CN110930317A - Ghost imaging method based on convolutional neural network - Google Patents
Ghost imaging method based on convolutional neural network Download PDFInfo
- Publication number
- CN110930317A CN110930317A CN201911046280.9A CN201911046280A CN110930317A CN 110930317 A CN110930317 A CN 110930317A CN 201911046280 A CN201911046280 A CN 201911046280A CN 110930317 A CN110930317 A CN 110930317A
- Authority
- CN
- China
- Prior art keywords
- image
- convolutional neural
- network
- neural network
- resolution
- 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
- 238000003384 imaging method Methods 0.000 title claims abstract description 64
- 238000013527 convolutional neural network Methods 0.000 title claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 18
- 230000005855 radiation Effects 0.000 claims abstract description 6
- 238000001514 detection method Methods 0.000 claims description 17
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000036544 posture Effects 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 5
- 238000012935 Averaging Methods 0.000 claims description 3
- 230000004913 activation Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 2
- 238000000034 method Methods 0.000 description 16
- 238000005070 sampling Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V8/00—Prospecting or detecting by optical means
- G01V8/10—Detecting, e.g. by using light barriers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T9/00—Image coding
- G06T9/002—Image coding using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
Abstract
A ghost imaging method based on a convolutional neural network comprises the following steps: step 1, taking an object image to be recognized or imaged as a target of network training, taking an original object image under low resolution and each angle posture as a sample of the network training, and training by using a convolutional neural network; step 2, detecting a target area by using a random radiation field, receiving an echo signal, and acquiring a low-resolution target image; and 3, inputting the low-resolution target image into the trained network, and quickly outputting the high-resolution target image. The invention has short data acquisition time, low requirement on the quality of the input image, high imaging speed, high-quality target image still obtained under the condition of poor effect of the input image and strong anti-interference capability.
Description
Technical Field
The invention belongs to the field of ghost imaging, and particularly relates to a ghost imaging method based on a convolutional neural network.
Background
Ghost imaging, also known as quantum imaging or correlation imaging, is a novel imaging technique developed on the basis of quantum entanglement. In 1995, it was first completed by the Chinese scientists Shih inkstone and Pittman at the university of Marylan, USA. Compared with the traditional imaging technology, the ghost imaging has the characteristics of no lens imaging, strong disturbance resistance, no localization and the like, and has good application prospect in the aspects of remote sensing imaging, weak light detection, medical imaging, security inspection, penetrating scattering medium imaging and the like. The mechanism of ghost imaging is different from the traditional imaging, and the ghost imaging is imaged by using a photon high-order correlation mode. In the imaging process, the target is continuously irradiated by the detection speckles to form barrel detection data, the information detected by the detection speckles and the barrel is known, and correlation operation is performed by using the detection speckles and barrel detection signals, which is a traditional basic correlation method in the ghost imaging field. In addition, algorithms of the compressed sensing class are also applied in ghost imaging signal processing. The compressed sensing method utilizes a random matrix to detect the target, and random speckles projected in ghost imaging also belong to the random matrix. Therefore, ghost imaging naturally satisfies the conditions required by compressed sensing theory, and the L1 norm is usually chosen for solution.
However, both of these methods have some problems. The basic correlation method utilizes high-order correlation of photons to image through coincidence operation, the algorithm is simple and visual, but a large amount of speckles are needed to perform projection accumulation on a target, the sampling times are large, the data acquisition time is too long, and the imaging quality is poor. Meanwhile, although the algorithm can bring stronger disturbance resistance than that of the traditional imaging, the detection signal-to-noise ratio of the algorithm is rapidly reduced along with the increase of the number of pixel points required by imaging, and the imaging quality is influenced. Therefore, in order to solve the problems of excessive sampling times of basic correlation algorithms and the like, researchers introduce a compressed sensing technology into ghost imaging, but the compressed sensing algorithms are high in calculation complexity, so that the algorithms are long in time consumption, and the wide application of the compressed sensing algorithms in the ghost imaging field is also limited.
Disclosure of Invention
Aiming at the problem that the existing ghost imaging technology can not take the sampling rate, the calculation efficiency and the imaging quality into consideration, the invention provides the ghost imaging method based on the convolutional neural network, which has the advantages of short data acquisition time, quick imaging, high image quality and strong anti-interference capability.
In order to achieve the purpose, the invention has the following technical scheme:
a ghost imaging method based on a convolutional neural network comprises the following steps:
and 3, inputting the low-resolution target image into the trained network, and quickly outputting the high-resolution target image.
Preferably, in an embodiment of the ghost imaging method based on the convolutional neural network, in step 2, a bucket detector or a single antenna is used to receive the echo signal.
Preferably, in an embodiment of the ghost imaging method based on the convolutional neural network, in step 2, the low-resolution target image is acquired by using a basic correlation algorithm or a compressive sensing algorithm.
Preferably, in an embodiment of the ghost imaging method based on the convolutional neural network, in step 2, the basic correlation algorithm expression is as follows:
wherein, g(2)In order to be the second-order degree of correlation,<·>for ensemble averaging, IbI (x, y) is the spatial distribution of target plane speckles for echo signals containing target information received by a bucket detector, and the above formula is in a normalized form; and restoring the object image through multiple coincidence operations by utilizing the correlation between the detection signal and the echo signal.
Preferably, an embodiment of the present invention relates to a convolutional neural network-based ghost imaging methodIn step 2, the compressed sensing algorithm detects the target by using the random matrix, and selects L1The norm is solved as follows:
wherein, A is the detection speckle, x is the original image, y is the bucket detection data, and lambda is the regularization factor.
Preferably, in an embodiment of the convolutional neural network-based ghost imaging method according to the present invention, the ghost imaging expression of the target image output in step 3 is: y is Ax + b; wherein y is the received echo signal, A is the detected speckle, x is the original image of the object to be restored, and b is noise; in the ideal case, the original image x is recovered by solving the above equation, without considering the influence of noise, since both y and a are known signals in the ghost imaging system.
Preferably, in an embodiment of the ghost imaging method based on the convolutional neural network, in step 3, the relation between the input and the output is:
where a is the output of the network, akIs an input to the network, wkIs the weight of the network, b is the bias, σ is the activation function of the network; taking the original images of the object in low resolution and various angular postures as input akAnd taking the high-resolution original image as an output a, and establishing a mapping relation between the high-resolution image and the low-resolution image through training.
Compared with the prior art, the invention has the following beneficial effects: based on the ghost imaging framework, the target is detected by utilizing the random radiation field, and the echo signal is received. Compared with the traditional basic correlation method and the traditional compressed sensing method, the method establishes a mapping relation between input and output in a network training mode, has high imaging speed and can quickly recover the target image. The invention establishes the pixel-level mapping relation between the low signal-to-noise ratio image and the high signal-to-noise ratio image in a pre-training mode, and can still obtain the high-quality target image under the condition of poor input image effect. Because the method can recover the target image with high quality under the condition of image input with low signal-to-noise ratio, compared with other traditional ghost imaging algorithms, the ghost imaging method based on the convolutional neural network has better anti-jamming capability.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on the drawings without creative efforts.
FIG. 1 is a flow chart of a ghost imaging method of the present invention;
FIG. 2 is a graph of experimental results of an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, also belong to the protection scope of the present invention.
Referring to fig. 1, the ghost imaging method based on the convolutional neural network of the present invention includes the following steps;
step 1: taking an object image to be recognized or imaged as a target of network training, taking an original object image under low resolution and each angle posture as a sample of the network training, and training by using a convolutional neural network;
step 2: detecting a target area by using a random radiation field, receiving an echo signal by using a barrel detector or a single antenna, and acquiring a low-resolution target image by using a basic correlation or compressed sensing algorithm;
and step 3: and inputting the low-resolution target image into the trained network, and quickly outputting the high-resolution target image.
Specifically, the imaging process of ghost imaging can be expressed as:
y＝Ax+b；
where y is the received echo signal, i.e., the bucket detection signal, a is the detected speckle, x is the original image of the object to be restored, and b is noise. Ideally, if noise is not considered, both y and a are known signals in a ghost imaging system, and the original image x of the object is recovered by solving the above equation.
The relationship between the inputs and outputs of a classical neural network can be expressed as:
where a is the output of the network, akIs an input to the network, wkIs the weight of the network, b is the bias, and σ is the activation function of the network. In the training process, the original object image with low resolution and various angle postures is used as an input akAnd taking the high-resolution original image as an output a, and establishing a mapping relation between the high-resolution image and the low-resolution image through training.
In the actual use process, based on a ghost imaging framework, a target is detected by using a random radiation field, echo signals detected by a barrel are received, and preprocessing is performed by using a basic correlation algorithm or a compressed sensing algorithm, wherein the basic correlation algorithm is expressed as follows:
wherein, g(2)In order to be the second-order degree of correlation,<·>for ensemble averaging, IbDetecting received echo signals containing target information for the tank, I (x, y)The above equation is a normalized form for the spatial distribution of the target planar speckle.
And restoring the object image through multiple coincidence operations by utilizing the correlation between the detection signal and the echo signal.
The compressed sensing algorithm detects the target by using a random matrix, and random speckles projected by ghost imaging also belong to the random matrix. Thus, ghost imaging naturally satisfies the conditions required by compressed sensing theory, and L is usually chosen1The norm is solved as follows:
wherein, A is the detection speckle, x is the original image, y is the bucket detection data, and lambda is the regularization factor.
And preprocessing by using a basic correlation algorithm or a compressed sensing algorithm, and performing primary reconstruction under the condition of low sampling rate to obtain a target image with low resolution. And inputting the target image with low resolution into the trained convolutional neural network, and quickly outputting the target image with high quality.
Examples
The experimental target is a toy airplane model, the distance between the target and the light source is 27.5 cm, the distance between the target and the receiver is 45 cm, and the distance between the light source and the receiver is 17.5 cm. The power of the light source is 30 muW/cm2The method comprises the steps of continuously projecting random speckles with different sizes of 64 x 64 pixels, receiving 300000 echo signals, randomly selecting 200 samples for testing, and randomly selecting 100 samples for testing. The network was trained at 20% sampling rate and tested at 5%, 10%, 15% and 20% sampling rate.
The experimental results are shown in fig. 2, which are input and output comparison results at a sampling rate of 5% -20% from left to right, and the rightmost side is the target original image photographed by the CCD. It can be seen that, no matter what sampling rate, the quality of the output image is greatly improved relative to the input image, and the image becomes clearer and clearer along with the improvement of the sampling rate.
Table 1 shows the comparison of the basic correlation and compressed sensing algorithm in the present invention and the ghost imaging. Two indexes of peak signal-to-noise ratio (PSNR) and time consumed by the algorithm are selected for comparison.
TABLE 1
It can be seen that the PSNR of the method of the present invention increases with the increase of the sampling rate, and when the sampling rate increases to 20%, the PSNR is substantially the same as those of the other two methods, but the time consumption of the algorithm is much shorter than that of the other two methods.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (7)
1. A ghost imaging method based on a convolutional neural network is characterized by comprising the following steps:
step 1, taking an object image to be recognized or imaged as a target of network training, taking an original object image under low resolution and each angle posture as a sample of the network training, and training by using a convolutional neural network;
step 2, detecting a target area by using a random radiation field, receiving an echo signal, and acquiring a low-resolution target image;
and 3, inputting the low-resolution target image into the trained network, and quickly outputting the high-resolution target image.
2. A convolutional neural network-based ghost imaging method as claimed in claim 1, wherein said step 2 uses a bucket detector or a single antenna to receive the echo signal.
3. A convolutional neural network-based ghost imaging method as claimed in claim 1, wherein said step 2 is performed by using a basic correlation algorithm or a compressive sensing algorithm to obtain a low resolution target image.
4. A convolutional neural network-based ghost imaging method as claimed in claim 1, wherein the basic correlation algorithm expression in step 2 is:
wherein, g(2)In order to be the second-order degree of correlation,<·>for ensemble averaging, IbI (x, y) is the spatial distribution of target plane speckles for echo signals containing target information received by a bucket detector, and the above formula is in a normalized form; and restoring the object image through multiple coincidence operations by utilizing the correlation between the detection signal and the echo signal.
5. A convolutional neural network-based ghost imaging method as claimed in claim 1, wherein the compressed sensing algorithm in step 2 utilizes a random matrix to detect the target, and selects L1The norm is solved as follows:
wherein, A is the detection speckle, x is the original image, y is the bucket detection data, and lambda is the regularization factor.
6. A convolutional neural network-based ghost imaging method as claimed in claim 1, wherein the ghost imaging expression of the target image output in step 3 is: y is Ax + b; wherein y is the received echo signal, A is the detected speckle, x is the original image of the object to be restored, and b is noise; in the ideal case, the original image x is recovered by solving the above equation, without considering the influence of noise, since both y and a are known signals in the ghost imaging system.
7. A convolutional neural network-based ghost imaging method as claimed in claim 1, wherein the relationship between the input and the output in step 3 is:
where a is the output of the network, akIs an input to the network, wkIs the weight of the network, b is the bias, σ is the activation function of the network; taking the original images of the object in low resolution and various angular postures as input akAnd taking the high-resolution original image as an output a, and establishing a mapping relation between the high-resolution image and the low-resolution image through training.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911046280.9A CN110930317A (en) | 2019-10-30 | 2019-10-30 | Ghost imaging method based on convolutional neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911046280.9A CN110930317A (en) | 2019-10-30 | 2019-10-30 | Ghost imaging method based on convolutional neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110930317A true CN110930317A (en) | 2020-03-27 |
Family
ID=69849888
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911046280.9A Pending CN110930317A (en) | 2019-10-30 | 2019-10-30 | Ghost imaging method based on convolutional neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110930317A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111626285A (en) * | 2020-05-27 | 2020-09-04 | 北京环境特性研究所 | Character recognition system and method |
| CN111833248A (en) * | 2020-06-19 | 2020-10-27 | 西北大学 | Super-resolution ghost imaging method and system based on partial Hadamard matrix |
| CN112435189A (en) * | 2020-11-23 | 2021-03-02 | 湖北工业大学 | Computed ghost imaging method and system based on self-coding network |
| CN112528731A (en) * | 2020-10-27 | 2021-03-19 | 西安交通大学 | Plane wave beam synthesis method and system based on double-regression convolutional neural network |
| CN112802145A (en) * | 2021-01-27 | 2021-05-14 | 四川大学 | Color calculation ghost imaging method based on deep learning |
| CN113099207A (en) * | 2021-03-31 | 2021-07-09 | 吉林工程技术师范学院 | Phase modulation-based micro-lens array type deep learning three-dimensional ghost imaging method |
| CN113393392A (en) * | 2021-06-11 | 2021-09-14 | 清华大学深圳国际研究生院 | Dynamic target ghost imaging system and method based on neural network |
| CN113706408A (en) * | 2021-08-11 | 2021-11-26 | 西安交通大学 | Ghost imaging denoising method and device based on noise reduction convolutional neural network |
| CN114494055A (en) * | 2022-01-19 | 2022-05-13 | 西安交通大学 | Ghost imaging method, system, equipment and storage medium based on circulating neural network |
-
2019
- 2019-10-30 CN CN201911046280.9A patent/CN110930317A/en active Pending
Non-Patent Citations (2)
| Title |
|---|
| YUCHEN HE ETC.: "Ghost Imaging Based on Deep Learning", 《SCIENTIFIC REPORTS》 * |
| 梁礼明: "《优化方法导论》", 30 September 2017, 北京理工大学出版社 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111626285A (en) * | 2020-05-27 | 2020-09-04 | 北京环境特性研究所 | Character recognition system and method |
| CN111833248A (en) * | 2020-06-19 | 2020-10-27 | 西北大学 | Super-resolution ghost imaging method and system based on partial Hadamard matrix |
| CN111833248B (en) * | 2020-06-19 | 2023-06-16 | 西北大学 | Super-resolution ghost imaging method and system based on partial Hadamard matrix |
| CN112528731A (en) * | 2020-10-27 | 2021-03-19 | 西安交通大学 | Plane wave beam synthesis method and system based on double-regression convolutional neural network |
| CN112528731B (en) * | 2020-10-27 | 2024-04-05 | 西安交通大学 | Plane wave beam synthesis method and system based on dual regression convolutional neural network |
| CN112435189A (en) * | 2020-11-23 | 2021-03-02 | 湖北工业大学 | Computed ghost imaging method and system based on self-coding network |
| CN112802145A (en) * | 2021-01-27 | 2021-05-14 | 四川大学 | Color calculation ghost imaging method based on deep learning |
| CN113099207B (en) * | 2021-03-31 | 2022-09-02 | 吉林工程技术师范学院 | Phase modulation-based micro-lens array type deep learning three-dimensional ghost imaging method |
| CN113099207A (en) * | 2021-03-31 | 2021-07-09 | 吉林工程技术师范学院 | Phase modulation-based micro-lens array type deep learning three-dimensional ghost imaging method |
| CN113393392A (en) * | 2021-06-11 | 2021-09-14 | 清华大学深圳国际研究生院 | Dynamic target ghost imaging system and method based on neural network |
| CN113706408A (en) * | 2021-08-11 | 2021-11-26 | 西安交通大学 | Ghost imaging denoising method and device based on noise reduction convolutional neural network |
| CN114494055A (en) * | 2022-01-19 | 2022-05-13 | 西安交通大学 | Ghost imaging method, system, equipment and storage medium based on circulating neural network |
| CN114494055B (en) * | 2022-01-19 | 2024-02-06 | 西安交通大学 | Ghost imaging method, ghost imaging system, ghost imaging equipment and ghost imaging storage medium based on cyclic neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110930317A (en) | Ghost imaging method based on convolutional neural network | |
| CN112597864B (en) | Monitoring video anomaly detection method and device | |
| CN106447668B (en) | The small target detecting method restored under IR Scene based on grab sample and sparse matrix | |
| CN105931196A (en) | Fourier optical modeling-based coded aperture camera image restoration method | |
| CN107833182A (en) | The infrared image super resolution ratio reconstruction method of feature based extraction | |
| WO2020036782A2 (en) | Methods and systems for object recognition in low illumination conditions | |
| Guo et al. | An anchor-free network with density map and attention mechanism for multiscale object detection in aerial images | |
| Ding et al. | Uncertainty-based thin cloud removal network via conditional variational autoencoders | |
| Li et al. | Infrared dim target detection method inspired by human vision system | |
| CN115731174A (en) | Infrared small target detection method and device based on image information entropy and multi-scale local contrast measurement | |
| Wang et al. | Single image haze removal via attention-based transmission estimation and classification fusion network | |
| Chen et al. | Space-based infrared aerial target detection method via interframe registration and spatial local contrast | |
| CN105608674B (en) | A kind of image enchancing method based on image registration, interpolation and denoising | |
| CN113706408A (en) | Ghost imaging denoising method and device based on noise reduction convolutional neural network | |
| Pan et al. | Detection method for small and dim targets from a time series of images observed by a space-based optical detection system | |
| Zhou et al. | Infrared small target detection via learned infrared patch-image convolutional network | |
| Zhang et al. | Infrared small target detection based on multidirectional cumulative measure | |
| CN107807444A (en) | A kind of projecting apparatus " street corner imaging " method based on difference | |
| Zhou et al. | Dim and small target detection based on their living environment | |
| Tehseen et al. | Enhanced imaging through turbid water based on quadrature lock-in discrimination and retinex aided by adaptive gamma function for illumination correction | |
| CN110044474B (en) | Correlation imaging optimization method based on interpolation algorithm | |
| Lv et al. | Turn a silicon camera into an ingaas camera | |
| Gunawan et al. | Inferring the level of visibility from hazy images | |
| Duncan et al. | Relational entropy-based saliency detection in images and videos | |
| Han et al. | Dual discriminators generative adversarial networks for unsupervised infrared super-resolution |
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: 20200327 |
|
| RJ01 | Rejection of invention patent application after publication |