CN108898597B - Method for identifying painting and calligraphy based on smart phone - Google Patents
Method for identifying painting and calligraphy based on smart phone Download PDFInfo
- Publication number
- CN108898597B CN108898597B CN201810692348.XA CN201810692348A CN108898597B CN 108898597 B CN108898597 B CN 108898597B CN 201810692348 A CN201810692348 A CN 201810692348A CN 108898597 B CN108898597 B CN 108898597B
- Authority
- CN
- China
- Prior art keywords
- image
- matching
- painting
- points
- calligraphy
- 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.)
- Active
Links
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Collating Specific Patterns (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method for identifying painting and calligraphy based on a smart phone, which comprises two steps of filing and authentication; firstly, a plurality of detail textures which can be used for matching are positioned in a painting and calligraphy sample by an image feature detection method, and then consistency comparison of the textures is realized under the limited magnification of a mobile phone camera, so that consistency of the painting and calligraphy image and the sample in a database is judged. The invention can identify the painting and calligraphy based on the smart phone, has lower application cost and simple flow operation, can meet the identification requirements of common consumer groups at any time and any place, has three-step matching operation and has very high identification result accuracy, thereby having higher market popularization and application value.
Description
Technical Field
The invention relates to a method for identifying calligraphy and painting certificates, in particular to a method for identifying calligraphy and painting certificates based on a smart phone.
Background
The painting and calligraphy authentication is a way for identifying painting and calligraphy works in folk. Most of the existing identification methods are manual identification methods, but if the meanings of the calligraphy and painting and the classification of the painting are not understood, the identification is not carried out. In order to clear the obstacle, people usually need to understand the knowledge, otherwise, the authenticity of the painting cannot be identified, the price is not accurate, and the basis is lost during appreciation.
American and Dutch researchers have studied the difference characteristic of the effects of painting and brush stroke with the help of the computer algorithm, have developed a set of AI artificial intelligence systems that can discern the painting true and false, this system can discern whether the painting is a counterfeit article with the help of the simple effect of brush stroke, although easy to operate, the rate of accuracy is higher, but also there is the defect that the fabrication cost is high, the sexual valence is lower too, not suitable for ordinary consumer groups. Therefore, the development of a novel intelligent painting and calligraphy authentication method is still a hot problem for people to focus on research.
Disclosure of Invention
In order to solve the defects of the technology, the invention provides a method for identifying the painting and calligraphy based on a smart phone.
In order to solve the technical problems, the invention adopts the technical scheme that: a method for identifying calligraphy and painting based on a smart phone comprises two steps of recording and authenticating;
firstly, recording:
a. shooting a painting and calligraphy sample image to enable the shot breadth to cover most of the area of the painting and calligraphy sample;
b. extracting feature points on the graph and obtaining respective feature attribute description of the feature points;
c. screening the characteristic points, and selecting more than 4 stable characteristic points more uniformly;
d. then selecting a plurality of salient feature points which are easy to distinguish by people;
e. manually selecting one of the remarkable feature points, moving the mobile phone to the feature point, amplifying the feature point after the mobile phone moves to the height specified by the picture, turning on a flash lamp, then shooting an image, and focusing and shooting a plurality of sequence images by a mobile camera;
f. automatically selecting the clearest image from the sequence images, extracting the feature points and the attribute description of the image again according to the method, and storing the feature points and the attribute description;
g. selecting a block region with rich detail texture of small blocks from a painting and calligraphy sample image as a matching candidate point, storing the characteristic image after characteristic transformation such as gradient transformation, and marking the offset position and size of the small blocks;
h. repeating the processes of the steps e to g, and selecting a plurality of different micro characteristic matching candidate points;
i. storing the characteristic points and the attribute description, and the small block offset position and size into a set file directory, and recording the file directory as template image data;
II, authentication:
a. shooting a painting and calligraphy sample image to be authenticated, and enabling the shot breadth to cover most of the painting and calligraphy image;
b. extracting feature points on the graph and obtaining respective feature attribute description of the feature points; screening the characteristic points, and selecting more than 4 stable characteristic points more uniformly;
c. matching the stable characteristic points with the characteristic points in the template and performing consistency comparison between the characteristic points; if the matching success point number is smaller than the specified proportion, the matching is considered to be failed, and the authentication is quitted; if the matching is not successful, calculating the position, angle and scale deviation of the current sample image and the template image, and continuing the next step;
d. displaying the matched candidate points of the shot magnified image on the template image or the current sample image, manually selecting one of the candidate points, moving the camera to be close to the candidate point, turning on a flash lamp after the candidate point is magnified, then shooting the image, and moving the camera to focus on a plurality of shooting sequence images;
e. automatically selecting the clearest image from the sequence images, extracting the feature points and the attribute description of the image again according to the method, and storing the feature points and the attribute description;
f. matching the feature points in the last step with the feature points in the template again, and if the matching is successful, calculating the position, angle and scale deviation of the current sample image and the template; if the matching fails, the point is changed again and the operation is repeated;
g. according to the matching result, image correction is carried out on the position, angle and scale deviation of the current sample image and the template to obtain the position of a matching candidate block, corresponding to a matching candidate point in the template, in the current sample image, and the characteristic image is obtained after characteristic transformation, such as gradient transformation, is carried out on the current sample image as the template image;
h. moving the matching candidate blocks up, down, left and right by a small amount, calculating the correlation coefficient between the characteristic images, and observing whether the correlation coefficient exceeds a set threshold value; if yes, matching is successful; otherwise, the matching fails;
i. and selecting the matching points according to the reliability requirement, if the successfully matched points are larger than a set proportion threshold, considering that the consistency check is successful, and otherwise, considering that the consistency check is failed.
Further, the feature points extracted in the first step and the second step include SIFT points, corner points, and SURF points.
Furthermore, the candidate position for shooting the amplified image is displayed in the wide template image or the current sample image, so that the user can conveniently select the shooting position of the amplified image.
Further, the specific way of automatically selecting the clearest image in the first step and the second step is as follows: the definition of the image is determined by the gradient change or the high-frequency component of the image.
The invention obtains the accurate positioning of the detail texture at a certain position in the painting and calligraphy works through two times of feature point matching, and then obtains the final work consistency authentication by calculating the correlation coefficient between feature image blocks.
The invention can identify the painting and calligraphy based on the smart phone, has lower application cost and simple flow operation, can meet the identification requirements of common consumer groups at any time and any place, has three-step matching operation and has very high identification result accuracy, thereby having higher market popularization and application value.
Detailed Description
The present invention will be described in further detail with reference to specific embodiments.
A method for identifying calligraphy and painting based on a smart phone comprises two steps of recording and authenticating;
firstly, recording:
a. shooting a painting and calligraphy sample image to enable the shot breadth to cover most of the area of the painting and calligraphy sample;
b. extracting feature points on the graph, such as SIFT points, corner points, SURF points and the like, and obtaining respective feature attribute description;
c. screening the characteristic points, and selecting more than 4 stable characteristic points more uniformly; the selection of stable characteristic points needs at least 4, but can be hundreds or more in order to improve the precision, so that more characteristic points can be selected as much as possible if possible, and the characteristic points and the characteristic description thereof are stored;
d. then selecting a plurality of salient feature points which are easy to distinguish by people;
e. manually selecting one of the remarkable feature points, moving the mobile phone to the feature point, amplifying the feature point after the mobile phone moves to the height specified by the picture, turning on a flash lamp, then shooting an image, and focusing and shooting a plurality of sequence images by a mobile camera;
f. judging the definition of the image through the gradient change of the image or the high-frequency component, automatically selecting the clearest image from the sequence images, extracting the feature points (such as SIFT and the like) of the image and attribute description thereof again according to the method, and storing the feature points and attribute description;
g. selecting a block region with rich detail texture of small blocks from a painting and calligraphy sample image as a matching candidate point, storing the characteristic image after characteristic transformation such as gradient transformation, and marking the offset position and size of the small blocks;
h. repeating the processes of the steps e to g, and selecting a plurality of different micro characteristic matching candidate points;
i. storing the characteristic points and the attribute description, and the small block offset position and size into a set file directory, and recording the file directory as template image data;
II, authentication:
a. shooting a painting and calligraphy sample image to be authenticated, and enabling the shot breadth to cover most of the painting and calligraphy image;
b. extracting feature points on the graph and obtaining respective feature attribute description of the feature points; screening the characteristic points, and selecting more than 4 stable characteristic points more uniformly; if possible, more feature points can be selected as much as possible, and the feature points and feature descriptions thereof are stored;
c. matching the stable characteristic points with the characteristic points in the template and performing consistency comparison between the characteristic points; if the matching success point number is smaller than the specified proportion, the matching is considered to be failed, and the authentication is quitted; if the matching is not successful, calculating the position, angle and scale deviation of the current sample image and the template image, and continuing the next step;
d. displaying matching candidate points of the shot magnified image on the template image or the current sample image (the candidate positions of the shot magnified image can be displayed in the wide template image or the current sample image, so that a user can conveniently select the shooting position of the magnified image), manually selecting one of the candidate points (one of the candidate points can be selected arbitrarily), moving the camera to be close to the candidate point, turning on a flash lamp after the candidate point is magnified, then shooting the image, and moving the camera to focus on a plurality of shooting sequence images;
e. automatically selecting the clearest image from the sequence images, extracting the feature points and the attribute description of the image again according to the method, and storing the feature points and the attribute description;
f. matching the feature points in the last step with the feature points in the template again, and if the matching is successful, calculating the position, angle and scale deviation of the current sample image and the template; if the matching fails, the point is changed again and the operation is repeated;
g. according to the matching result, image correction is carried out on the position, angle and scale deviation of the current sample image and the template to obtain the position of a matching candidate block, corresponding to a matching candidate point in the template, in the current sample image, and the characteristic image is obtained after characteristic transformation, such as gradient transformation, is carried out on the current sample image as the template image;
h. moving the matching candidate blocks up, down, left and right by a small amount, calculating the correlation coefficient between the characteristic images, and observing whether the correlation coefficient exceeds a set threshold value; if yes, matching is successful; otherwise, the matching fails;
i. and selecting the matching points according to the reliability requirement, if the successfully matched points are larger than a set proportion threshold, considering that the consistency check is successful, and otherwise, considering that the consistency check is failed.
The method positions a plurality of detail textures which can be used for matching in the painting and calligraphy sample through the detection method of the image characteristics, and realizes consistency comparison of the textures under the limited magnification factor of the mobile phone camera, thereby judging the consistency of the painting and calligraphy image and the sample in the database.
The invention adopts three steps to record and authenticate the painting and calligraphy, namely the extraction and matching of the characteristics of a wide image, the extraction and matching of the characteristics of an amplified image and the related matching of the image or the characteristic image.
The above embodiments are not intended to limit the present invention, and the present invention is not limited to the above examples, and those skilled in the art may make variations, modifications, additions or substitutions within the technical scope of the present invention.
Claims (4)
1. A method for identifying calligraphy and painting certificates based on a smart phone is characterized by comprising the following steps: the method comprises two steps of filing and authenticating;
firstly, recording:
a. shooting a painting and calligraphy sample image to enable the shot breadth to cover most of the painting and calligraphy sample;
b. extracting feature points on the graph and obtaining respective feature attribute description of the feature points;
c. screening the characteristic points, and selecting more than 4 stable characteristic points more uniformly;
d. then selecting a plurality of salient feature points which are easy to distinguish;
e. manually selecting one of the salient feature points, moving the mobile phone to the feature point, amplifying the mobile phone to a height specified by a picture, turning on a flash lamp, then shooting an image, and focusing and shooting a plurality of sequence images by a mobile camera to obtain a detailed image;
f. extracting feature points again from the detail image, obtaining respective feature attribute description of the feature points, and storing the feature point positions and the attribute description of the feature points;
g. selecting a small block area in the detail image as a matching position, performing gradient transformation on the small block image at the position to form a gradient image, storing the gradient image, and marking the offset position and size of the small block area relative to the detail image;
h. repeating the processes of the steps e to g, and selecting a plurality of different micro characteristic matching candidate points;
i. storing characteristic points and attribute description of shot painting and calligraphy sample images and detailed images and offset positions and sizes of gradient images into a set file directory, and recording the characteristic points and attribute description and the offset positions and the sizes as template image data;
II, authentication:
a. shooting a painting and calligraphy sample image to be authenticated, and enabling the shot breadth to cover most of the area of the painting and calligraphy sample image to obtain a wide image of the painting and calligraphy work to be authenticated;
b. extracting feature points on the wide image and obtaining respective feature attribute description of the feature points;
c. matching the characteristic points on the wide image with the characteristic points in the template image and performing matching between the characteristic points; if the matching success point number is smaller than the specified proportion threshold value, the matching is considered to be failed, the matching is considered not to belong to the same work, and the authentication is quit; if the matching is not successful, calculating mapping parameters including position, angle and scale deviation between the wide image and the template image, and continuing the next step;
d. calling a template image, and displaying the position of the shot detail image in the template image; manually selecting one of the works, moving the camera to the position, turning on a flash lamp, and shooting a detail clear image of the work to be authenticated after amplification;
e. automatically selecting the clearest image from the detail clear images in the step d, and shooting other detail images in the work to be authenticated again according to the method in the authentication step d;
f. matching the characteristic points of the detail image in the template image with the characteristic points of the detail clear image of the authentication work, and if the matching is successful, calculating the mapping relation parameters between the template image and the detail clear image of the work to be authenticated; if the matching fails, the authentication steps b to e are repeated by replacing points again;
g. according to the matching result in the step f, carrying out image correction on the position, angle and scale deviation of the current sample image and the template to obtain the position of a matching candidate block, which corresponds to a matching candidate point in the template and is positioned in the current sample image, and carrying out feature transformation on the current sample image to obtain a feature image as same as the template image;
h. moving the matching candidate blocks up, down, left and right by a small amount, calculating the correlation coefficient between the characteristic images, and observing whether the correlation coefficient exceeds a set threshold value; if yes, matching is successful; otherwise, the matching fails;
i. if the matching of most of small images is successful, the small images are regarded as the same work, and the authentication is successful; otherwise, the work is regarded as different works, and the authentication fails.
2. The smart phone based calligraphy and painting certification method according to claim 1, wherein: the feature points extracted in the first step and the second step comprise SIFT points, corner points and SURF points.
3. The smart phone based calligraphy and painting certification method according to claim 2, wherein: and displaying the candidate position for shooting the amplified image in the wide template image or the current sample image, so that a user can conveniently select the shooting position of the amplified image.
4. The smart phone based calligraphy and painting certification method according to claim 3, wherein: the specific mode for automatically selecting the clearest image in the first step and the second step is as follows: the definition of the image is determined by the gradient change or the high-frequency component of the image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810692348.XA CN108898597B (en) | 2018-06-28 | 2018-06-28 | Method for identifying painting and calligraphy based on smart phone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810692348.XA CN108898597B (en) | 2018-06-28 | 2018-06-28 | Method for identifying painting and calligraphy based on smart phone |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108898597A CN108898597A (en) | 2018-11-27 |
| CN108898597B true CN108898597B (en) | 2021-10-15 |
Family
ID=64347085
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810692348.XA Active CN108898597B (en) | 2018-06-28 | 2018-06-28 | Method for identifying painting and calligraphy based on smart phone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108898597B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110428404B (en) * | 2019-07-25 | 2021-11-23 | 北京邮电大学 | Artificial intelligence-based auxiliary culture and auxiliary appreciation formulation system |
| CN110909750B (en) * | 2019-11-14 | 2022-08-19 | 展讯通信（上海）有限公司 | Image difference detection method and device, storage medium and terminal |
| CH717252A2 (en) * | 2020-03-23 | 2021-09-30 | 4Art Holding Ag | Process for the recognition of surfaces. |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103116755A (en) * | 2013-01-27 | 2013-05-22 | 深圳市书圣艺术品防伪鉴定有限公司 | Automatic painting and calligraphy authenticity degree detecting system and method thereof |
| CN105323208A (en) * | 2014-06-07 | 2016-02-10 | 李季 | Method for tracing calligraphy and painting identity tag |
| CN105654216A (en) * | 2014-11-15 | 2016-06-08 | 北京翰易得科技有限公司 | Calligraphy and painting management and selling system based on intelligent handset platform |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8963805B2 (en) * | 2012-01-27 | 2015-02-24 | Microsoft Corporation | Executable virtual objects associated with real objects |
-
2018
- 2018-06-28 CN CN201810692348.XA patent/CN108898597B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103116755A (en) * | 2013-01-27 | 2013-05-22 | 深圳市书圣艺术品防伪鉴定有限公司 | Automatic painting and calligraphy authenticity degree detecting system and method thereof |
| CN105323208A (en) * | 2014-06-07 | 2016-02-10 | 李季 | Method for tracing calligraphy and painting identity tag |
| CN105654216A (en) * | 2014-11-15 | 2016-06-08 | 北京翰易得科技有限公司 | Calligraphy and painting management and selling system based on intelligent handset platform |
Non-Patent Citations (2)
| Title |
|---|
| Increment of Antenna Efficiency of the Mobile Phone Antenna by EMI Slot;TAEHO, SON 等;《The Journal Of Korean Institute of Electromagnetic Engineering and Science》;20081231;第19卷(第9期);第1037-1044页 * |
| 新媒体技术在博物馆中的技术应用与开发设计研究;许聪;《中国优秀硕士学位论文全文数据库 信息科技辑》;20170215(第2期);I144-75 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108898597A (en) | 2018-11-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110334585B (en) | Table identification method, apparatus, computer device and storage medium | |
| CN107358596B (en) | Vehicle loss assessment method and device based on image, electronic equipment and system | |
| CN108898597B (en) | Method for identifying painting and calligraphy based on smart phone | |
| CN108288012B (en) | Artwork filing verification method and system based on mobile phone | |
| CN105809651B (en) | Image significance detection method based on the comparison of edge non-similarity | |
| CN110427746A (en) | Sliding block verifies code verification method, device, storage medium and computer equipment | |
| CN106169080A (en) | A kind of combustion gas index automatic identifying method based on image | |
| CN105184225B (en) | A kind of multinational banknote image recognition methods and device | |
| CN108921160B (en) | Book identification method, electronic equipment and storage medium | |
| CN116822623B (en) | Method, device, equipment and storage medium for generating countermeasures network joint training | |
| CN110826571B (en) | Image traversal algorithm for rapid image identification and feature matching | |
| CN109784379B (en) | Updating method and device of textile picture feature library | |
| CN107133260A (en) | The matching and recognition method and device of a kind of landmark image | |
| CN114235815A (en) | Method for detecting surface defects of outdoor electrical equipment of converter station based on scene filtering | |
| CN112734747B (en) | Target detection method and device, electronic equipment and storage medium | |
| CN110084269A (en) | Characteristics of image optimization method based on cloud computing | |
| CN105608423A (en) | Video matching method and device | |
| CN108898059A (en) | Flowers recognition methods and its equipment | |
| CN115880683B (en) | Urban waterlogging ponding intelligent water level detection method based on deep learning | |
| CN110031471B (en) | Method, system and device for analyzing surface defect growth of large-caliber optical element | |
| CN109657083B (en) | Method and device for establishing textile picture feature library | |
| CN111627059A (en) | Method for positioning center point position of cotton blade | |
| CN108629733B (en) | Method and device for acquiring high-resolution image | |
| CN105930813B (en) | A method of detection composes a piece of writing this under any natural scene | |
| CN115760695A (en) | Image anomaly identification method based on depth vision model |
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 |