CN109706616A - A kind of cloth cover defective vision detection system of double rib warp loom - Google Patents
A kind of cloth cover defective vision detection system of double rib warp loom Download PDFInfo
- Publication number
- CN109706616A CN109706616A CN201910001622.9A CN201910001622A CN109706616A CN 109706616 A CN109706616 A CN 109706616A CN 201910001622 A CN201910001622 A CN 201910001622A CN 109706616 A CN109706616 A CN 109706616A
- Authority
- CN
- China
- Prior art keywords
- cloth cover
- double rib
- warp loom
- rib warp
- detection system
- 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
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Looms (AREA)
Abstract
The present invention relates to warp knitting machine fabric testing equipment fields, and in particular to a kind of cloth cover defective vision detection system of double rib warp loom, including double rib warp loom and the cloth cover defects detection mechanism being arranged on double rib warp loom;The double rib warp loom includes the lathe top plane above knocking-over bar and knocking-over bar;Cloth cover defects detection mechanism includes the power supply being electrically connected to each other, control device, photographic device, human-computer interaction interface and storage device;The photographic device includes bracket component and CCD camera assembly.The camera of the cloth cover defects detection mechanism of the cloth cover defective vision detection system of double rib warp loom of the present invention is located between the knocking-over bar of double rib warp loom and lathe top plane above knocking-over bar, can it is optimal carry out cloth planar defect Image Acquisition, it is easy for installation flexible, the production of tricot machine is not influenced and guarantees image quality.False detection rate substantially reduces compared with prior art, mitigates the working strength of worker, greatly improves product quality.
Description
Technical field
The present invention relates to warp knitting machine fabric testing equipment fields, and in particular to a kind of cloth planar defect view of double rib warp loom
Feel detection system.
Background technique
Tricot machine is classified according to needle bed, can be divided into single needle bar warp knitting machine and double rib warp loom.And in recent years, with people
Improvement of living standard, people it is higher and higher to the designs and varieties and quality requirement of warp knit product.Due to the complication of demand
And diversification.The fabric of single needle bar warp knitting machine has been difficult meet the needs of people, the fabric of double rib warp loom flower abundant
Color effect can knit out double rib, cylindraceous fabric and various fuzz fabrics etc..Two-needle bar application at present is more and more general
Time.
But in the actual production process, due to broken string, the problems such as yarn greasy dirt, frequently can lead to if lag is found the problem
Made finished product cloth is scrapped.Therefore we need to find these defects in cloth production process.Reach defect
It is found in first time.To reduce economic loss.But, low efficiency at high cost by manual identified, and enter 2018, artificial intelligence
Energy technology is maked rapid progress, and increasingly mature.Artificial intelligence technology can be described as at a tremendous pace in the development of all trades and professions.Weaving
Industry belongs to the typical labour intensive profession in China, and in the case where current lack of labor, artificial intelligence technology is to weaving
The development of industry specifies direction.How to combine artificial intelligence technology to carry out cloth cover defects detection is that textile industry letter is to be solved
Technical problem.
Summary of the invention
The technical problems to be solved by the present invention are: providing a kind of intelligent, efficiency and the high two-needle bar warp of image quality
The cloth cover defective vision detection system of volume machine.
In order to solve the above-mentioned technical problem, the technical solution adopted by the present invention are as follows: a kind of cloth of double rib warp loom is provided
Planar defect vision detection system, including double rib warp loom and the cloth cover defects detection mechanism being arranged on double rib warp loom;
The double rib warp loom includes the lathe top plane above knocking-over bar and knocking-over bar;
Cloth cover defects detection mechanism includes the power supply, control device, photographic device, human-computer interaction circle being electrically connected to each other
Face and storage device;
The photographic device includes bracket component and CCD camera assembly;The bracket component is fixed under the plane of lathe top
Side, the bracket component include guide rail and driving device, and the driving device is for driving CCD camera assembly back and forth to slide along guide rail
Dynamic, the CCD camera assembly is towards knocking-over bar.
The beneficial effects of the present invention are: the cloth cover defective vision detection system of double rib warp loom of the invention utilizes people
Work intellectual technology carries out object crawl to cloth cover with mobile camera, and using moving camera shooting technology, multi-angled shooting improves figure
As shooting quality and camera quantity is reduced, is constantly performed physical exercise to the database of the feature extraction of system, then system is according to number
The image grabbed is determined according to library, to determine cloth with the presence or absence of defect.Cloth cover defects detection mechanism of the invention
Camera is located between the knocking-over bar of double rib warp loom and lathe top plane above knocking-over bar, can optimal carry out cloth cover
The Image Acquisition of defect, it is easy for installation flexible, the production of tricot machine is not influenced and guarantees image quality.Erroneous detection compared with prior art
Rate substantially reduces, and mitigates the working strength of worker, greatly improves product quality.
Detailed description of the invention
Fig. 1 is the structural representation of the cloth cover defective vision detection system of the double rib warp loom of the specific embodiment of the invention
Figure;
Fig. 2 is the knot of the part of the cloth cover defective vision detection system of the double rib warp loom of the specific embodiment of the invention
Structure schematic diagram;
Label declaration:
1, knocking-over bar;2, lathe top plane;3, guide rail;4, industrial camera;5, motor;6, lead screw;7, slide unit.
Specific embodiment
To explain the technical content, the achieved purpose and the effect of the present invention in detail, below in conjunction with embodiment and cooperate attached
Figure is explained.
The most critical design of the present invention is: the camera of cloth cover defects detection mechanism of the invention is located at two-needle bar warp knit
Between the knocking-over bar of machine and lathe top plane above knocking-over bar, can the optimal Image Acquisition for carrying out cloth planar defect, installation
It is convenient, flexible, the production of tricot machine is not influenced and guarantees image quality.
Please refer to Fig. 1 and Fig. 2, a kind of cloth cover defective vision detection system of double rib warp loom, including two-needle bar warp
Volume machine and the cloth cover defects detection mechanism being arranged on double rib warp loom;The double rib warp loom includes knocking-over bar 1 and takes off
The lathe top plane 2 of 1 top of girth sheets;
Cloth cover defects detection mechanism includes the power supply, control device, photographic device, human-computer interaction circle being electrically connected to each other
Face and storage device;
The photographic device includes bracket component and CCD camera assembly;The bracket component is fixed under lathe top plane 2
Side, the bracket component include guide rail 3 and driving device, and the driving device is for driving CCD camera assembly reciprocal along guide rail 3
Sliding, the CCD camera assembly is towards knocking-over bar 1.
The course of work of the cloth cover defective vision detection system of above-mentioned double rib warp loom: it is set by human-computer interaction interface
The movement speed of good CCD camera assembly, for fabric during mobile on tricot machine, driving device drives CCD camera assembly edge to lead
Rail is mobile, and mobile CCD camera assembly simultaneously carries out Image Acquisition;The image transmitting taken is filled to main control unit according to storage
Database in setting carries out image analysis comparison and processing, and then identifies to cloth planar defect and type.
The cloth cover defective vision detection system of above-mentioned double rib warp loom the utility model has the advantages that using artificial intelligence technology, fortune
Object crawl is carried out to cloth cover with mobile camera, multi-angled shooting is carried out using moving camera shooting technology and improves image taking matter
Amount, and the camera quantity used is reduced, the database constantly extracted to the cloth cover defect characteristic of system performs physical exercise, then system
The image grabbed is determined according to database, to determine cloth with the presence or absence of defect.Cloth cover defects detection of the invention
The camera of mechanism is located between the knocking-over bar of double rib warp loom and lathe top plane above knocking-over bar, can optimal position
The Image Acquisition for carrying out cloth planar defect is set, it is easy for installation flexible, the production of tricot machine is not influenced and guarantees image quality.Compared to existing
There is technology false detection rate to substantially reduce, mitigates the working strength of worker, greatly improve product quality.
Further, in the cloth cover defective vision detection system of above-mentioned double rib warp loom, the cloth cover defects detection
Mechanism further includes warning light, and the warning light is to issue audible and visible alarm when control device finds cloth planar defect.By upper description
It is found that control device can issue alarm signal according to obtained recognition result control warning light, or tricot machine is shut down.
Further, in the cloth cover defective vision detection system of above-mentioned double rib warp loom, the CCD camera assembly packet
Include industrial camera 4, sensor pixel 2,000,000, the resolution ratio 1080P of the industrial camera 4, frame speed 120.
Further, in the cloth cover defective vision detection system of above-mentioned double rib warp loom, the CCD camera assembly is also
Including LED light source, the LED light source provides illumination for cloth cover.By upper description it is found that the setting of above-mentioned LED light source makes camera
The picture quality that component obtains is higher.
Further, in the cloth cover defective vision detection system of above-mentioned double rib warp loom, the control device includes
FPGA main control chip, neuroid unit built in the FPGA main control chip;
Further, in the cloth cover defective vision detection system of above-mentioned double rib warp loom, the driving device includes
Motor 5, the lead screw 6 being rotatablely connected with motor 5, the slide unit 7 connecting with wire rod thread, the screw rod drive slide unit 7 along 3 line of guide rail
Property sliding, the CCD camera assembly and slide unit 7 be connected, and the motor 5 is electrically connected with control device.
Further, in the cloth cover defective vision detection system of above-mentioned double rib warp loom, the cloth cover defects detection
Mechanism further includes the printer being electrically connected with control device, and the printer is for printing the image for having defective cloth cover.
By upper description it is found that the setting of printer can be convenient operator studies the image of the defective cloth cover of tool
Compare or stay shelves.
Further, in the cloth cover defective vision detection system of above-mentioned double rib warp loom, the human-computer interaction interface
For touch display screen.
By it is upper description it is found that touch display screen setting can real-time display defect cloth cover situation, when facilitating operator the first
Between find and reject defect cloth cover;
Further, in the cloth cover defective vision detection system of above-mentioned double rib warp loom, the storage device includes
SD card.
By upper description it is found that the setting of SD card is saved to carry out direct copying and transfer to cloth cover defect database
The process of repetition training improves production efficiency.
Embodiment one
Existing double rib warp loom lacks effective measures in process of production and detects to cloth planar defect, so that
Many defects are just found in the finished product stage, are had a significant impact to the examination and test of products and maintenance cost.The present invention utilizes artificial intelligence
Technology carries out object crawl to cloth with high speed camera, constantly performs physical exercise to system.Then system is according to the figure grabbed
Piece is determined, to determine cloth with the presence or absence of defect.The present invention does not specify tricot machine producer and model.
A kind of cloth cover defective vision detection system the present embodiment provides double rib warp loom include double rib warp loom and
Cloth cover defects detection mechanism on double rib warp loom is set;The double rib warp loom includes on knocking-over bar 1 and knocking-over bar 1
The lathe top plane 2 of side;Cloth cover defects detection mechanism includes the power supply being electrically connected to each other, control device, photographic device, people
Machine interactive interface and storage device;The photographic device includes bracket component and CCD camera assembly;The bracket component is fixed on
2 lower section of lathe top plane, the bracket component include guide rail 3 and driving device, and the driving device is for driving camera group
Part reciprocatingly slides along guide rail 3, and the CCD camera assembly is towards knocking-over bar 1.
The present embodiment provides a kind of power supplys of the cloth cover defective vision detection system of double rib warp loom using high performance
Switching Power Supply is mainly powered to system;
Control device passes through people using the main control chip based on FPGA technology, built-in neuroid unit (NPU)
The study of work intelligent depth, makes the operation of system more efficient.CCD camera assembly is using sensor pixel 2,000,000, resolution ratio
1080P, the High performance industrial camera mould group of frame speed 120.Human-computer interaction interface is for real-time display defect cloth cover situation, conveniently
Operator has found at the first time and rejects defect cloth cover;SD card is saved to carry out direct copying to cloth cover defect database
The process of repetition training.Improve production efficiency.The quick operation of high speed high definition camera mould group is driven by chain.
Vision detection system of the invention is installed on double rib warp loom, so in image acquisition process, by
In unavoidably will cause using the detection of high speed camera movable type, there is jitter phenomenon in shooting video sequence and image is fuzzy existing
As, and in defects detection, need image that there is preferable stability, so stabilization algorithm is needed, in actual photographed
Kinematic parameter compensates, and promotes the quality of shooting image, achievees the purpose that stabilization.
Currently, common video stabilization algorithm includes optical flow method and characteristic method both algorithms.The main think of of optical flow method
Road and step are as follows: the light stream vector in video between consecutive frame is calculated first, then according to previous step calculated result and fortune
Dynamic analysis is counted and is calculated to the current global relevant parameter for becoming pupil.Finally, by this parameter to video sequence
Initial parameter carries out revision compensation.The advantages of optical flow method is the motion vector that each pixel can be obtained by calculation;The disadvantage is that
Calculation amount is very big, it is difficult to guarantee the real-time of subsequent foreign bodies detection.
The thinking and step of characteristic method are as follows: first correlated characteristic point is acquired in current frame image, and by these characteristic points
Matching operation is carried out with the characteristic point of next frame, according to matched as a result, the global parameter to capture apparatus calculates, then
Filtering, finally revises compensation according to original image parameter value of the filtered motion parameter to shooting.It, should compared with optical flow method
The advantages of algorithm is that calculation amount is smaller, the disadvantage is that this method treatment effect is bad when the edge line for shooting image is fuzzy.
Therefore the present invention carries out collection apparatus using characteristic method.Cloth cover defect detecting system in the present invention is movement
Formula system, by that then according to relative motion principle, using cloth cover as background, and will lack in real time to cloth cover image information collecting
Fall into the target for being considered as movement.Therefore for Systems for optical inspection, the research core of cloth cover defects detection algorithm is movement mesh
Mark detection algorithm.In this algorithm, the machine-learning process of the feature of cloth planar defect be it is crucial, system is needed to cloth cover and can
The defect that can occur carries out a degree of training.The debugging of trained length and system and use time are proportional.It calculates
The method that method has used " feature representation " to foreign matter, it is the progress vector quantization, sparse coding in most basic feature base
Or other expression ways to be to form the last feature of piece image, such as it is damaged, lose line, greasy dirt etc..Respectively to RGB image and depth
Degree feature extraction denseSIFT feature simultaneously carries out local restriction uniform enconding to form the character representation of Kinect image pair, so
Afterwards validity feature is learnt to similar foreign matter and the spot for being easy to report by mistake based on convolutional neural networks algorithm automatically and identified,
And database is used to carry out training pattern to further increase discrimination, effectively cloth planar defect target detection can be come out.
The above description is only an embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalents made by bright specification and accompanying drawing content are applied directly or indirectly in relevant technical field, similarly include
In scope of patent protection of the invention.
Claims (9)
1. a kind of cloth cover defective vision detection system of double rib warp loom, which is characterized in that including double rib warp loom and set
Set the cloth cover defects detection mechanism on double rib warp loom;The double rib warp loom includes above knocking-over bar and knocking-over bar
Lathe top plane;
Cloth cover defects detection mechanism include the power supply being electrically connected to each other, control device, photographic device, human-computer interaction interface and
Storage device;
The photographic device includes bracket component and CCD camera assembly;The bracket component is fixed below the plane of lathe top, institute
Stating bracket component includes guide rail and driving device, and the driving device is for driving CCD camera assembly to reciprocatingly slide along guide rail, institute
CCD camera assembly is stated towards knocking-over bar.
2. the cloth cover defective vision detection system of double rib warp loom according to claim 1, which is characterized in that the cloth
Planar defect testing agency further includes warning light, and the warning light is alert to issue acousto-optic when control device finds cloth planar defect
Report.
3. the cloth cover defective vision detection system of double rib warp loom according to claim 1, which is characterized in that described to take the photograph
As head assembly includes industrial camera, sensor pixel 2,000,000, the resolution ratio 1080P of the industrial camera, frame speed 120.
4. the cloth cover defective vision detection system of double rib warp loom according to claim 3, which is characterized in that described to take the photograph
As head assembly further includes LED light source, the LED light source provides illumination for cloth cover.
5. the cloth cover defective vision detection system of double rib warp loom according to claim 1, which is characterized in that the control
Device processed includes FPGA main control chip, neuroid unit built in the FPGA main control chip.
6. the cloth cover defective vision detection system of double rib warp loom according to claim 1, which is characterized in that the drive
Dynamic device includes motor, and with the lead screw of motor rotation connection and the slide unit connecting with wire rod thread, the screw rod drives slide unit edge
Guide rail linear sliding, the CCD camera assembly and slide unit are connected, and the motor is electrically connected with control device.
7. the cloth cover defective vision detection system of double rib warp loom according to claim 1, which is characterized in that the cloth
Planar defect testing agency further includes the printer being electrically connected with control device, and the printer has defective cloth cover for printing
Image.
8. the cloth cover defective vision detection system of double rib warp loom according to claim 1, which is characterized in that the people
Machine interactive interface is touch display screen.
9. the cloth cover defective vision detection system of double rib warp loom according to claim 1, which is characterized in that described to deposit
Storage device includes SD card.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910001622.9A CN109706616A (en) | 2019-01-02 | 2019-01-02 | A kind of cloth cover defective vision detection system of double rib warp loom |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910001622.9A CN109706616A (en) | 2019-01-02 | 2019-01-02 | A kind of cloth cover defective vision detection system of double rib warp loom |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109706616A true CN109706616A (en) | 2019-05-03 |
Family
ID=66260721
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910001622.9A Pending CN109706616A (en) | 2019-01-02 | 2019-01-02 | A kind of cloth cover defective vision detection system of double rib warp loom |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109706616A (en) |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB1047717A (en) * | 1962-09-05 | 1966-11-09 | Lewis Henry Colton | Improvements in and relating to the monitoring of textile webs |
EP0496086A2 (en) * | 1990-12-27 | 1992-07-29 | Compax Corp. | Method and apparatus for analyzing fabric conditions |
CN202175829U (en) * | 2010-12-27 | 2012-03-28 | 中国船舶重工集团公司第七一五研究所 | On-line real-time detection system for gray fabric flaw based on machine vision |
CN102839496A (en) * | 2012-08-30 | 2012-12-26 | 无锡信捷电气股份有限公司 | Visual broken line detection system for warp knitting machine |
CN102967606A (en) * | 2012-11-02 | 2013-03-13 | 海宁市科威工业电子科技有限公司 | Textile machine fabric defect visual inspection system |
CN103116351A (en) * | 2012-12-28 | 2013-05-22 | 福州科迪电子技术有限公司 | Spinning defective cloth detection camera and detection system thereof |
JP2014167456A (en) * | 2013-01-30 | 2014-09-11 | Toyama Prefecture | Fabric defect inspection method and apparatus |
CN104404711A (en) * | 2014-11-27 | 2015-03-11 | 常州驰网智能检测技术有限公司 | Fabric cover defect detection device |
CN206902375U (en) * | 2017-07-06 | 2018-01-19 | 福建中联博创机器人科技股份有限公司 | A kind of novel detection device for being used to detect double rib warp loom fabric |
CN108318496A (en) * | 2018-05-09 | 2018-07-24 | 海宁市科威工业电子科技有限公司 | One kind being used for the online fabric defects detection device of tricot machine |
CN209307576U (en) * | 2019-01-02 | 2019-08-27 | 福建杰嘉科技有限公司 | A kind of cloth cover defective vision detection system of double rib warp loom |
-
2019
- 2019-01-02 CN CN201910001622.9A patent/CN109706616A/en active Pending
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB1047717A (en) * | 1962-09-05 | 1966-11-09 | Lewis Henry Colton | Improvements in and relating to the monitoring of textile webs |
EP0496086A2 (en) * | 1990-12-27 | 1992-07-29 | Compax Corp. | Method and apparatus for analyzing fabric conditions |
CN202175829U (en) * | 2010-12-27 | 2012-03-28 | 中国船舶重工集团公司第七一五研究所 | On-line real-time detection system for gray fabric flaw based on machine vision |
CN102839496A (en) * | 2012-08-30 | 2012-12-26 | 无锡信捷电气股份有限公司 | Visual broken line detection system for warp knitting machine |
CN102967606A (en) * | 2012-11-02 | 2013-03-13 | 海宁市科威工业电子科技有限公司 | Textile machine fabric defect visual inspection system |
CN103116351A (en) * | 2012-12-28 | 2013-05-22 | 福州科迪电子技术有限公司 | Spinning defective cloth detection camera and detection system thereof |
JP2014167456A (en) * | 2013-01-30 | 2014-09-11 | Toyama Prefecture | Fabric defect inspection method and apparatus |
CN104404711A (en) * | 2014-11-27 | 2015-03-11 | 常州驰网智能检测技术有限公司 | Fabric cover defect detection device |
CN206902375U (en) * | 2017-07-06 | 2018-01-19 | 福建中联博创机器人科技股份有限公司 | A kind of novel detection device for being used to detect double rib warp loom fabric |
CN108318496A (en) * | 2018-05-09 | 2018-07-24 | 海宁市科威工业电子科技有限公司 | One kind being used for the online fabric defects detection device of tricot machine |
CN209307576U (en) * | 2019-01-02 | 2019-08-27 | 福建杰嘉科技有限公司 | A kind of cloth cover defective vision detection system of double rib warp loom |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN202175829U (en) | On-line real-time detection system for gray fabric flaw based on machine vision | |
CN205643191U (en) | Fabric quality on -line measuring equipment based on machine vision | |
CN206321577U (en) | A kind of fabric defect detection means | |
CN102967606A (en) | Textile machine fabric defect visual inspection system | |
CN110184659B (en) | Detection device for chemical fiber spinning process yarn path | |
CN202869985U (en) | Machine-vision-based beam-warping and yarn-breaking online monitoring system | |
CN109385876A (en) | A kind of the intelligent quality management system and its method of woven fabric | |
CN102830120A (en) | Machine vision-based beaming broken yarn on-line monitoring system and implementation method thereof | |
CN207866722U (en) | Mobile Automatic Detection of Fabric Defects system | |
CN209307576U (en) | A kind of cloth cover defective vision detection system of double rib warp loom | |
CN111748872A (en) | Device for detecting and removing foreign fiber objects in textile fibers | |
CN110308147B (en) | Intelligent inspection method for chemical fiber spinning process yarn path | |
CN204422432U (en) | Cotton defect detection system | |
CN109706616A (en) | A kind of cloth cover defective vision detection system of double rib warp loom | |
CN111826747A (en) | Method for removing foreign fiber in opening and picking cotton and intelligent bale plucker with foreign fiber detection function used in method | |
CN111748873A (en) | Device for grabbing and removing foreign fiber objects in textile fibers | |
CN210796704U (en) | Detection apparatus for chemical fiber spinning technology silk road | |
CN210775245U (en) | Intelligent inspection device for chemical fiber spinning process yarn path | |
CN107726995A (en) | A kind of knitting needle automatic detection device based on machine vision | |
CN208023291U (en) | A kind of perching machine detecting device of clothes processing | |
CN110672635A (en) | Cloth defect detection device and real-time detection method | |
CN109825944B (en) | Online fabric weaving defect detection method based on line laser | |
CN205711109U (en) | A kind of CCD electronic imaging foreign fiber piece-rate system | |
CN208860762U (en) | Automatic detection yarn hits reed device and slashing system | |
CN208607161U (en) | A kind of device of novel testing product defect |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190503 |