CN112801930A - Method and equipment for detecting warp and weft defects of fabric in backlight based on machine vision - Google Patents
Method and equipment for detecting warp and weft defects of fabric in backlight based on machine vision Download PDFInfo
- Publication number
- CN112801930A CN112801930A CN201911028346.1A CN201911028346A CN112801930A CN 112801930 A CN112801930 A CN 112801930A CN 201911028346 A CN201911028346 A CN 201911028346A CN 112801930 A CN112801930 A CN 112801930A
- Authority
- CN
- China
- Prior art keywords
- image
- filtering
- filter
- column
- row
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30124—Fabrics; Textile; Paper
Abstract
The invention discloses a method and a device for detecting warp and weft flaws in backlight of a fabric based on machine vision, wherein the method comprises the following steps: acquiring a backlight image of the fabric; filtering the backlight image by adopting a single-row filtering method and a single-column filtering method respectively to obtain a single-row filtering image and a single-column filtering image; making a difference between the single-row filtering image and the single-column filtering image to obtain a filtering difference image; performing Gaussian filtering on the filtering difference image to obtain a Gaussian filtering image; and carrying out canny edge detection based on the Gaussian filtering image to obtain the coordinates of the edge points of the longitude and latitude flaws. According to the method and the device for detecting the warp and weft defects of the fabric based on the machine vision, the warp and weft defects of the fabric can be efficiently, accurately and inexpensively detected in a detection mode without human intervention basically, the high quality of a fabric product is ensured, and the defect detection rate is extremely high and the cost is low.
Description
Technical Field
The invention relates to a warp and weft flaw detection technology of a fabric, in particular to a method and equipment for detecting warp and weft flaws of the fabric in a backlight mode based on machine vision.
Background
The development of domestic fabric and warp knitting industry is fast, but how to improve and ensure the product quality and reduce the production cost is always the key point of related enterprise research. The quality assessment, price and income of enterprises of fabrics such as wool fabrics have great influence on the defects of the fabric products.
At present, most of the defects of the fabrics are still detected by adopting a manual visual inspection mode. The disadvantages of manual detection are: the limitation of visual precision and easy fatigue cause unstable detection; the detection accuracy is not high, and the detection omission easily occurs; the labor cost and the like are high.
Especially, when the backlight defect detection is performed, due to the light transmittance of the fabric, a large amount of light is distributed in a point manner, and the defect detection accuracy may be seriously disturbed due to the characteristic. But at the same time, the light transmission at the non-flaw points is obviously different from the light transmission at the longitude and latitude flaws in distribution.
In view of the above, it is highly desirable to design a new automatic fabric back-lighting warp and weft defect detection technology.
Disclosure of Invention
The invention aims to overcome the defects that the detection accuracy of the existing fabric flaw detection technology is not high and the stable high quality of a product is difficult to ensure, and provides a novel machine vision-based method and equipment for detecting the warp and weft flaws in the fabric in a backlight mode.
The invention solves the technical problems by adopting the following technical scheme:
the invention provides a method for detecting warp and weft defects in a backlight of a fabric based on machine vision, which is characterized by comprising the following steps of:
acquiring a backlight image of the fabric;
respectively adopting a single-row filtering method and a single-column filtering method to filter the backlight image so as to obtain a single-row filtering image and a single-column filtering image;
making a difference between the single-row filtering image and the single-column filtering image to obtain a filtering difference making image;
performing Gaussian filtering on the filtering difference image to obtain a Gaussian filtering image;
and carrying out canny edge detection based on the Gaussian filtering image to obtain the coordinates of the edge points of the longitude and latitude flaws.
According to some embodiments of the present invention, in the filtering process of the backlight image using the single-line filtering method and the single-column filtering method, a single-line filter and a single-column filter are respectively configured, wherein widths of the single-line filter and the single-column filter are respectively configured as a plurality of pixels.
According to some embodiments of the invention, the single row of filters and the single column of filters are configured to have the same width.
According to some embodiments of the invention, the single row filter and the single column filter are each configured to have a width of 5-20 pixels.
The invention also provides a machine vision-based backlight warp and weft defect detection device for the fabric, which is characterized by comprising the following components:
the device comprises a light source and a backlight shooting device, wherein the light source and the backlight shooting device are configured to be capable of shooting and acquiring a backlight image of a fabric to be detected by the backlight shooting device;
the single-row filtering module and the single-column filtering module are configured to be capable of respectively performing filtering processing on the backlight image so as to obtain a single-row filtering image and a single-column filtering image;
a differencing module configured to be capable of differencing the single row filtered image and the single column filtered image to obtain a filtered differencing image;
a Gaussian filter module configured to Gaussian filter the filtered difference image to obtain a Gaussian filtered image;
a canny edge detection module configured to perform canny edge detection based on the Gaussian filtered image to obtain edge point coordinates of longitude and latitude flaws.
According to some embodiments of the invention, the single row and column filter modules are configured to be capable of constructing a single row filter and a single column filter, respectively, wherein the single row and column filters are a plurality of pixels in width.
According to some embodiments of the invention, the single row of filters and the single column of filters have the same width.
According to some embodiments of the invention, the single row filter and the single column filter each have a width of 5-20 pixels.
On the basis of the common knowledge in the field, the above preferred conditions can be combined randomly to obtain the preferred embodiments of the invention.
The positive progress effects of the invention are as follows:
according to the method and the device for detecting the warp and weft defects of the fabric based on the machine vision, the warp and weft defects of the fabric can be efficiently, accurately and inexpensively detected in a detection mode without human intervention basically, the high quality of a fabric product is ensured, and the defect detection rate is extremely high and the cost is low.
Drawings
Fig. 1 is a flowchart of a method for machine vision based backlit warp and weft defect detection of a fabric according to a preferred embodiment of the present invention.
Detailed Description
The following detailed description of the preferred embodiments of the present invention, taken in conjunction with the accompanying drawings, is intended to be illustrative, and not restrictive, and any other similar items may be considered within the scope of the present invention.
In the following detailed description, directional terms, such as "left", "right", "upper", "lower", "front", "rear", and the like, are used with reference to the orientation as illustrated in the drawings. The components of various embodiments of the present invention can be positioned in a number of different orientations and the directional terminology is used for purposes of illustration and is in no way limiting.
Referring to fig. 1, a method for detecting a warp and weft defect in a fabric based on machine vision according to a preferred embodiment of the present invention includes:
acquiring a backlight image of the fabric;
respectively adopting a single-row filtering method and a single-column filtering method to filter the backlight image so as to obtain a single-row filtering image and a single-column filtering image;
making a difference between the single-row filtering image and the single-column filtering image to obtain a filtering difference making image;
performing Gaussian filtering on the filtering difference image to obtain a Gaussian filtering image;
and carrying out canny edge detection based on the Gaussian filtering image to obtain the coordinates of the edge points of the longitude and latitude flaws.
In consideration of the obvious difference in distribution of light transmission at non-defect points of the fabric and light transmission at warp and weft defects in the backlight defect detection, the backlight warp and weft defect detection algorithm according to the preferred embodiment of the invention can achieve the defect detection rate and detection accuracy which are obviously higher than those of the existing detection method while reducing the detection cost.
Canny edge detection, which generally includes gradient value and gradient direction calculation for an image, filtering for non-maximum values of the gradient values to filter out points that do not belong to an edge, uses upper and lower thresholds together to detect and judge edge points or edge coordinates. The canny edge detection technique is per se prior art and will not be described further herein.
Wherein, in filtering the backlight image using a single-line filtering method and a single-column filtering method, a single-line filter and a single-column filter are respectively configured, wherein widths of the single-line filter and the single-column filter are respectively configured as a plurality of pixels.
For example, when the above-mentioned method for detecting warp and weft defects in backlight is applied to some application examples of wool fabrics, it can be specifically designed to have the following steps:
step one, acquiring a backlight image of the fabric, wherein the backlight image can be recorded as an image I (x, y). .
And step two, single-line filtering to obtain a single-line filtering image R (x, y). Here, for example, the width of a warp and weft flaw of a wool fabric to be detected is 15 pixels or more, so when constructing a filter, the filter width is selected to be 15. A filter constructed as followsFiltering the image by using the filterTo a single-line filtered image R (x, y) ═ g (x) I (x, y).
And step three, single-row filtering to obtain a single-row filtering image C (x, y). Constructing a single-row filter as above, may for example construct the filter to have a width of 15, resulting in a filterThe image is filtered using this filter to obtain a single filtered image C (x, y) ═ m (x) × I (x, y).
And step four, performing subtraction to obtain a filtered difference image S (x, y), wherein S (x, y) is abs (R (x, y) -C (x, y)).
And step five, Gaussian filtering to obtain a Gaussian filtering image M (x, y). By gaussian filtering the image S (x, y), some thin line disturbances to the detection can be filtered out.
And sixthly, detecting a canny edge based on the Gaussian filtered image M (x, y), and performing canny edge detection on the M (x, y) to return the coordinates of the flaw edge point.
Optionally, the single row of filters and the single column of filters are configured to have the same width. Also optionally, the single row filter and the single column filter are each configured to have a width of 5-20 pixels.
According to the method and the device for detecting the warp and weft defects of the fabric based on the machine vision, the warp and weft defects of the fabric can be efficiently, accurately and inexpensively detected in a detection mode without human intervention basically, the high quality of a fabric product is ensured, and the defect detection rate is extremely high and the cost is low.
In some application examples of the wool fabric, the detection rate of warp and weft flaws in practical tests can reach more than 94%, and the coordinates of the finally obtained flaw edge points have high precision and reliability.
The machine vision based fabric backlighting warp and weft defect detection apparatus according to some preferred embodiments of the present invention comprises:
the device comprises a light source and a backlight shooting device, wherein the light source and the backlight shooting device are configured to be capable of shooting and acquiring a backlight image of a fabric to be detected by the backlight shooting device;
the single-row filtering module and the single-column filtering module are configured to be capable of respectively performing filtering processing on the backlight image so as to obtain a single-row filtering image and a single-column filtering image;
a differencing module configured to be capable of differencing the single row filtered image and the single column filtered image to obtain a filtered differencing image;
a Gaussian filter module configured to Gaussian filter the filtered difference image to obtain a Gaussian filtered image;
a canny edge detection module configured to perform canny edge detection based on the Gaussian filtered image to obtain edge point coordinates of longitude and latitude flaws.
Wherein the single row and column filter modules are configured to be capable of constructing a single row and column filter, respectively, wherein the single row and column filter are a plurality of pixels in width.
Preferably, the single row of filters and the single column of filters have the same width.
Preferably, the single row filter and the single column filter each have a width of 5-20 pixels.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that these are by way of example only, and that the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the spirit and scope of the invention, and these changes and modifications are within the scope of the invention.
Claims (8)
1. The method for detecting the warp and weft defects of the fabric in the backlight based on the machine vision is characterized by comprising the following steps of:
acquiring a backlight image of the fabric;
respectively adopting a single-row filtering method and a single-column filtering method to filter the backlight image so as to obtain a single-row filtering image and a single-column filtering image;
making a difference between the single-row filtering image and the single-column filtering image to obtain a filtering difference making image;
performing Gaussian filtering on the filtering difference image to obtain a Gaussian filtering image;
and carrying out canny edge detection based on the Gaussian filtering image to obtain the coordinates of the edge points of the longitude and latitude flaws.
2. The backlight longitude and latitude defect detection method according to claim 1, wherein in the filtering process of the backlight image by the single-row filtering method and the single-column filtering method, a single-row filter and a single-column filter are respectively constructed, wherein widths of the single-row filter and the single-column filter are respectively constructed as a plurality of pixels.
3. The backlight longitude and latitude flaw detection method of claim 2 wherein said single row filter and said single column filter are configured to have the same width.
4. The backlight longitude and latitude flaw detection method of claim 2 wherein said single row filter and said single column filter are each configured to have a width of 5-20 pixels.
5. The utility model provides a longitude and latitude flaw check out test set is shaded of fabric based on machine vision which characterized in that, longitude and latitude flaw check out test set is shaded includes:
the device comprises a light source and a backlight shooting device, wherein the light source and the backlight shooting device are configured to be capable of shooting and acquiring a backlight image of a fabric to be detected by the backlight shooting device;
the single-row filtering module and the single-column filtering module are configured to be capable of respectively performing filtering processing on the backlight image so as to obtain a single-row filtering image and a single-column filtering image;
a differencing module configured to be capable of differencing the single row filtered image and the single column filtered image to obtain a filtered differencing image;
a Gaussian filter module configured to Gaussian filter the filtered difference image to obtain a Gaussian filtered image;
a canny edge detection module configured to perform canny edge detection based on the Gaussian filtered image to obtain edge point coordinates of longitude and latitude flaws.
6. The backlit latitude and longitude flaw detection apparatus of claim 5 wherein the single row and column filter modules are configured to be capable of constructing a single row filter and a single column filter, respectively, wherein the single row and column filters are a plurality of pixels in width.
7. The backlit warp and weft defect detection apparatus of claim 6, wherein the single row filter and the single column filter have the same width.
8. The backlit latitude and longitude imperfection detection device of claim 6, wherein the single row filter and the single column filter each have a width of 5-20 pixels.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911028346.1A CN112801930A (en) | 2019-10-28 | 2019-10-28 | Method and equipment for detecting warp and weft defects of fabric in backlight based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911028346.1A CN112801930A (en) | 2019-10-28 | 2019-10-28 | Method and equipment for detecting warp and weft defects of fabric in backlight based on machine vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112801930A true CN112801930A (en) | 2021-05-14 |
Family
ID=75803282
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911028346.1A Pending CN112801930A (en) | 2019-10-28 | 2019-10-28 | Method and equipment for detecting warp and weft defects of fabric in backlight based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112801930A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116165216A (en) * | 2023-03-16 | 2023-05-26 | 苏州鼎纳自动化技术有限公司 | Liquid crystal display micro scratch flaw 3D detection method, system and computing equipment |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101035946A (en) * | 2004-04-19 | 2007-09-12 | 3M创新有限公司 | Apparatus and method for the automated marking of defects on webs of material |
CN101866427A (en) * | 2010-07-06 | 2010-10-20 | 西安电子科技大学 | Method for detecting and classifying fabric defects |
CN103217436A (en) * | 2013-03-06 | 2013-07-24 | 京东方科技集团股份有限公司 | Backlight module group defect detection method and equipment |
CN103369347A (en) * | 2012-03-05 | 2013-10-23 | 苹果公司 | Camera blemish defects detection |
CN104933728A (en) * | 2015-07-13 | 2015-09-23 | 天津理工大学 | Mixed motion target detection method |
CN105203547A (en) * | 2015-08-26 | 2015-12-30 | 李云栋 | Cloth flaw detection method and device based on intelligent visual sensor |
CN105740876A (en) * | 2014-12-10 | 2016-07-06 | 阿里巴巴集团控股有限公司 | Image preprocessing method and device |
CN109300127A (en) * | 2018-09-21 | 2019-02-01 | 中兴新通讯有限公司 | Defect inspection method, device, computer equipment and storage medium |
-
2019
- 2019-10-28 CN CN201911028346.1A patent/CN112801930A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101035946A (en) * | 2004-04-19 | 2007-09-12 | 3M创新有限公司 | Apparatus and method for the automated marking of defects on webs of material |
CN101866427A (en) * | 2010-07-06 | 2010-10-20 | 西安电子科技大学 | Method for detecting and classifying fabric defects |
CN103369347A (en) * | 2012-03-05 | 2013-10-23 | 苹果公司 | Camera blemish defects detection |
CN103217436A (en) * | 2013-03-06 | 2013-07-24 | 京东方科技集团股份有限公司 | Backlight module group defect detection method and equipment |
CN105740876A (en) * | 2014-12-10 | 2016-07-06 | 阿里巴巴集团控股有限公司 | Image preprocessing method and device |
CN104933728A (en) * | 2015-07-13 | 2015-09-23 | 天津理工大学 | Mixed motion target detection method |
CN105203547A (en) * | 2015-08-26 | 2015-12-30 | 李云栋 | Cloth flaw detection method and device based on intelligent visual sensor |
CN109300127A (en) * | 2018-09-21 | 2019-02-01 | 中兴新通讯有限公司 | Defect inspection method, device, computer equipment and storage medium |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116165216A (en) * | 2023-03-16 | 2023-05-26 | 苏州鼎纳自动化技术有限公司 | Liquid crystal display micro scratch flaw 3D detection method, system and computing equipment |
CN116165216B (en) * | 2023-03-16 | 2023-08-04 | 苏州鼎纳自动化技术有限公司 | Liquid crystal display micro scratch flaw 3D detection method, system and computing equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105158258B (en) | A kind of bamboo cane detection method of surface flaw based on computer vision | |
CN107678192B (en) | Mura defect detection method based on machine vision | |
Watson | 31.1: Invited paper: The spatial standard observer: A human vision model for display inspection | |
CN109345528B (en) | Display screen defect detection method and device based on human visual characteristics | |
CN113989228A (en) | Method for detecting defect area of color texture fabric based on self-attention | |
CN109727233B (en) | LCD defect detection method | |
CN115330784B (en) | Cloth surface defect detection method | |
CN105241638A (en) | Vision-based quick LED module brightness uniformity detection method | |
CN107515481B (en) | Display panel detection method and device | |
CN104933978A (en) | Detection method and detection apparatus for display panel | |
CN106054421A (en) | Liquid crystal display panel defect detecting method and device | |
CN109886960A (en) | The method of glass edge defects detection based on machine vision | |
CN115100200B (en) | Optical fiber defect detection method and system based on optical means | |
CN103792705A (en) | Detecting method and detecting device for detecting substrate defects | |
CN108445020B (en) | Glass substrate defect aggregation identification method | |
CN112819844B (en) | Image edge detection method and device | |
CN104568956A (en) | Machine vision based detection method for strip steel surface defects | |
CN104459421A (en) | Diode polarity detecting method and system | |
CN115861310B (en) | Method for detecting textile defects on surface of bed sheet | |
CN105044127A (en) | OLED micro-display defect detection device and detection method | |
CN112801930A (en) | Method and equipment for detecting warp and weft defects of fabric in backlight based on machine vision | |
CN102811363A (en) | Method for measuring 3D (Three-Dimensional) crosstalk value of liquid crystal display | |
CN109426013B (en) | Method and device for analyzing and detecting and repairing defects of color film substrate | |
CN105371955B (en) | A kind of dye aberration detection device and detection method | |
CN109613004A (en) | Defect display methods in a kind of inspection of backlight |
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 | ||
CB02 | Change of applicant information |
Address after: Room 5124, building C, 555 Dongchuan Road, Minhang District, Shanghai 200241 Applicant after: SHANGHAI LYNUC NUMERICAL CONTROL TECHNOLOGY CO.,LTD. Address before: Room 5124, building C, 555 Dongchuan Road, Minhang District, Shanghai 200241 Applicant before: SHANGHAI LYNUC CNC TECHNOLOGY Co.,Ltd. |
|
CB02 | Change of applicant information |