CN109858536A - A method of the offline automatic detection long filament silk end of reel bar silk - Google Patents
A method of the offline automatic detection long filament silk end of reel bar silk Download PDFInfo
- Publication number
- CN109858536A CN109858536A CN201910056393.0A CN201910056393A CN109858536A CN 109858536 A CN109858536 A CN 109858536A CN 201910056393 A CN201910056393 A CN 201910056393A CN 109858536 A CN109858536 A CN 109858536A
- Authority
- CN
- China
- Prior art keywords
- silk
- region
- volume
- training
- long filament
- 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
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000001514 detection method Methods 0.000 title abstract description 18
- 238000012549 training Methods 0.000 claims abstract description 56
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 30
- 238000012512 characterization method Methods 0.000 claims abstract 2
- 238000007689 inspection Methods 0.000 claims description 19
- 238000012360 testing method Methods 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 230000003068 static effect Effects 0.000 claims description 3
- 239000004744 fabric Substances 0.000 claims description 2
- 230000001737 promoting effect Effects 0.000 abstract 1
- 230000004913 activation Effects 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000001537 neural effect Effects 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000003475 lamination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 230000009291 secondary effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of methods of the offline automatic detection long filament silk end of reel bar silk, the surface that silk is rolled up is divided into multiple regions, after the appearance images for acquiring each region, it is separately input in the corresponding disaggregated model in each region, by disaggregated model output characterization, whether there is or not the labels of tail silk fault, multiple regions are n region, i=1, 2, ..., n, i corresponding disaggregated model in region is the LeNet convolutional neural networks after database training, database includes multiple appearance images and its corresponding label, multiple appearance images are the appearance images of multiple volume surface region i, image is seen in addition respectively when training and label is input item and target output item, multiple volumes are the set for the silk volume that surface region i has tail silk fault without the silk volume and surface region i of tail silk fault.The method of the present invention detection efficiency is high, and accuracy rate is high, there is fabulous promotional value.
Description
Technical field
The invention belongs to chemical fibre detection technique fields, are related to a kind of method of the offline automatic detection long filament silk end of reel bar silk.
Background technique
Tail silk is the top of long filament silk volume and visible single or more disengaging Yu Sijuan main body of independence that tail end occurs
Long filament, be common a kind of fault in long filament silk volume, the appearance and quality of silk volume can be largely affected by, such as rear
Often lead to the interruption of production in procedure because of the presence of tail silk, therefore long filament silk volume will be rolled up before packing
Fill the tail silk detection of appearance.
The detection of tail silk at this stage is detected using artificial eye, is needed artificial to each volume package
It is checked to determine whether silk volume has tail silk, can not only expend a large amount of artificial and time in this way, influence production efficiency,
And the fatigue of human eye and the subjectivity of people can also influence the accuracy rate of tail silk detection to a certain degree.
Therefore, the method for studying a kind of high offline automatic detection long filament silk end of reel bar silk of detection efficiency has particularly significant
Meaning.
Summary of the invention
The purpose of the invention is to overcome above-mentioned problems of the prior art, provide a kind of detection efficiency it is high from
The method that line detects the long filament silk end of reel bar silk automatically.
A kind of method of the offline inspection long filament silk end of reel bar silk, by silk volume surface (be easy to appear the surface of tail silk,
Because tail silk is that the local surfaces rolled up in silk occur, do not have to the overall surface that silk is rolled up being divided into multiple regions, only
Need the local surfaces that tail silk will likely occur to be divided into multiple regions, be conducive in this way improve detection efficiency) be divided into it is multiple
Region after the appearance images for acquiring each region, is separately input in the corresponding disaggregated model in each region, defeated by disaggregated model
Characterize that whether there is or not the labels of tail silk fault out；
Multiple regions are n region, and i=1,2 ..., n, i corresponding disaggregated model in region is after database training
LeNet convolutional neural networks, database include multiple appearance images and its corresponding label, multiple appearance images are multiple volumes
The appearance images of surface region i, see image respectively when training in addition and label is input item and target output item to LeNet convolution
Neural network is trained；LeNet convolutional neural networks are the convolutional neural networks forms of earliest period, and structure is simple, training
At low cost, the used time is few, and tail silk is typically all clearly independently of the fault except silk volume, relatively easy to judge, utilization
LeNet convolutional neural networks, which are trained, rapidly to be identified, time used time is greatly shortened；
The multiple silk volume is the silk that surface region i has tail silk fault without the silk volume and surface region i of tail silk fault
The set of volume.
Image detecting method of the invention can be very good as a kind of touchless detection means to long filament silk volume
Appearance is analyzed and processed without causing secondary effect to silk volume appearance because of contact long filament silk volume, and utilizes database
LeNet convolutional neural networks after image training can reach as disaggregated model and normally be wrapped and contain tail to long filament silk volume
The quick, intelligent identification of the package of bar silk, in addition, appearance images obtained in detection process can also be continually added to instruction
Practice in sample constantly to improve corresponding disaggregated model.
As a preferred technical scheme:
A kind of method of the offline inspection long filament silk end of reel bar silk as described above, uses when establishing different disaggregated models
Multiple volumes are identical or different.
The sum of a kind of method of the offline inspection long filament silk end of reel bar silk as described above, the multiple silk volume is greater than
2000。
A kind of method of the offline inspection long filament silk end of reel bar silk as described above, in the multiple silk volume surface region i without
It is 1:1 that silk volume and the surface region i of tail silk fault, which have the quantity ratio of the silk volume of tail silk fault,.The present invention will be in multiple volumes
Surface region i without tail silk fault silk volume with surface region i have tail silk fault silk volume quantity ratio be set as 1:1 be for
The uneven of elimination positive negative sample influences on brought by the training of subsequent neural network and evaluation, can be good at accurate
Rate index evaluates constructed neural network.
A kind of method of the offline inspection long filament silk end of reel bar silk as described above, the label are 0 and 1,0 representative without tail
Silk fault, 1 representative have tail silk fault.
A kind of method of the offline inspection long filament silk end of reel bar silk as described above, the foundation of the corresponding disaggregated model of region i
Steps are as follows:
(1) after acquiring the appearance images of multiple volume surface region i and being translated into gray level image, 90% is randomly selected
Appearance images as training sample, remaining appearance images are as test sample；
(2) the corresponding label of each sample is determined one by one；
(3) respectively using appearance images and its corresponding label as input item and target output item, using training sample training
LeNet convolutional neural networks obtain the LeNet convolutional neural networks after training sample training, and LeNet convolutional neural networks are by rolling up
Lamination, pond layer and full articulamentum composition, belong to a kind of structure of determination, the training method of LeNet convolutional neural networks and often
Advise convolutional neural networks it is almost the same, training termination condition it is the same with conventional convolution neural network, be all error be less than etc.
In desired value, desired value can be set according to actual needs；The appearance images of training sample and corresponding label just start to instruct
LeNet convolutional neural networks are all inputted when practicing, for being trained to an ideal network, and test sample only needs to input
Appearance images, LeNet convolutional neural networks after appearance images input training sample training, can export corresponding code,
Namely reality output label；
(4) after whole training sample training, by the LeNet convolutional Neural after test sample input training sample training
Its corresponding label simultaneously is compared to obtain by network with the label of the LeNet convolutional neural networks output after training sample training
Classification accuracy；It is 0 without tail silk fault that each test sample, which has corresponding label, and having tail silk fault is 1, test
When can by the appearance images of test sample input enter, then export a label, by the label of output with script it is corresponding
Label compares, if unanimously illustrating that the judgement of neural network is that correctly, each test sample can carry out judging once, finally
It can obtain a classification accuracy；
(5) judge whether classification accuracy is greater than 97%, if it is, obtaining disaggregated model；Conversely, then entering next
Step；
(6) return step (4) after the LeNet convolutional neural networks parameter after adjusting training sample training, such as adjustment batch
Size and various activation primitive types etc., the ginsengs such as the size and step-length, the type of pond layer and step-length of also adjustable convolution kernel
Number, or increase return step (1) after the quantity of the multiple silk volume.
A kind of method of the offline inspection long filament silk end of reel bar silk as described above, n=4, the appearance images in each region
It is acquired by 4 cameras, 1 camera lens rolls up top surface towards silk, and 1 camera lens rolls up bottom surface, 2 camera lens towards silk
Side, and the same side that cloth is rolled up setting in silk up and down are rolled up towards silk.
A kind of method of the offline inspection long filament silk end of reel bar silk as described above, the region that 4 cameras surround are figure
As pickup area, the silk volume is delivered to static after image acquisition region by plate conveyor.
The utility model has the advantages that
(1) method of a kind of offline inspection long filament silk end of reel bar silk of the invention, detection efficiency is high, and accuracy rate is high, solves
The prior art manual detection efficiency low problem low with accuracy rate；
(2) method of a kind of offline inspection long filament silk end of reel bar silk of the invention can offline whether there is long filament package
Tail silk fault is detected, easy to operate.
Detailed description of the invention
Fig. 1 is that long filament silk of the invention rolls up Image Acquisition schematic diagram；
1- camera a, 2- camera b, 3- camera c, 4- camera d, 5- long filament silk volume, 6- plate conveyor, 7- taper ingot.
Specific implementation method
Below with reference to specific implementation method, the present invention is further explained.It should be understood that these embodiments are merely to illustrate this hair
It is bright rather than limit the scope of the invention.In addition, it should also be understood that, after reading the content taught by the present invention, art technology
Personnel can make various changes or modifications the present invention, and such equivalent forms equally fall within the application the appended claims and limited
Fixed range.
The method of the offline inspection long filament silk end of reel bar silk of the invention, steps are as follows:
The first step, as shown in Figure 1,6 conveying filament silk of plate conveyor volume 5 is to image acquisition region, image acquisition region is phase
The region that machine a 1, camera b 2, camera c 3 and camera d 4 are surrounded, wherein the camera lens of camera a 1 and camera d 4 respectively for
Silk volume top surface and silk roll up bottom surface, camera b 2 and 3 camera lens of camera c and roll up side upper end and side lower end, and 2 He of camera b towards silk
Camera c 3 is located at the same side of silk volume, and taper ingot 7 is equipped in plate conveyor 6, and 7 tip radius of taper ingot is less than on long filament silk volume 5
The both ends of the radius of bobbin for winding filaments, plate conveyor 6 and 7 contact area of taper ingot are hollow processing, it is ensured that fortune
Input board 6 and taper ingot 7 will not interfere acquisition of the camera d 4 to long filament silk 5 bottom end images of volume；
Second step acquires image, the motion mode of plate conveyor 6 with camera a 1, camera b 2, camera c 3 and camera d 4
To be intermittent, when long filament silk volume 5 be in image acquisition region when, plate conveyor 6 is static, to four groups of cameras simultaneously take pictures after after
Reforwarding send next group of long filament silk volume 5 to image acquisition region to carry out Image Acquisition, to obtain the outside drawing of multiple long filament silk volumes
Picture, the sum of multiple volumes are greater than 2000, nothing in camera a 1, camera b 2, camera c 3 and the corresponding pickup area of camera d 4
The silk volume and the quantity ratio for the silk volume for having tail silk fault of tail silk fault are 1:1；
Third step establishes disaggregated model a for camera a acquired image；
(1) after converting gray level image for the appearance images of the collected multiple volumes of camera a, randomly select 90% it is outer
Image is seen as training sample, remaining appearance images are as test sample；
(2) the corresponding label of each sample is determined one by one, and wherein label is that 0 and 1,0 representative is represented without tail silk fault, 1
There is tail silk fault；
(3) respectively using appearance images and its corresponding label as input item and target output item, using training sample training
LeNet convolutional neural networks obtain the LeNet convolutional neural networks after training sample training, and LeNet convolutional neural networks include
13 convolutional layers and 3 full articulamentums, are finally softmax layers, for exporting the label of image, LeNet convolutional neural networks
Parameter include activation primitive " sigmoid " activation primitive, further include the size and step-length, the type and step of pond layer of convolution kernel
It is long etc., in training process, using the appearance images of training sample and its corresponding label as input item and target output item
LeNet convolutional neural networks are input to, are reused after convolutional layer, pond layer and full articulamentum as softmax layers respectively
Classifying and then adjusting Network In Network convolutional neural networks is the LeNet convolutional Neural constructed after training sample training
Network；
(4) after by training sample all training, after being trained using test sample to constructed training sample
The classification accuracy of LeNet convolutional neural networks is tested, i.e., the appearance images input of test sample is entered, then trained
LeNet convolutional neural networks after sample training export a label, by the label of output label corresponding with its script compared with,
If identical, the judgement of the LeNet convolutional neural networks after illustrating training sample training is correctly on the contrary then mistake, often
A test sample all can once be judged, to obtain classification accuracy；
(5) judge whether classification accuracy is greater than 97%, if it is, obtaining disaggregated model a；Conversely, then entering next
Step；
(6) parameter of the LeNet convolutional neural networks after adjusting training sample training is as replaced " sigmoid " activation primitive
Be changed to " ReLU " activation primitive, adjust convolution kernel size and step-length, the type and step-length of pond layer after return step (4), or
Person utilizes return step (1) after the more silks volume appearance images of camera a acquisition；
4th step is established for the image of camera b, camera c and camera d acquisition using method identical with third step respectively
Disaggregated model b, disaggregated model c and disaggregated model d are used multiple when establishing disaggregated model b, disaggregated model c and disaggregated model d
Silk volume can also be not identical as disaggregated model a, but higher using identical silk volume Detection accuracy；
5th step carries out image using camera a 1, camera b 2, camera c 3 and the long filament silk of 4 pairs of camera d production volume and adopts
Input is separately input into disaggregated model a, disaggregated model b, disaggregated model c and disaggregated model after collecting and being translated into gray level image
In d, if all output 0 of 4 disaggregated models, illustrate that the package is normal package, there are tails conversely, then illustrating long filament silk volume
Bar silk fault.
In addition, during carrying out actual classification using the disaggregated model of building, the length that can constantly will test
Silk silk volume image adds in total sample of training and test to constantly improve disaggregated model.
Claims (8)
1. a kind of method of the offline inspection long filament silk end of reel bar silk acquires it is characterized in that: the surface that silk is rolled up is divided into multiple regions
It after the appearance images in each region, is separately input in the corresponding disaggregated model in each region, is had by disaggregated model output characterization
Label without tail silk fault；
Multiple regions are n region, and i=1,2 ..., n, i corresponding disaggregated model in region is LeNet volume after database training
Product neural network, database include multiple appearance images and its corresponding label, multiple appearance images are multiple volume surface districts
The appearance images of domain i see image respectively when training in addition and label are input item and target output item；
The multiple silk volume is the silk volume that surface region i has tail silk fault without the silk volume and surface region i of tail silk fault
Set.
2. a kind of method of the offline inspection long filament silk end of reel bar silk according to claim 1, which is characterized in that establish different
Disaggregated model when multiple volumes using it is identical or different.
3. a kind of method of the offline inspection long filament silk end of reel bar silk according to claim 1, which is characterized in that the multiple
The sum of silk volume is greater than 2000.
4. a kind of method of the offline inspection long filament silk end of reel bar silk according to claim 1, which is characterized in that the multiple
It is 1:1 that silk volume of the surface region i without tail silk fault, which has the quantity ratio of the silk volume of tail silk fault with surface region i, in silk volume.
5. a kind of method of the offline inspection long filament silk end of reel bar silk according to claim 1, which is characterized in that the label
It represents for 0 and 1,0 without tail silk fault, 1 representative has tail silk fault.
6. a kind of method of the offline inspection long filament silk end of reel bar silk according to claim 5, which is characterized in that i pairs of region
The establishment step for the disaggregated model answered is as follows:
(1) appearance images of multiple volume surface region i are acquired and after being translated into gray level image, randomly select 90% it is outer
Image is seen as training sample, remaining appearance images are as test sample；
(2) the corresponding label of each sample is determined one by one；
(3) respectively using appearance images and its corresponding label as input item and target output item, using training sample training LeNet
Convolutional neural networks obtain the LeNet convolutional neural networks after training sample training；
(4) by the LeNet convolutional neural networks after test sample input training sample training and by its corresponding label and training
The label of LeNet convolutional neural networks output after sample training is compared to obtain classification accuracy；
(5) judge whether classification accuracy is greater than 97%, if it is, obtaining disaggregated model；Conversely, then entering in next step；
(6) return step (4) after the LeNet convolutional neural networks parameter after adjusting training sample training, or increase described more
Return step (1) after the quantity of a volume.
7. a kind of method of the offline inspection long filament silk end of reel bar silk according to claim 1, which is characterized in that n=4, institute
The appearance images for stating each region are acquired by 4 cameras, and 1 camera lens rolls up top surface towards silk, and 1 camera lens is towards silk
Bottom surface is rolled up, 2 camera lens roll up side, and the same side that cloth is rolled up setting in silk up and down towards silk.
8. a kind of method of the offline inspection long filament silk end of reel bar silk according to claim 7, which is characterized in that described 4
The region that camera surrounds is image acquisition region, and silk volume is delivered to static after image acquisition region by plate conveyor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910056393.0A CN109858536A (en) | 2019-01-22 | 2019-01-22 | A method of the offline automatic detection long filament silk end of reel bar silk |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910056393.0A CN109858536A (en) | 2019-01-22 | 2019-01-22 | A method of the offline automatic detection long filament silk end of reel bar silk |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109858536A true CN109858536A (en) | 2019-06-07 |
Family
ID=66895515
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910056393.0A Pending CN109858536A (en) | 2019-01-22 | 2019-01-22 | A method of the offline automatic detection long filament silk end of reel bar silk |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109858536A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113899696A (en) * | 2021-08-30 | 2022-01-07 | 中国纺织科学研究院有限公司 | On-line detection device for broken filaments on surface of filament package |
| CN113971650A (en) * | 2020-07-22 | 2022-01-25 | 富泰华工业（深圳）有限公司 | Product flaw detection method, computer device and storage medium |
| CN114092382A (en) * | 2020-08-07 | 2022-02-25 | 富泰华工业（深圳）有限公司 | Product flaw marking device and method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107123114A (en) * | 2017-04-21 | 2017-09-01 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of cloth defect inspection method and device based on machine learning |
| CN107169956A (en) * | 2017-04-28 | 2017-09-15 | 西安工程大学 | Yarn dyed fabric defect detection method based on convolutional neural networks |
| US20180211373A1 (en) * | 2017-01-20 | 2018-07-26 | Aquifi, Inc. | Systems and methods for defect detection |
| CN109145985A (en) * | 2018-08-21 | 2019-01-04 | 佛山职业技术学院 | A kind of detection and classification method of Fabric Defects Inspection |
-
2019
- 2019-01-22 CN CN201910056393.0A patent/CN109858536A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180211373A1 (en) * | 2017-01-20 | 2018-07-26 | Aquifi, Inc. | Systems and methods for defect detection |
| CN107123114A (en) * | 2017-04-21 | 2017-09-01 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of cloth defect inspection method and device based on machine learning |
| CN107169956A (en) * | 2017-04-28 | 2017-09-15 | 西安工程大学 | Yarn dyed fabric defect detection method based on convolutional neural networks |
| CN109145985A (en) * | 2018-08-21 | 2019-01-04 | 佛山职业技术学院 | A kind of detection and classification method of Fabric Defects Inspection |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113971650A (en) * | 2020-07-22 | 2022-01-25 | 富泰华工业（深圳）有限公司 | Product flaw detection method, computer device and storage medium |
| US12260537B2 (en) | 2020-07-22 | 2025-03-25 | Hon Hai Precision Industry Co., Ltd. | Defect detection method, computer device and storage medium |
| CN114092382A (en) * | 2020-08-07 | 2022-02-25 | 富泰华工业（深圳）有限公司 | Product flaw marking device and method |
| CN113899696A (en) * | 2021-08-30 | 2022-01-07 | 中国纺织科学研究院有限公司 | On-line detection device for broken filaments on surface of filament package |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111274860B (en) | Recognition method for online automatic tobacco grade sorting based on machine vision | |
| US12223633B2 (en) | Fabric defect detection method based on multi-modal deep learning | |
| CN103971342B (en) | A kind of image noise detection method based on convolutional neural networks | |
| CN109858536A (en) | A method of the offline automatic detection long filament silk end of reel bar silk | |
| CN117348831B (en) | Picture adjustment method and system for liquid crystal display screen | |
| CN107067043A (en) | A kind of diseases and pests of agronomic crop detection method | |
| CN113077450B (en) | Cherry grading detection method and system based on deep convolutional neural network | |
| CN111462058B (en) | A rapid detection method for effective rice panicles | |
| CN108428231A (en) | A kind of multi-parameter Part Surface Roughness learning method based on random forest | |
| CN115908351B (en) | Yarn quality prediction system and method | |
| CN109035248A (en) | Defect detection method, apparatus, terminal device, server and storage medium | |
| CN115546141A (en) | A small sample Mini LED defect detection method and system based on multi-dimensional measurement | |
| CN116883333A (en) | Textile surface defect detection method based on deep learning | |
| CN110813795A (en) | Device and method for detecting breakage of chemical fiber spindle paper tube on line | |
| CN112132803A (en) | A method for egg freshness detection based on convolutional neural network | |
| CN117132174B (en) | Model training method and system applied to quality detection of industrial assembly line | |
| CN112150416A (en) | Coating defect detection system based on deep learning | |
| CN116630735A (en) | Silkworm disease detection and prevention and control service system for industrial sericulture production line based on artificial intelligence | |
| CN115170881A (en) | Liquor picking method and system based on deep learning | |
| CN106501278B (en) | Surface of the light tube defect classification method and system based on invariable rotary textural characteristics | |
| CN109903267A (en) | A method of based on image processing techniques test network wire network degree | |
| CN114092441A (en) | A product surface defect detection method and system based on dual neural network | |
| CN115100204B (en) | Cloth defect detection and discrimination method based on Fourier transform | |
| CN117495795A (en) | A textile defect detection method and system | |
| CN112775032B (en) | Fruit sorting method and device and flexible robot |
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: 20190607 |