CN114998311A - Part precision detection method based on homomorphic filtering - Google Patents
Part precision detection method based on homomorphic filtering Download PDFInfo
- Publication number
- CN114998311A CN114998311A CN202210816252.6A CN202210816252A CN114998311A CN 114998311 A CN114998311 A CN 114998311A CN 202210816252 A CN202210816252 A CN 202210816252A CN 114998311 A CN114998311 A CN 114998311A
- Authority
- CN
- China
- Prior art keywords
- frequency
- circle
- pixel points
- circumference
- radius
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning 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
- G06T2207/20024—Filtering details
-
- 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]
Abstract
The invention relates to the technical field of data processing, in particular to a part precision detection method based on homomorphic filtering. The method is characterized in that a data processing method after data acquisition is improved, and different most applicable homomorphic filter cutoff frequencies are given for different images. The method provided by the invention solves the technical problem that the surface defects of the parts cannot be efficiently and accurately determined in the prior art by improving the data processing method, improves the efficiency and accuracy of part precision detection, can be integrated into an artificial intelligence system in the production field, and can be used as an artificial intelligence optimization operation system, an artificial intelligence middleware and the like for developing computer vision software.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a part precision detection method based on homomorphic filtering.
Background
Because the precision parts play an important role in the whole mechanical structure of the machine, and some slight differences can cause the machine to have problems, when the precision parts are detected, the collected precision part images for completing the defect detection of the precision parts have high requirements, and the precision part images are required to reflect the surface defects and the contour information of the precision parts as accurately as possible and display the details of the parts as more as possible.
Because the precision part is metal, so inevitably can lead to the precision part image luminance uneven that gathers because of the reflection of light, consequently in order to ensure to detect the precision, prior art generally adopts homomorphic filtering to carry out enhancement processing to the image to solve the uneven problem of illumination.
However, when the homomorphic filtering is used for enhancing the images, because the most suitable homomorphic filtering cutoff frequency of each image is different, the prior art detection method for evaluating the homomorphic filtering cutoff frequency by experience cannot ensure the enhancement processing effect of the homomorphic filtering, and the detection result is inaccurate; if the most suitable homomorphic filtering cutoff frequency of each image is determined, a large amount of experimental calculation is needed in the prior art, so that the calculation cost is increased, and the detection efficiency is low.
Disclosure of Invention
The invention provides a part precision detection method based on homomorphic filtering, which is used for solving the problem that the prior art can not efficiently and accurately adopt homomorphic filtering to finish part precision detection, and adopts the following technical scheme:
the invention discloses a part precision detection method based on homomorphic filtering, which comprises the following steps of:
identifying the surface of the part to obtain a part surface image;
enhancing the part surface image by adopting a homomorphic filtering method to obtain an enhanced part surface image;
the method for determining the cut-off frequency of the filter function used in the homomorphic filtering process comprises the following steps:
performing time-frequency conversion on the part surface image to obtain a spectrogram of the part surface image;
sequentially increasing the circle radius by taking the highest highlight central point of the spectrogram as the circle center and taking the circle center as a starting point according to a set step length, and calculating the proportion value of pixel points containing brightness information on the circumference corresponding to each circle radius in all the pixel points on the circumference;
calculating the difference value of the ratio values corresponding to any two adjacent circle radiuses, determining the maximum difference value of all the difference values, further determining the two adjacent circle radiuses corresponding to the maximum difference value, and taking the frequency value corresponding to the smaller radius of the two adjacent circle radiuses corresponding to the maximum difference value on a spectrogram as the cut-off frequency of a filter function;
and inputting the enhanced part surface image into a neural network model, and judging whether the part surface has defects or not.
The invention has the beneficial effects that:
the invention obtains the surface data of the part from the image based on the surface image of the part obtained by identification, then carries out data processing analysis on the obtained surface data of the part, and provides a cutoff frequency calculation method which is most suitable for the current surface image of the part when the surface image of each part is enhanced by using a homomorphic filtering method based on the frequency spectrogram of the surface image of the part obtained by conversion. Obviously, the method is integrated into an artificial intelligence system in the production field, or used as an artificial intelligence optimization operation system and an artificial intelligence middleware, or used for developing computer vision software, and can obviously improve the precision detection efficiency and accuracy of parts.
Further, the process of sequentially calculating the proportion value of the pixel points containing the brightness information in all the pixel points on the circumference corresponding to each circle radius is as follows:
performing Hough circle detection on the frequency spectrogram based on the circle center, and performing binarization processing on pixel points in the frequency spectrogram to reassign the pixel points to gray values 1 and 0, wherein the pixel points with the gray values of 1 represent the pixel points containing brightness information, and the pixel points with the gray values of 0 represent the pixel points not containing the brightness information;
taking the circle center as a starting point, taking one pixel point as a step length, gradually increasing the circle radius, and counting the proportion of the pixel point with the gray value of 1 on the circumference corresponding to different circle radii to all the pixel points on the circumference:
in the formulaRepresenting the number of all pixel points on the circumference when the radius of the circle is r,indicating the number of pixels having a gray value of 1 on the circumference when the radius of the circle is r,and the ratio of the pixel point with the gray value of 1 to all the pixel points on the circumference is expressed when the radius of the circle is r.
Further, the filter function is:
wherein, the first and the second end of the pipe are connected with each other,in order to be a function of the filtering,andrespectively a maximum amplitude and a minimum amplitude,,,in order to cut-off the frequency of the frequency,and the frequency difference between the frequency of a pixel point (U, V) in the frequency spectrum graph of the part surface image and the central frequency is obtained.
Further, the neural network model is a DNN neural network model.
Drawings
FIG. 1 is a flow chart of the part precision detection method based on homomorphic filtering according to the present invention.
Detailed Description
The following describes a part precision detection method based on homomorphic filtering in detail with reference to the accompanying drawings and embodiments.
The method comprises the following steps:
the embodiment of the part precision detection method based on homomorphic filtering is shown in figure 1, and the specific process is as follows:
step one, collecting the surface image of the part by using image collecting equipment.
And performing pattern recognition by using related electronic equipment, such as an industrial camera, so as to obtain an image of the surface of the part.
The gray scale processing is performed on the obtained part surface image by using a weighted average method to obtain a part surface gray scale image, but other gray scale processing methods in the prior art may be adopted.
Because the part is made of metal and the metal surface can reflect light, the collected part surface image can show the condition that the brightness of the reflection part is very high and the rest part is darker, and correspondingly, the gray level image of the grayed part surface has the condition that the gray level value of the reflection part is large and the gray level value of the rest part is small.
Because the identified part surface image has the phenomenon of uneven illumination, homomorphic filtering is needed to be adopted for enhancing the part surface image.
And step two, performing enhancement processing on the surface image of the part by adopting an improved homomorphic filtering method.
1. And processing the part surface image by using homomorphic filtering.
The obtained gray image of the surface of the part is regarded as being formed by combining an incident component and a reflection component, and the image is recorded as:
Wherein, the first and the second end of the pipe are connected with each other,in order to be the incident component of the light,is the reflected component.
The incident component belongs to a low-frequency part with slow change, the gray level in a low-frequency area changes slowly, the gray level in an area is approximately the same in the image, for example, a strong light direct-irradiating area, and in order to solve the problem of uneven illumination, the frequency of the part is a part needing to be reduced; the reflection component corresponds to a high frequency region that changes rapidly, and the high frequency region reflects details and edge portions, and needs to be enhanced in order to enhance the details. Therefore, when the image is enhanced by using homomorphic filtering, the low-frequency information is reduced, and the high-frequency information is increased, so that the image details are highlighted, and more information is displayed.
The reflected component and the incident component are closely connected in the time domain, and the direct processing of the time domain is not good, so that homomorphic filtering converts the reflected component and the incident component into the frequency domain for processing, and logarithms are taken on two sides of the equal sign of formula (1):
obtaining a formula (2), and then performing Fourier transform on two equal-sign sides of the formula (2):
at this time, the gray image of the surface of the part in the formula (3)The image is a spectrogram, the middle of the image is a high-brightness low-frequency area, and the outer side of the image is a dark high-frequency area. After Fourier transform processing, the gray level image of the surface of the part can be considered to be obtained by adding a high-frequency image and a low-frequency image, and a filter function is adopted subsequentlyFor imagesAnd (3) carrying out filtering treatment:
by separately aligning imagesThe high-frequency image and the low-frequency image in the middle are processed, so that the low-frequency information is reduced and the high-frequency information is increased.
Then, performing inverse fourier transform on the processed image, namely, formula (4), to obtain:
then, the index is taken from the formula (5):
2. And optimizing the determination mode of the filter function used by homomorphic filtering.
In the above process of processing the surface image of the part by using homomorphic filtering, it can be seen that the most important in the whole homomorphic filtering process is to use the filter, i.e. the filtering functionSpecifically, the cutoff frequency of the filter function is determined.
For spectrogramsIn other words, the middle highlight region is a low-frequency region, the outer dark region is a high-frequency region, and the center of the most middle highlight center point of the spectrogram is recorded as the center of a circleAnd extending outwards, the points closer to the center of the circle have smaller corresponding frequencies, and the points farther away from the center of the circle have larger corresponding frequencies, that is, the corresponding frequencies are different according to the selected radius from the center of the circle.
According to the above characteristics of the spectrogram, the invention provides a method for determining the homomorphic filtering cutoff frequency:
to be provided withPerforming Hough circle detection on the spectrogram for the circle center, performing binarization processing on pixel points in the spectrogram, and assigning the pixel points to two classes, namely pixel points with a gray value of 1 and pixel points with a gray value of 0, wherein the pixel points with a gray value of 1 representAnd the pixel points containing the brightness information and the pixel points with the gray value of 0 represent the pixel points without the brightness information.
Taking a pixel point as a step length, and countingThe proportion of the pixel points with the gray value of 1 on the circumference corresponding to different radiuses as the circle center to all the pixel points on the circumference is as follows:
in the formulaRepresenting the number of all pixel points on the circumference formed with radius r,indicating the number of pixels with a gray value of 1 on the circumference with radius r.And the ratio of the pixel point with the gray value of 1 to all the pixel points on the circumference when the radius is r is represented.
For the spectrogram, the spectrogram consists of a central highlight part and a peripheral dark part, and there is no strict standard for distinguishing high and low frequenciesAnd sequentially making differences according to the sequence, and determining the maximum difference obtained by making the differences:
it indicates the ratio to any two adjacent valuesThe maximum value obtained by the difference is taken,representing a radius ofThe proportion of the pixel point with the gray value of 1 on the corresponding circumference to all the pixel points on the circumference,representing a radius ofThe ratio and radius of the corresponding pixel point with the gray value of 1 on the circumference to all the pixel points on the circumferenceCompared with radiusIs one pixel point long, radiusThe corresponding frequency is the low frequency dividing frequency.
For homomorphic filtering, the required parameters are the cut-off frequency, the maximum amplitude and the minimum amplitude, wherein the cut-off frequency means that the filter function does not increase any more within a certain radial distance, and is recorded asIn the invention, the Hough circle is used to determine the boundary frequency of high and low frequencies, which is actually the cut-off frequency, i.e. the cut-off frequencyTaking the value as a radiusThe corresponding frequency value.
The maximum amplitude and the minimum amplitude respectively represent the enhancement effect and the suppression effect, and the maximum amplitude and the minimum amplitude are respectively set as,Order maximum amplitudeEnsuring that the high frequency region can be enhanced to a minimum amplitudeGuarantee again when can restraining low frequency information and guarantee that low frequency information can pass through the wave filter, the wave filter main part is gaussian filter, has consequently obtained the wave filter of this scheme:
so far, the cut-off frequency of homomorphic filtering which is most suitable for the image is determined by Hough change on the frequency domain, thereby determining homomorphic filtering functions used by the invention.
And step three, detecting whether the part has defects by using a neural network.
After the surface image of the part is enhanced, the precision of the part can be detected based on the enhanced surface image of the part, and whether the surface of the part has defects or not can be judged. The invention uses a neural network bounding box target detection method to detect the part defect part.
The neural network is a DNN neural network, and the specific content is as follows:
1) the network adopts an Encoder-Decoder form, the input of the network is an enhanced part surface image, and the output is a center point of the bounding box, the width and height dimensions of the regressed bounding box and a corresponding probability value.
2) The training set used by the network is an image of the surface of the part with a defect.
3) Downsampling the part surface image input into the neural network by convolution and pooling operations to extract spatial features to identify where there is a defect, enclosing the possible defect with a bounding box, and wherein the probability of being a defect is given.
After the network processing is completed, several bounding boxes are obtained, and probability values corresponding to the bounding boxes are obtained, and if the probability value corresponding to a certain bounding box is greater than 70%, the area in the bounding box is considered to have defects.
Therefore, the neural network can be adopted to judge whether the surface of the part has defects.
On the whole, the most suitable filter function cut-off frequency of each image needing homomorphic filtering can be correspondingly given out by the method, so that the enhancement effect of each image is optimal, and compared with the conventional calculation method, the calculation method of the homomorphic filtering cut-off frequency provided by the invention reduces the calculation amount, and finally, the method can more efficiently and accurately adopt homomorphic filtering to finish part precision detection.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and are not limited thereto; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.
Claims (4)
1. A part precision detection method based on homomorphic filtering is characterized by comprising the following steps:
identifying the surface of the part to obtain a part surface image;
enhancing the part surface image by adopting a homomorphic filtering method to obtain an enhanced part surface image;
the method for determining the cut-off frequency of the filtering function used in the homomorphic filtering process comprises the following steps:
performing time-frequency conversion on the part surface image to obtain a spectrogram of the part surface image;
sequentially increasing the circle radius by taking the highest highlight central point of the spectrogram as the circle center and taking the circle center as a starting point according to a set step length, and calculating the proportion value of pixel points containing brightness information on the circumference corresponding to each circle radius in all the pixel points on the circumference;
calculating the difference value of the ratio values corresponding to any two adjacent circle radiuses, determining the maximum difference value of all the difference values, further determining the two adjacent circle radiuses corresponding to the maximum difference value, and taking the frequency value corresponding to the smaller radius of the two adjacent circle radiuses corresponding to the maximum difference value on a spectrogram as the cut-off frequency of a filter function;
and inputting the enhanced part surface image into a neural network model, and judging whether the part surface has defects or not.
2. The method of claim 1, wherein the process of sequentially calculating the percentage of the pixels containing luminance information to all pixels on the circumference corresponding to each circle radius is as follows:
performing Hough circle detection on the frequency spectrogram based on the circle center, and performing binarization processing on pixel points in the frequency spectrogram to reassign the pixel points to gray values 1 and 0, wherein the pixel points with the gray values of 1 represent the pixel points containing brightness information, and the pixel points with the gray values of 0 represent the pixel points not containing the brightness information;
taking the circle center as a starting point, taking one pixel point as a step length, gradually increasing the circle radius, and counting the proportion of the pixel point with the gray value of 1 on the circumference corresponding to different circle radii to all the pixel points on the circumference:
in the formulaThe number of all the pixel points on the circumference when the radius of the circle is r,indicating the number of pixels with a gray value of 1 on the circumference when the radius of the circle is r,and the ratio of the pixel point with the gray value of 1 to all the pixel points on the circumference is expressed when the radius of the circle is r.
3. The method for detecting precision of parts based on homomorphic filtering according to claim 1 or 2, wherein the filtering function is:
wherein the content of the first and second substances,in order to be a function of the filtering,andrespectively a maximum amplitude and a minimum amplitude,,,in order to cut-off the frequency of the frequency,and the frequency difference between the frequency of a pixel point (U, V) in the frequency spectrum graph of the part surface image and the central frequency is obtained.
4. The homomorphic filtering-based part precision detection method of claim 1, wherein the neural network model is a DNN neural network model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210816252.6A CN114998311A (en) | 2022-07-11 | 2022-07-11 | Part precision detection method based on homomorphic filtering |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210816252.6A CN114998311A (en) | 2022-07-11 | 2022-07-11 | Part precision detection method based on homomorphic filtering |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114998311A true CN114998311A (en) | 2022-09-02 |
Family
ID=83019519
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210816252.6A Pending CN114998311A (en) | 2022-07-11 | 2022-07-11 | Part precision detection method based on homomorphic filtering |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114998311A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116576806A (en) * | 2023-04-21 | 2023-08-11 | 深圳市磐锋精密技术有限公司 | Precision control system for thickness detection equipment based on visual analysis |
CN117635610B (en) * | 2024-01-25 | 2024-04-09 | 青岛正大正电力环保设备有限公司 | Visual detection method for oil leakage of oil pipe of hydraulic tensioning mechanism |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103499590A (en) * | 2013-10-17 | 2014-01-08 | 福州大学 | Method and system for detecting and screening end defects in circular parts |
CN105389814A (en) * | 2015-11-03 | 2016-03-09 | 浙江工业大学 | Air bubble detection method for air tightness test |
CN106373104A (en) * | 2016-09-07 | 2017-02-01 | 河海大学常州校区 | Adaptive enhancement method of rock boring image |
CN109741314A (en) * | 2018-12-29 | 2019-05-10 | 广州博通信息技术有限公司 | A kind of visible detection method and system of part |
CN109829876A (en) * | 2018-05-30 | 2019-05-31 | 东南大学 | Carrier bar on-line detection device of defects and method based on machine vision |
CN113538432A (en) * | 2021-09-17 | 2021-10-22 | 南通蓝城机械科技有限公司 | Part defect detection method and system based on image processing |
CN113989233A (en) * | 2021-10-28 | 2022-01-28 | 太原理工大学 | Image-based part surface roughness support vector machine detection method and system |
-
2022
- 2022-07-11 CN CN202210816252.6A patent/CN114998311A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103499590A (en) * | 2013-10-17 | 2014-01-08 | 福州大学 | Method and system for detecting and screening end defects in circular parts |
CN105389814A (en) * | 2015-11-03 | 2016-03-09 | 浙江工业大学 | Air bubble detection method for air tightness test |
CN106373104A (en) * | 2016-09-07 | 2017-02-01 | 河海大学常州校区 | Adaptive enhancement method of rock boring image |
CN109829876A (en) * | 2018-05-30 | 2019-05-31 | 东南大学 | Carrier bar on-line detection device of defects and method based on machine vision |
CN109741314A (en) * | 2018-12-29 | 2019-05-10 | 广州博通信息技术有限公司 | A kind of visible detection method and system of part |
CN113538432A (en) * | 2021-09-17 | 2021-10-22 | 南通蓝城机械科技有限公司 | Part defect detection method and system based on image processing |
CN113989233A (en) * | 2021-10-28 | 2022-01-28 | 太原理工大学 | Image-based part surface roughness support vector machine detection method and system |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116576806A (en) * | 2023-04-21 | 2023-08-11 | 深圳市磐锋精密技术有限公司 | Precision control system for thickness detection equipment based on visual analysis |
CN116576806B (en) * | 2023-04-21 | 2024-01-26 | 深圳市磐锋精密技术有限公司 | Precision control system for thickness detection equipment based on visual analysis |
CN117635610B (en) * | 2024-01-25 | 2024-04-09 | 青岛正大正电力环保设备有限公司 | Visual detection method for oil leakage of oil pipe of hydraulic tensioning mechanism |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116168026B (en) | Water quality detection method and system based on computer vision | |
CN114998311A (en) | Part precision detection method based on homomorphic filtering | |
CN115829883B (en) | Surface image denoising method for special-shaped metal structural member | |
CN115272346A (en) | PCB production process online detection method based on edge detection | |
CN115619793B (en) | Power adapter appearance quality detection method based on computer vision | |
CN113570605B (en) | Defect detection method and system based on liquid crystal display panel | |
CN116309537B (en) | Defect detection method for oil stain on surface of tab die | |
CN115359044B (en) | Metal part surface scratch detection method based on image enhancement | |
CN115330767A (en) | Method for identifying production abnormity of corrosion foil | |
CN114782329A (en) | Bearing defect damage degree evaluation method and system based on image processing | |
TWI765442B (en) | Method for defect level determination and computer readable storage medium thereof | |
CN116703909B (en) | Intelligent detection method for production quality of power adapter | |
CN114998325B (en) | Air conditioner radiating tube welding defect detection method | |
CN116363126B (en) | Welding quality detection method for data line USB plug | |
CN115082477B (en) | Semiconductor wafer processing quality detection method based on light reflection removing effect | |
CN116542982A (en) | Departure judgment device defect detection method and device based on machine vision | |
CN107545549B (en) | Method for estimating scattered focus point spread function based on one-dimensional spectrum curve | |
CN111814852A (en) | Image detection method, image detection device, electronic equipment and computer-readable storage medium | |
CN117689655B (en) | Metal button surface defect detection method based on computer vision | |
CN116883408B (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
CN116433700B (en) | Visual positioning method for flange part contour | |
CN116363136B (en) | On-line screening method and system for automatic production of motor vehicle parts | |
CN112396580A (en) | Circular part defect detection method | |
CN115761606A (en) | Box electric energy meter identification method and device based on image processing | |
CN115564705A (en) | Burr detection precision enhancing method and system for precision standard part |
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: 20220902 |