CN112633366A - Jade recognition method based on feature matching - Google Patents
Jade recognition method based on feature matching Download PDFInfo
- Publication number
- CN112633366A CN112633366A CN202011520731.0A CN202011520731A CN112633366A CN 112633366 A CN112633366 A CN 112633366A CN 202011520731 A CN202011520731 A CN 202011520731A CN 112633366 A CN112633366 A CN 112633366A
- Authority
- CN
- China
- Prior art keywords
- jade
- ellipse
- feature
- matching
- disc
- 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.)
- Granted
Links
Images
Classifications
-
- 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
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/757—Matching configurations of points or features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a jade article identification method based on feature matching, which comprises the steps of collecting pictures before and after the jade article is returned by using a detector, and fitting a disc for supporting the jade article based on the sizes of an ellipse center and a major and minor axis to determine the position of the disc; utilizing orb characteristics to extract characteristics of the refunding jade article picture and pictures in a database, positioning characteristic key points in the pictures and calculating a corresponding orb descriptor; selecting parameters to reduce the length of the major axis and the minor axis of the ellipse by utilizing the profile curve of the original ellipse to obtain an ellipse area contained in the disc; whether the matching is successful or not is judged by utilizing the number of the feature points, the identification problem in the returned goods of the jade articles can be solved by adopting the feature extraction and matching based on opencv, the accuracy is high, the speed is high, and the accuracy of the contour extraction is improved; and substituting the coordinate information of the extracted features into an ellipse equation by using a method of reducing the major axis and the minor axis of the fitting ellipse, removing descriptors of feature points falling outside the ellipse, and improving the matching accuracy.
Description
Technical Field
The invention relates to the technical field of jade article identification, in particular to a jade article identification method based on feature matching.
Background
From ancient times to present, jade articles are enjoyed by people in China, and therefore, the consumption of the jade articles gradually develops into a fashion. However, in the current return and exchange process of the jade article, the manual inspection method is still adopted for rechecking the jade article, which not only increases a great amount of labor cost, but also has low inspection efficiency and long time. In order to improve the jade article identification efficiency in the return process, the invention provides a jade article identification method based on feature matching, aiming at quickly detecting whether an uploaded jade article picture is the same as a filed jade article picture before sale, shortening the return time and improving the consumption experience of customers.
The current main jade article identification technology mainly comprises manual auditing, identity jade cards, identity two-dimensional codes and the like. Wherein:
identity jade card identification: the jade plate with the positive surface engraved with the identification marks such as patterns or characters and the like is utilized, and the jade plate has the related information of the jade article;
identity two-dimensional code: by utilizing the identity two-dimensional code picture carried by the jade article, the information of the jade article can be obtained by scanning through a mobile phone or other electronic equipment so as to confirm whether the jade article is the same article sold.
However, the manual inspection of the jade article is mainly affected by the personal skill of the person to be inspected, the inspection equipment and the like, and the inspection result is easily affected by external factors such as human beings and the like.
The main disadvantage of identity-based badges and identity two-dimensional codes is that the certificates are typically hung on or sold with the jades, where there is a possibility that the jades may be exchanged but still use the original identity certificates.
The computer vision related technology has made great development and breakthrough in the field of target re-identification, and jade identification in jade article return is still mainly manual at present, so that the image related technology is applied to jade article identification, the efficiency of return and exchange can be greatly improved, and the influence of human factors on the identification result can be avoided.
Disclosure of Invention
The invention provides a jade article identification method based on feature matching, aiming at the defects of the prior art, and the method utilizes a method for extracting picture features to carry out matching, and judges whether the returned jade article picture is consistent with the pictures in a sample library by comparing and matching the returned jade article picture with the pictures in the sample library to obtain whether the jade article is the originally sold jade article. The method not only overcomes the influence of human factors in the prior art, but also solves the problem that whether the jade plate is the original jade article or not is difficult to judge by the methods of jade plate and two-dimensional code identification and the like, and has high detection speed and high identification accuracy.
The technical scheme of the invention is as follows: a jade article identification method based on feature matching comprises the following steps:
s1), before sale, putting the jade article to be sold into a detector, acquiring the picture of the jade article by using a commodity detector, and then storing the picture in a database for comparison of the characteristics of the jade article returned later;
s2), disc positioning
S201), reading in the image of the returned jade article in an HSV format by using opencv;
s202), fitting a disk for supporting the jade ware based on the sizes of the center and the major and minor axes of the ellipse to determine the position of the disk, and setting the lightness of the area outside the disk to be 0;
s3), feature extraction
S301), extracting features of the refunding jade picture and pictures in the database by using orb features, positioning feature key points in the pictures and calculating corresponding orb descriptors;
s302), selecting corresponding parameters to reduce the length of the major axis and the minor axis of the ellipse by using the contour curve of the original ellipse to obtain an ellipse area contained in the disc;
s303), substituting the coordinate information of the characteristic key points into an ellipse equation:
removing feature descriptors of key points outside the ellipse, and avoiding the influence of edge features of the disc on jade matching;
s4), feature matching
S401), using a violence matcher BFMatcher, taking the features in the refunding jade picture, matching the features with the features of the pictures in a database one by one, and returning the subscripts and the distance of two feature points closest to each feature;
s402), calculating the difference between two feature points closest to the feature points of the original image to judge whether the feature points are successfully matched;
s403), calculating the number of successfully matched feature points in the two pictures, setting a threshold, and if the number of successfully matched feature points is greater than or equal to the threshold, successfully matching the two pictures, namely, returning the jade ware to be sold is consistent;
otherwise, the jade article is inconsistent with the product sold.
Further, in step S1), in the collecting process, the jade is put into a disc of the detector, and pictures of different angles of the front and back surfaces of the jade are collected.
Further, step S1), collecting the returned jade article picture, placing the returned jade article picture into a disc of the detector after receiving the returned jade article, and taking the jade article pictures of different surfaces in the process of rotating the disc.
Further, step S301)), creating an ORB descriptor using the function ORB _ create () of opencv;
then the detectAndComputer method in opencv is used to locate the key points of the picture and calculate their corresponding ORB descriptors.
Further, in step S302), the parameters are selected to be 0.1-0.5, i.e. the length of the major and minor axes of the ellipse is reduced by 0.1-0.5 units each time.
The invention has the beneficial effects that:
1. the method has the advantages that the jade article identification technology based on picture processing is used, so that the labor cost is saved, the identification problem in jade article return and exchange can be solved by adopting the feature extraction and matching based on opencv, and whether manual coring and actual selling are consistent or not is replaced;
2. the method has high identification accuracy and high speed, is different from the defects of the prior identity jade plate and two-dimensional code identification, can directly identify the jade article, can greatly avoid the interference of external factors, simultaneously, the identification process only needs to upload the jade article picture to be identified to a server for feature matching, has high identification speed, can reduce the detection time in the goods returning and changing process, and improves the consumption experience of customers;
3. the method has strong expandability and wide application range of the identification problem in goods return, avoids the influence of the edge characteristics of the disc on the object identification, and only performs characteristic matching identification on the object on the disc, so the method can be expanded to the goods return of a plurality of common objects, such as ornaments like silverware and the like;
4. according to the invention, the size and the position of the disc are fitted by directly utilizing the outline of the ellipse by utilizing the relatively fixed information of the disc, so that the calculated amount is reduced, and the accuracy of outline extraction is improved;
5. the invention uses the method of reducing the long and short axes of the fitting ellipse, substitutes the extracted coordinate information of the characteristics into the ellipse equation, removes the descriptors of the characteristic points outside the ellipse, thereby improving the matching accuracy;
6. according to the method, the manual identification in the jade article returning and changing process is replaced by the operations such as image processing, the position of a disc in a detector is fitted, and then the brightness of an area outside the disc is set to be 0 so as to reduce the influence of other areas on matching; and finally, removing the influence of the edge characteristics of the disc to perform characteristic matching on the objects on the disc to obtain whether the objects are the same object.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention.
Detailed Description
The following further describes embodiments of the present invention with reference to the accompanying drawings:
example 1
As shown in fig. 1, the present embodiment provides a method for identifying a jade article based on feature matching, which includes the following steps:
s1), before sale, putting the jade article to be sold into a detector, acquiring the picture of the jade article by using a commodity detector, and then storing the picture in a database for comparison of the characteristics of the jade article returned later;
s2), disc positioning
S201), reading in the refunding jade picture in an HSV format by using opencv;
s202), fitting a disk for bearing the jade ware based on the sizes of the ellipse center and the major and minor axes to determine the position of the disk, and setting the lightness of the area outside the disk to be 0, namely displaying the area outside the disk to be black, so as to avoid the influence of the internal space features in the detector outside the disk on the subsequent feature extraction; in the embodiment, the size and the position of the disc are fitted by directly utilizing the relatively fixed information of the disc and the outline of the ellipse, so that the calculated amount is reduced, and the accuracy of outline extraction is improved;
s3), feature extraction
S301), extracting features of the refunding jade picture and pictures in the database by using orb features, positioning feature key points in the pictures and calculating corresponding orb descriptors;
s302), by utilizing the contour curve of the original ellipse, the length of the major axis and the minor axis of the ellipse is reduced by a parameter of 0.2 to obtain an ellipse area contained in the disc;
s303), substituting the coordinate information of the characteristic key points into an ellipse equation:
the feature descriptors of key points outside the ellipse are removed, so that the influence of the edge features of the disc on jade matching is avoided, the coordinate information of the extracted features is substituted into an ellipse equation by using a method for reducing the major axis and the minor axis of a fitting ellipse, and the descriptors of the feature points outside the ellipse are removed, so that the matching accuracy is improved;
s4), feature matching
S401), using a violence matcher BFMatcher, taking the features in the refunding jade picture, matching the features with the features of the pictures in a database one by one, and returning the subscripts and the distance of two feature points closest to each feature;
s402), calculating the difference between two feature points closest to the feature points of the original image to judge whether the feature points are successfully matched;
s403), calculating the number of feature points successfully matched in the two pictures, and setting a threshold, for example, the threshold determined in this embodiment is 8,
if the number of the successfully matched feature points is larger than or equal to the threshold value, the two pictures are successfully matched, namely the returned jade ware is consistent with the sold jade ware; otherwise, the jade article is inconsistent with the product sold.
The embodiment utilizes a disc edge feature point removal mode. The central disc of the detector is fitted through an elliptic equation, and the influence of the edge characteristics of the disc on the matching of objects on the disc is removed by using a mode of reducing the major axis and the minor axis of the ellipse, so that the matching precision is improved.
The jade article return and exchange identification process comprises the steps of firstly carrying out ellipse fitting on a disc and carrying out feature extraction, then removing edge features of the disc, and finally determining whether the jade articles are the same article or not in a feature matching mode.
Example 2
The embodiment provides a jade article identification method based on feature matching, which comprises the following steps:
s1), before sale, putting the jade article to be sold into a detector, rotating a disc, and acquiring pictures of the jade article at different angles by using a commodity detector, wherein 6-10 images at different angles can be acquired in the embodiment and then stored in a database for feature comparison of the subsequently returned jade article;
after receiving the degraded jade article, putting the jade article into a commodity detector, shooting 36 pictures at different angles, turning over the jade article, and shooting 36 pictures to obtain 72 pictures;
s2), disc positioning
S201), reading in the returned jade picture in an HSV format by using opencv;
s202), fitting a disk for bearing the jade ware based on the sizes of the ellipse center and the major and minor axes to determine the position of the disk, and setting the lightness of the area outside the disk to be 0, namely displaying the area outside the disk to be black, so as to avoid the influence of the internal space features in the detector outside the disk on the subsequent feature extraction; in the embodiment, the size and the position of the disc are fitted by directly utilizing the relatively fixed information of the disc and the outline of the ellipse, so that the calculated amount is reduced, and the accuracy of outline extraction is improved;
s3), feature extraction
S301), performing feature extraction on the refund jade picture and pictures in the database by using ORB features, locating feature key points in the pictures and calculating corresponding ORB descriptors, in this embodiment, creating ORB descriptors by using a function ORB _ create () of opencv, and then locating the key points of the pictures by using a detectAndComputer method in opencv and calculating their corresponding ORB descriptors.
S302), by utilizing the contour curve of the original ellipse, the length of the major axis and the minor axis of the ellipse is reduced by 0.2 to obtain an elliptical area contained in the disc, in the embodiment, each pair of the major axis and the minor axis of the ellipse is compared once, and the length of the major axis and the minor axis of the ellipse is reduced once, and each time, the reduction is 0.2 unit;
s303), substituting the coordinate information of the characteristic key points into an ellipse equation:
the feature descriptors of key points outside the ellipse are removed, so that the influence of the edge features of the disc on jade matching is avoided, the coordinate information of the extracted features is substituted into an ellipse equation by using a method for reducing the major axis and the minor axis of a fitting ellipse, and the descriptors of the feature points outside the ellipse are removed, so that the matching accuracy is improved;
s4), feature matching
S401), using a violence matcher BFMatcher, taking the features in the refunding jade picture, matching the features with the features of the pictures in a database one by one, and returning the subscripts and the distance of two feature points closest to each feature;
s402), calculating the difference between two feature points closest to the feature points of the original image to judge whether the feature points are successfully matched;
s403), calculating the number of feature points successfully matched in the two pictures, and setting a threshold, for example, the threshold determined in this embodiment is 8,
if the number of the successfully matched feature points is larger than or equal to the threshold value, the two pictures are successfully matched, namely the returned jade ware is consistent with the sold jade ware; otherwise, the jade article is inconsistent with the product sold.
In the embodiment, the same jade article is identified by utilizing an artificial identification mode, and the two types of jade articles are compared to find that:
the method has the advantages that the jade article identification technology based on picture processing is used, so that labor cost is saved, the identification problem in jade article return and exchange can be solved by adopting the feature extraction and matching based on opencv, and whether manual coring and compaction are consistent with the sold jade articles or not is replaced;
the method has high identification accuracy and high speed, is different from the defects of the prior identity jade plate and two-dimensional code identification, can directly identify the jade article, can greatly avoid the interference of external factors, simultaneously, the identification process only needs to upload the jade article picture to be identified to a server for feature matching, has high identification speed, can reduce the detection time in the goods returning and changing process, and improves the consumption experience of customers;
the method has the advantages of strong expandability and wide application range of the identification problem in the goods return and exchange, avoids the influence of the edge characteristics of the disc on the object identification, and only performs characteristic matching identification on the object on the disc. Therefore, the method can be expanded to the return of a plurality of common objects, such as ornaments like silverware and the like.
The foregoing embodiments and description have been presented only to illustrate the principles and preferred embodiments of the invention, and various changes and modifications may be made therein without departing from the spirit and scope of the invention as hereinafter claimed.
Claims (5)
1. A jade article identification method based on feature matching is characterized by comprising the following steps:
s1), before sale, putting the jade article to be sold into a detector, acquiring the picture of the jade article by using a commodity detector, and then storing the picture in a database for comparison of the characteristics of the jade article returned later;
s2), disc positioning
S201), reading in the image of the returned jade article in an HSV format by using opencv;
s202), fitting a disk for supporting the jade ware based on the sizes of the center and the major and minor axes of the ellipse to determine the position of the disk, and setting the lightness of the area outside the disk to be 0;
s3), feature extraction
S301), extracting features of the refunding jade picture and pictures in the database by using orb features, positioning feature key points in the pictures and calculating corresponding orb descriptors;
s302), selecting corresponding parameters to reduce the length of the major axis and the minor axis of the ellipse by using the contour curve of the original ellipse to obtain an ellipse area contained in the disc;
s303), substituting the coordinate information of the characteristic key points into an ellipse equation:
removing feature descriptors of key points outside the ellipse, and avoiding the influence of edge features of the disc on jade matching;
s4), feature matching
S401), using a violence matcher BFMatcher, taking the features in the refunding jade picture, matching the features with the features of the pictures in a database one by one, and returning the subscripts and the distance of two feature points closest to each feature;
s402), calculating the difference between two feature points closest to the feature points of the original image to judge whether the feature points are successfully matched;
s403), calculating the number of successfully matched feature points in the two pictures, setting a threshold, and if the number of successfully matched feature points is greater than or equal to the threshold, successfully matching the two pictures, namely, returning the jade ware to be sold is consistent;
otherwise, the jade article is inconsistent with the product sold.
2. The method for identifying a jade article based on feature matching as claimed in claim 1, wherein: step S1), in the collection process, the jade is put into a disc of the detector, and pictures of the front and back surfaces of the jade at different angles are collected.
3. The method for identifying a jade article based on feature matching as claimed in claim 2, wherein: step S1), the method also comprises the steps of collecting the returned jade ware picture, placing the returned jade ware into a disc of the detector after receiving the returned jade ware, and shooting the jade ware pictures on different surfaces in the process of rotating the disc.
4. The method for identifying a jade article based on feature matching as claimed in claim 1, wherein: step S301)), an ORB descriptor is created using a function ORB _ create () of opencv;
then the detectAndComputer method in opencv is used to locate the key points of the picture and calculate their corresponding ORB descriptors.
5. The method for identifying a jade article based on feature matching as claimed in claim 1, wherein: in step S302), the selected parameter is 0.1-0.5, that is, the length of the major axis and the minor axis of the ellipse is respectively reduced by 0.1-0.5 unit each time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011520731.0A CN112633366B (en) | 2020-12-21 | 2020-12-21 | Jade article identification method based on feature matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011520731.0A CN112633366B (en) | 2020-12-21 | 2020-12-21 | Jade article identification method based on feature matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112633366A true CN112633366A (en) | 2021-04-09 |
| CN112633366B CN112633366B (en) | 2023-05-12 |
Family
ID=75320370
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011520731.0A Active CN112633366B (en) | 2020-12-21 | 2020-12-21 | Jade article identification method based on feature matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112633366B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113409148A (en) * | 2021-07-21 | 2021-09-17 | 苏州维格纳信息科技有限公司 | Transaction information security maintenance system based on block chain |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109635818A (en) * | 2018-10-26 | 2019-04-16 | 平安科技（深圳）有限公司 | The anti-counterfeit of seals method of inspection, device and computer readable storage medium |
| CN109724990A (en) * | 2019-01-08 | 2019-05-07 | 上海大学 | The quick positioning and detection method in coding region in a kind of packing box label |
| CN110490211A (en) * | 2019-08-23 | 2019-11-22 | 郑州航空工业管理学院 | A kind of button identification method based on feature extraction |
| CN112084365A (en) * | 2020-09-11 | 2020-12-15 | 上海幻维数码创意科技有限公司 | Real-time image retrieval method of network camera based on OpenCV and CUDA acceleration |
-
2020
- 2020-12-21 CN CN202011520731.0A patent/CN112633366B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109635818A (en) * | 2018-10-26 | 2019-04-16 | 平安科技（深圳）有限公司 | The anti-counterfeit of seals method of inspection, device and computer readable storage medium |
| CN109724990A (en) * | 2019-01-08 | 2019-05-07 | 上海大学 | The quick positioning and detection method in coding region in a kind of packing box label |
| CN110490211A (en) * | 2019-08-23 | 2019-11-22 | 郑州航空工业管理学院 | A kind of button identification method based on feature extraction |
| CN112084365A (en) * | 2020-09-11 | 2020-12-15 | 上海幻维数码创意科技有限公司 | Real-time image retrieval method of network camera based on OpenCV and CUDA acceleration |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113409148A (en) * | 2021-07-21 | 2021-09-17 | 苏州维格纳信息科技有限公司 | Transaction information security maintenance system based on block chain |
| CN113409148B (en) * | 2021-07-21 | 2022-03-04 | 南京理工大学 | Transaction information security maintenance system based on block chain |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112633366B (en) | 2023-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108921939A (en) | A kind of method for reconstructing three-dimensional scene based on picture | |
| CN107808120A (en) | Glasses localization method, device and storage medium | |
| CN106327534B (en) | A kind of tire inner wall Texture Recognition based on locating piece | |
| CN111445459A (en) | Image defect detection method and system based on depth twin network | |
| CN105243376A (en) | Living body detection method and device | |
| CN101131728A (en) | Face shape matching method based on Shape Context | |
| CN107609489A (en) | Calligraphy writing path evaluation device, method and electronic equipment | |
| CN102324042A (en) | Visual identifying system and visual identity method | |
| CN110263662B (en) | Human body contour key point and key part identification method based on grading | |
| CN110717885A (en) | Customer number counting method and device, electronic equipment and readable storage medium | |
| CN103488966A (en) | Intelligent mobile phone capable of identifying real-name ticket information | |
| CN112633366B (en) | Jade article identification method based on feature matching | |
| CN113454649B (en) | Target detection method, apparatus, electronic device, and computer-readable storage medium | |
| CN111191708A (en) | Automatic sample key point marking method, device and system | |
| CN102163343B (en) | Three-dimensional model optimal viewpoint automatic obtaining method based on internet image | |
| CN111507177B (en) | Identification method and device for metering turnover cabinet | |
| CN112258540A (en) | Image corner detection method based on nonlinear direction derivative | |
| CN109254663B (en) | Using method of auxiliary reading robot for books of children | |
| CN106156774A (en) | Image processing method and image processing system | |
| CN111415378A (en) | Image registration method for automobile glass detection and automobile glass detection method | |
| CN115019152A (en) | Image shooting integrity judgment method and device | |
| CN111382703B (en) | Finger vein recognition method based on secondary screening and score fusion | |
| CN109472279B (en) | Article identification method and device based on image processing | |
| CN114255398A (en) | Method and device for extracting and matching features of satellite video image | |
| CN115994997A (en) | Secondhand luxury trading platform, system and method with authentication function |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |