CN108363961A - Bridge pad disease recognition method based on transfer learning between convolutional neural networks - Google Patents
Bridge pad disease recognition method based on transfer learning between convolutional neural networks Download PDFInfo
- Publication number
- CN108363961A CN108363961A CN201810066766.8A CN201810066766A CN108363961A CN 108363961 A CN108363961 A CN 108363961A CN 201810066766 A CN201810066766 A CN 201810066766A CN 108363961 A CN108363961 A CN 108363961A
- Authority
- CN
- China
- Prior art keywords
- convolutional neural
- neural networks
- bridge pad
- disease
- pad disease
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/29—Graphical models, e.g. Bayesian networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
A kind of bridge pad disease recognition method based on transfer learning between convolutional neural networks of the present invention, includes the following steps:Bridge pad disease photo is obtained, and label information is assigned for every photo;With the method for image procossing, increase the data volume for training convolutional neural networks;All pictures in training set and test set are scaled to the cromogram of predefined size, and carry out image preprocessing;Obtain trained convolutional neural networks model on other data sets;In a manner of knowledge migration, the convolutional neural networks model with automatic identification bridge pad disease function is obtained.A kind of bridge pad disease automatic identifying method based on convolutional neural networks of the present invention, by knowledge migration pattern drill convolutional neural networks in precision, there is apparent advantage in convergence rate and greatly reduce the data volume needed for trained neural network, the bridge pad disease of and data complicated for disease scene not easily collecting has certain realistic meaning.
Description
Technical field
The present invention relates to civil engineering and artificial intelligence crossing domain, specifically a kind of bridge based on convolutional neural networks
Bearing damage automatic identifying method.
Background technology
With the fast development of China's infrastructure construction in recent years, building industry development is rapid, a large amount of road and bridge
Construction finishes, and the thing followed is late detection and maintenance work.Bridge pad is the important structure for connecting bridge upper and lower part structure
Part is to count for much where the throat of a bridge block, once there is disease, does not such as find and handle in time, will influence structure
Stress and traffic safety.The detection work main path or artificial detection of bridge pad at present, this method take,
Arduously and traffic can be influenced.Some build remote mountains in, marine bridge is difficult to be realized by the method for artificial detection, or be difficult
Ensure the safety of bridge machinery personnel.Therefore, there is an urgent need for a kind of automatic testing methods of bridge pad disease.
Convolutional neural networks are one kind of artificial neural network, have become the research hotspot in present image identification field.
The shared network structure of its weights is allowed to more similar and biological neural network, reduces the quantity of weights.Convolutional neural networks
Input of the multidimensional image as neural network can be directly used, feature extraction sum number complicated in tional identification algorithm is avoided
According to reconstruction process and there is higher recognition accuracy.
The research of Matthew D.Zeiler et al. illustrates convolutional neural networks and learns to be characterized in layering, a width
Image always provides network inputs with pixel value array form, and first layer learns to be characterized as marginal information;Network it is higher
The basic pattern that layer detection marginal information rearranges in a specific manner;The higher of network combines basic pattern pair
Answer the component of typical objects;This means some rudimentary spies that the convolutional neural networks of this training on other data sets learn
Sign can also move in the identification mission of bridge pad disease.
Invention content
Goal of the invention:In order to overcome the deficiencies in the prior art, the present invention to provide a kind of based on convolutional neural networks
Bridge pad disease automatic identifying method.
Technical solution:In order to solve the above technical problems, a kind of bridge pad disease based on convolutional neural networks of the present invention
Evil automatic identifying method, includes the following steps:
S1:Bridge pad disease photo is obtained, and label information is assigned for every photo, label information corresponds to photo institute
The bridge pad Damage Types of category;
S2:With the method for image procossing, increase the data volume for training convolutional neural networks;
S3:The bridge pad disease photo of acquisition is divided into training set and test set, training set is for training convolutional nerve
Network, test set are used for test network, and all pictures in training set and test set are scaled to the cromogram of predefined size,
And carry out image preprocessing;
S4:Trained convolutional neural networks model on other data sets is obtained, and according to bridge pad
The requirement of disease recognition adjusts the output layer node number of the convolutional neural networks;
S5:In a manner of knowledge migration, with pretreated bridge pad photo training convolutional neural networks are passed through, ladder is used
Degree declines the weights that full articulamentum is adjusted with error backpropagation algorithm, obtains having automatic identification bridge pad disease function
Convolutional neural networks model.
In step S1, the data set of bearing damage photo should include the disease of multiple and different types.
In step S2, the amount of images for training convolutional neural networks is increased by the method for image procossing, for carrying
The generalization ability of high neural network.
In step S3, the size of picture is uniformly scaled to the picture that pixel size is 224 × 224.
In step S3, the preprocess method of image is:Calculate the sum of pixel value of all images then divided by image number
Measure the pixel value for subtracting the mean value image in every piece image to a mean value image.
In step S4, the advance trained model of acquisition is trained classical on ImageNet data sets
VGG16 convolutional neural networks models.
In step S4, the output layer node number of the convolutional neural networks is adjusted according to the requirement of bridge pad disease recognition
Refer to the classification number for the bridge pad disease that adjustment its network Softmax output layers are classified for N number of node, N expression needs,
Softmax layers of calculation formula are as follows:
Wherein, SiFor i-th of input only of this layer, SkFor k-th of net input of the layer.
In step S5, the method for carrying out knowledge migration to neural network is:By trained model is longitudinally drawn in advance
It is divided into each sub-network, when using bridge pad disease photo training, only training adjusts the weights of last full articulamentum and ties up
The weights for holding other each sub-networks are constant.
In step S5, gradient descent method the specific steps are:The gradient of each weights of counting loss function pair, from any point
Start, the negative direction along the gradient moves a distance, continues along gradient reverse direction operation a distance, in this way in new position
The weights of continuous update network.
In step S5, back-propagation algorithm the specific steps are:Convolutional Neural net is being updated using gradient descent method iteration
When the weights of each layer of network, gradient is propagated forward according to chain type Rule for derivation from last layer of network successively.
Advantageous effect:A kind of bridge pad disease automatic identifying method based on convolutional neural networks of the present invention, has
Following advantageous effect:
There is by knowledge migration pattern drill convolutional neural networks in precision, convergence rate apparent advantage and big
Reduce the data volume needed for trained neural network greatly, the bridge pad disease of and data complicated for disease scene not easily collecting
With certain realistic meaning.
Description of the drawings
Fig. 1 is proposed by the present invention a kind of based on the bridge pad disease for carrying out knowledge migration between different convolutional neural networks
The flow chart of automatic identifying method;
Fig. 2 is the longitudinally divided sub-network structure schematic diagrames of VGG16 that knowledge migration is carried out in the present invention.
Specific implementation mode
The invention will be further described below in conjunction with the accompanying drawings.
As shown in Figure 1, a kind of bridge pad disease automatic identifying method based on convolutional neural networks, including following step
Suddenly:
S1:Bridge pad disease photo is obtained, and label information is assigned for every photo, label information corresponds to photo institute
The bridge pad Damage Types of category;
S2:With the method for image procossing, the pixel values of the different color channels of such as random adjustment photo, affine transformation,
Flip horizontal, flip vertical etc. increase the data volume for training convolutional neural networks;
S3:The bridge pad disease photo of acquisition is divided into training set and test set, training set is for training convolutional nerve
Network, test set are used for test network, and all pictures in training set and test set are scaled to the cromogram of predefined size,
And carry out image preprocessing;
S4:Trained convolutional neural networks model on other data sets is obtained, and according to bridge pad
The requirement of disease recognition adjusts the output layer node number of the convolutional neural networks;
S5:In a manner of knowledge migration, with pretreated bridge pad photo training convolutional neural networks are passed through, ladder is used
Degree declines the weights that full articulamentum is adjusted with error backpropagation algorithm, obtains having automatic identification bridge pad disease function
Convolutional neural networks model.
In step S1, the data set of bearing damage photo should include the disease of multiple and different types.
In step S2, the amount of images for training convolutional neural networks is increased by the method for image procossing, for carrying
The generalization ability of high neural network.
In step S3, the size of picture uniformly scales the picture that pixel value is 224 × 224.
In step S3, the preprocess method of image is:Calculate the sum of pixel value of all images then divided by image number
Measure the pixel value for subtracting the mean value image in every piece image to a mean value image.
In step S4, the advance trained model of acquisition is trained classical on ImageNet data sets
VGG16 convolutional neural networks models, the structure chart of VGG16 network models are as shown in Figure 2.
In step S4, the output layer node number of the convolutional neural networks is adjusted according to the requirement of bridge pad disease recognition
Refer to the classification number for the bridge pad disease that adjustment its network Softmax output layers are classified for N number of node, N expression needs,
Softmax layers of calculation formula are as follows:
Wherein, SiFor i-th of input only of this layer, SkFor k-th of net input of the layer.
In step S5, the method for carrying out knowledge migration to neural network is:By trained model is longitudinally drawn in advance
It is divided into each sub-network, when using bridge pad disease photo training, only training adjusts the weights of last full articulamentum and ties up
The weights for holding other each sub-networks are constant.
In step S5, gradient descent method the specific steps are:The gradient of each weights of counting loss function pair, from any point
Start, the negative direction along the gradient moves a distance, continues along gradient reverse direction operation a distance, in this way in new position
The weights of continuous update network.
In step S5, back-propagation algorithm the specific steps are:Convolutional Neural net is being updated using gradient descent method iteration
When the weights of each layer of network, gradient is propagated forward according to chain type Rule for derivation from last layer of network successively.
The above is only a preferred embodiment of the present invention, it should be pointed out that:For the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (10)
1. a kind of bridge pad disease automatic identifying method based on convolutional neural networks, which is characterized in that include the following steps:
S1:Bridge pad disease photo is obtained, and label information is assigned for every photo, label information corresponds to belonging to photo
Bridge pad Damage Types;
S2:With the method for image procossing, increase the data volume for training convolutional neural networks;
S3:The bridge pad disease photo of acquisition is divided into training set and test set, training set is used for training convolutional neural networks,
Test set is used for test network, and all pictures in training set and test set is scaled to the cromogram of predefined size, goes forward side by side
Row image preprocessing;
S4:Trained convolutional neural networks model on other data sets is obtained, and according to bridge pad disease
The requirement of identification adjusts the output layer node number of the convolutional neural networks;
S5:In a manner of knowledge migration, with pretreated bridge pad photo training convolutional neural networks are passed through, using under gradient
Drop adjusts the weights of full articulamentum with error backpropagation algorithm, obtains the convolution with automatic identification bridge pad disease function
Neural network model.
2. the bridge pad disease automatic identifying method according to claim 1 based on convolutional neural networks, feature exist
In:In step S1, the data set of bearing damage photo should include the disease of multiple and different types.
3. the bridge pad disease automatic identifying method according to claim 1 based on convolutional neural networks, feature exist
In:In step S2, the amount of images for training convolutional neural networks is increased by the method for image procossing, for improving nerve
The generalization ability of network.
4. the bridge pad disease automatic identifying method according to claim 1 based on convolutional neural networks, feature exist
In:In step S3, the size of picture is uniformly scaled to the picture that pixel size is 224 × 224.
5. the bridge pad disease automatic identifying method according to claim 1 based on convolutional neural networks, feature exist
In:In step S3, the preprocess method of image is:It calculates the sum of pixel value of all images then divided by the quantity of image obtains
One mean value image subtracts the pixel value of the mean value image in every piece image.
6. the bridge pad disease automatic identifying method according to claim 1 based on convolutional neural networks, feature exist
In:In step S4, the advance trained model of acquisition is the VGG16 of trained classics on ImageNet data sets
Convolutional neural networks model.
7. the bridge pad disease automatic identifying method according to claim 1 based on convolutional neural networks, feature exist
In:In step S4, the output layer node number that the convolutional neural networks are adjusted according to the requirement of bridge pad disease recognition refers to
It is N number of node to adjust its network Softmax output layers, and N indicates the classification number for the bridge pad disease for needing to classify,
Softmax layers of calculation formula are as follows:
Wherein, SiFor i-th of input only of this layer, SkFor k-th of net input of the layer.
8. the bridge pad disease automatic identifying method according to claim 1 based on convolutional neural networks, feature exist
In:In step S5, the method for carrying out knowledge migration to neural network is:It is each by trained model is longitudinally divided in advance
A sub-network, when using bridge pad disease photo training, only training adjusts the weights of last full articulamentum and maintains other
The weights of each sub-network are constant.
9. the bridge pad disease automatic identifying method according to claim 1 based on convolutional neural networks, feature exist
In:In step S5, gradient descent method the specific steps are:The gradient of each weights of counting loss function pair, since any point,
Negative direction along the gradient moves a distance, continues along gradient reverse direction operation a distance in new position, continuous in this way
Update network weights.
10. the bridge pad disease automatic identifying method according to claim 1 based on convolutional neural networks, feature exist
In:In step S5, back-propagation algorithm the specific steps are:Each layer of convolutional neural networks is being updated using gradient descent method iteration
Weights when, gradient is propagated forward according to chain type Rule for derivation from last layer of network successively.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810066766.8A CN108363961A (en) | 2018-01-24 | 2018-01-24 | Bridge pad disease recognition method based on transfer learning between convolutional neural networks |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810066766.8A CN108363961A (en) | 2018-01-24 | 2018-01-24 | Bridge pad disease recognition method based on transfer learning between convolutional neural networks |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108363961A true CN108363961A (en) | 2018-08-03 |
Family
ID=63006912
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810066766.8A Pending CN108363961A (en) | 2018-01-24 | 2018-01-24 | Bridge pad disease recognition method based on transfer learning between convolutional neural networks |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108363961A (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109191445A (en) * | 2018-08-29 | 2019-01-11 | 极创智能(北京)健康科技有限公司 | Bone deformation analytical method based on artificial intelligence |
CN109508650A (en) * | 2018-10-23 | 2019-03-22 | 浙江农林大学 | A kind of wood recognition method based on transfer learning |
CN109753566A (en) * | 2019-01-09 | 2019-05-14 | 大连民族大学 | The model training method of cross-cutting sentiment analysis based on convolutional neural networks |
CN109784438A (en) * | 2018-12-28 | 2019-05-21 | 福建华闽通达信息技术有限公司 | A kind of bridge maintenance disease record, identification and treatment measures guidance method and system |
CN109800806A (en) * | 2019-01-14 | 2019-05-24 | 中山大学 | A kind of corps diseases detection algorithm based on deep learning |
CN109978847A (en) * | 2019-03-19 | 2019-07-05 | 东南大学 | Drag-line housing disease automatic identifying method based on transfer learning Yu drag-line robot |
CN110261108A (en) * | 2019-01-18 | 2019-09-20 | 北京化工大学 | Bearing fault method of identification when specified operating based on CNN color property figure |
CN110457982A (en) * | 2018-12-28 | 2019-11-15 | 中国科学院合肥物质科学研究院 | A kind of crop disease image-recognizing method based on feature transfer learning |
CN110555511A (en) * | 2019-07-24 | 2019-12-10 | 北京踏歌智行科技有限公司 | Method, device, electronic equipment and computer readable storage medium for identifying area |
CN110766045A (en) * | 2019-09-12 | 2020-02-07 | 深圳大学 | Underground drainage pipeline disease identification method, intelligent terminal and storage medium |
CN111191714A (en) * | 2019-12-28 | 2020-05-22 | 浙江大学 | Intelligent identification method for bridge appearance damage diseases |
CN112699736A (en) * | 2020-12-08 | 2021-04-23 | 江西省交通科学研究院 | Bridge bearing fault identification method based on space attention |
CN112749733A (en) * | 2020-11-27 | 2021-05-04 | 江西省交通科学研究院 | Bridge disease identification method based on mixed model and image pyramid |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105825511A (en) * | 2016-03-18 | 2016-08-03 | 南京邮电大学 | Image background definition detection method based on deep learning |
CN106971160A (en) * | 2017-03-23 | 2017-07-21 | 西京学院 | Winter jujube disease recognition method based on depth convolutional neural networks and disease geo-radar image |
CN106991666A (en) * | 2017-02-24 | 2017-07-28 | 中国科学院合肥物质科学研究院 | A kind of disease geo-radar image recognition methods suitable for many size pictorial informations |
CN106991439A (en) * | 2017-03-28 | 2017-07-28 | 南京天数信息科技有限公司 | Image-recognizing method based on deep learning and transfer learning |
-
2018
- 2018-01-24 CN CN201810066766.8A patent/CN108363961A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105825511A (en) * | 2016-03-18 | 2016-08-03 | 南京邮电大学 | Image background definition detection method based on deep learning |
CN106991666A (en) * | 2017-02-24 | 2017-07-28 | 中国科学院合肥物质科学研究院 | A kind of disease geo-radar image recognition methods suitable for many size pictorial informations |
CN106971160A (en) * | 2017-03-23 | 2017-07-21 | 西京学院 | Winter jujube disease recognition method based on depth convolutional neural networks and disease geo-radar image |
CN106991439A (en) * | 2017-03-28 | 2017-07-28 | 南京天数信息科技有限公司 | Image-recognizing method based on deep learning and transfer learning |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109191445A (en) * | 2018-08-29 | 2019-01-11 | 极创智能(北京)健康科技有限公司 | Bone deformation analytical method based on artificial intelligence |
CN109508650A (en) * | 2018-10-23 | 2019-03-22 | 浙江农林大学 | A kind of wood recognition method based on transfer learning |
CN109784438A (en) * | 2018-12-28 | 2019-05-21 | 福建华闽通达信息技术有限公司 | A kind of bridge maintenance disease record, identification and treatment measures guidance method and system |
CN110457982A (en) * | 2018-12-28 | 2019-11-15 | 中国科学院合肥物质科学研究院 | A kind of crop disease image-recognizing method based on feature transfer learning |
CN110457982B (en) * | 2018-12-28 | 2023-04-11 | 中国科学院合肥物质科学研究院 | Crop disease image identification method based on feature migration learning |
CN109753566B (en) * | 2019-01-09 | 2020-11-24 | 大连民族大学 | Model training method for cross-domain emotion analysis based on convolutional neural network |
CN109753566A (en) * | 2019-01-09 | 2019-05-14 | 大连民族大学 | The model training method of cross-cutting sentiment analysis based on convolutional neural networks |
CN109800806A (en) * | 2019-01-14 | 2019-05-24 | 中山大学 | A kind of corps diseases detection algorithm based on deep learning |
CN110261108A (en) * | 2019-01-18 | 2019-09-20 | 北京化工大学 | Bearing fault method of identification when specified operating based on CNN color property figure |
CN109978847A (en) * | 2019-03-19 | 2019-07-05 | 东南大学 | Drag-line housing disease automatic identifying method based on transfer learning Yu drag-line robot |
CN109978847B (en) * | 2019-03-19 | 2023-07-04 | 东南大学 | Automatic inhaul cable sleeve disease identification method based on transfer learning and inhaul cable robot |
CN110555511A (en) * | 2019-07-24 | 2019-12-10 | 北京踏歌智行科技有限公司 | Method, device, electronic equipment and computer readable storage medium for identifying area |
CN110766045A (en) * | 2019-09-12 | 2020-02-07 | 深圳大学 | Underground drainage pipeline disease identification method, intelligent terminal and storage medium |
CN110766045B (en) * | 2019-09-12 | 2023-07-07 | 深圳大学 | Underground drainage pipeline disease identification method, intelligent terminal and storage medium |
CN111191714A (en) * | 2019-12-28 | 2020-05-22 | 浙江大学 | Intelligent identification method for bridge appearance damage diseases |
CN112749733A (en) * | 2020-11-27 | 2021-05-04 | 江西省交通科学研究院 | Bridge disease identification method based on mixed model and image pyramid |
CN112699736A (en) * | 2020-12-08 | 2021-04-23 | 江西省交通科学研究院 | Bridge bearing fault identification method based on space attention |
CN112699736B (en) * | 2020-12-08 | 2024-06-07 | 江西省交通科学研究院有限公司 | Bridge bearing disease identification method based on spatial attention |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108363961A (en) | Bridge pad disease recognition method based on transfer learning between convolutional neural networks | |
CN105825511B (en) | A kind of picture background clarity detection method based on deep learning | |
CN108288269A (en) | Bridge pad disease automatic identifying method based on unmanned plane and convolutional neural networks | |
CN107145908B (en) | A kind of small target detecting method based on R-FCN | |
CN108921051B (en) | Pedestrian attribute identification network and technology based on cyclic neural network attention model | |
CN104978580B (en) | A kind of insulator recognition methods for unmanned plane inspection transmission line of electricity | |
CN107301387A (en) | A kind of image Dense crowd method of counting based on deep learning | |
CN109886141A (en) | A kind of pedestrian based on uncertainty optimization discrimination method again | |
CN110263705A (en) | Towards two phase of remote sensing technology field high-resolution remote sensing image change detecting method | |
CN107103754A (en) | A kind of road traffic condition Forecasting Methodology and system | |
CN109214441A (en) | A kind of fine granularity model recognition system and method | |
CN105760835A (en) | Gait segmentation and gait recognition integrated method based on deep learning | |
CN110163187A (en) | Remote road traffic sign detection recognition methods based on F-RCNN | |
CN108038821A (en) | A kind of image Style Transfer method based on production confrontation network | |
CN108596203A (en) | Optimization method of the pond layer in parallel to pantograph carbon slide surface abrasion detection model | |
CN104992223A (en) | Dense population estimation method based on deep learning | |
CN106203318A (en) | The camera network pedestrian recognition method merged based on multi-level depth characteristic | |
CN106803069A (en) | Crowd's level of happiness recognition methods based on deep learning | |
CN104952073B (en) | Scene Incision method based on deep learning | |
CN110163836A (en) | Based on deep learning for the excavator detection method under the inspection of high-altitude | |
CN107038416A (en) | A kind of pedestrian detection method based on bianry image modified HOG features | |
CN109325495A (en) | A kind of crop image segmentation system and method based on deep neural network modeling | |
CN109376676A (en) | Highway engineering site operation personnel safety method for early warning based on unmanned aerial vehicle platform | |
CN115457006B (en) | Unmanned aerial vehicle inspection defect classification method and device based on similarity consistency self-distillation | |
CN109408985A (en) | The accurate recognition methods in bridge steel structure crack based on computer vision |
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 |
Application publication date: 20180803 |
|
RJ01 | Rejection of invention patent application after publication |