Line defect detection method and system based on industrial product defects |
-
2022
- 2022-04-12 CN CN202210380919.2A patent/CN114937003A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116309563A (en) * | 2023-05-17 | 2023-06-23 | 成都数之联科技股份有限公司 | Method, device, medium, equipment and program product for detecting defect of panel edge |
| CN116309563B (en) * | 2023-05-17 | 2023-07-18 | 成都数之联科技股份有限公司 | Method, device, medium, equipment and program product for detecting defect of panel edge |
| CN118334033A (en) * | 2024-06-14 | 2024-07-12 | 成都图灵威视科技有限公司 | Line defect detection method and system based on industrial product defects |
| CN118334033B (en) * | 2024-06-14 | 2024-08-23 | 成都图灵威视科技有限公司 | Line defect detection method and system based on industrial product defects |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110766679B (en) | Lens contamination detection method and device and terminal equipment | |
| CN113781402B (en) | Method and device for detecting scratch defects on chip surface and computer equipment | |
| CN115908269B (en) | Visual defect detection method, visual defect detection device, storage medium and computer equipment | |
| CN105894036B (en) | A kind of characteristics of image template matching method applied to mobile phone screen defects detection | |
| CN111612781A (en) | Screen defect detection method and device and head-mounted display equipment | |
| TWI655586B (en) | Method and device for detecting specific identification image in predetermined area | |
| CN108416766B (en) | Double-side light-entering type light guide plate defect visual detection method | |
| CN109727233B (en) | LCD defect detection method | |
| CN115631117B (en) | Image enhancement method, device, detection system and storage medium for defect detection | |
| CN114937003A (en) | Multi-type defect detection system and method for glass panel | |
| CN111583223A (en) | Defect detection method, defect detection device, computer equipment and computer readable storage medium | |
| CN113646801A (en) | Defect detection method and device for defect image and computer readable storage medium | |
| WO2007061779A1 (en) | Shadow detection in images | |
| US5917957A (en) | Method of and apparatus for processing an image | |
| CN110060239B (en) | Defect detection method for bottle opening of bottle | |
| CN110648330B (en) | Defect detection method for camera glass | |
| CN108107611B (en) | Self-adaptive defect detection method and device and electronic equipment | |
| CN115205223A (en) | Visual detection method and device for transparent object, computer equipment and medium | |
| CN116883408B (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
| CN114998217A (en) | Method for determining defect grade of glass substrate, computer device and storage medium | |
| CN114298985B (en) | Defect detection method, device, equipment and storage medium | |
| JP2005165387A (en) | Method and device for detecting stripe defective of picture and display device | |
| US20240320806A1 (en) | Image processing method and apparatus, electronic device and computer readable storage medium | |
| CN112329572B (en) | Rapid static living body detection method and device based on frame and flash point | |
| CN113744200B (en) | Camera dirt detection method, device and equipment |
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 |