CN108152217B - Method for identifying writing and drawing tips - Google Patents
Method for identifying writing and drawing tips Download PDFInfo
- Publication number
- CN108152217B CN108152217B CN201711314760.XA CN201711314760A CN108152217B CN 108152217 B CN108152217 B CN 108152217B CN 201711314760 A CN201711314760 A CN 201711314760A CN 108152217 B CN108152217 B CN 108152217B
- Authority
- CN
- China
- Prior art keywords
- painting
- calligraphy
- hyperspectral
- detected
- original
- 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.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Processing Or Creating Images (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a method for identifying the tip of a painting and calligraphy pen. The method comprises the following steps: step 1, hyperspectral imaging; step 2, preprocessing a hyperspectral image; step 3, performing hyperspectral image fusion transformation; step 4, extracting and measuring the stroke features; step 5, distinguishing the similarity of the stroke features; the invention mainly solves the problem that the existing pen point analysis and identification by using human eyes or visible light images is difficult to counterfeit and identify; the method has the advantages that the hyperspectral imaging technology and the atlas analysis technology are used for obtaining and modeling the stroke point information, and the objective identification of the stroke point of the painting and calligraphy is completed in a more efficient, comprehensive and accurate mode by starting from two aspects of space and spectrum instead of single space information, so that the identification of the whole painting and calligraphy work is better assisted. The method is mainly applied to the identification of the penpoint penmanship of calligraphy, Chinese wash painting, Chinese color painting and foreign oil painting.
Description
Technical Field
The invention relates to the field of painting and calligraphy appraisal and evaluation, which is mainly applied to the field of painting and calligraphy transaction circulation, in particular to a hyperspectral appraisal method for a painting and calligraphy tip.
Background
The painting and calligraphy identification means that the authenticity and the artistic value of the works are reasonably and correctly judged while a painting and calligraphy artwork is appreciated. Has very important significance for painting and calligraphy preservation, research, transaction circulation and the like. The method is characterized in that the authenticity judgment is first, and the quality judgment is second. In the identification method, the basic information of writing and painting such as pen (tip), ink or color (pigment), and stamp method (composition) can be used for analysis and judgment by utilizing abundant writing and painting works and writer background knowledge.
The specific method for identifying the painting and calligraphy mainly comprises two aspects: first, the eye identification (expert identification), and second, the machine identification (computer professional means identification). Currently, the current methods are mainly based on the current identification and assisted by the machine identification. In the aspect of visual identification, the painting and calligraphy identification experts mainly achieve the purposes of identifying true and false, clear and non, breaking time and evaluating value through the analysis of artistic characteristics such as teachers' source, pen and ink characteristics, artistic features, time style and the like. In the aspect of machine identification, the characteristics of the work are mainly analyzed and processed by means of computers, chemistry, photoelectricity and the like, and are compared with objective history and prior characteristics of the work, if the characteristics are consistent with the historical objective characteristics of the author or the work, the characteristics are true, and if the characteristics are not consistent with the historical objective characteristics of the author or the work, the characteristics may be false. Because most of the objective historical prior information used by the current machine identification means is unknown and incomplete, the machine identification method is only used as a painting and calligraphy identification method which is assisted by expert identification.
As an important calligraphy and painting feature, the expression of calligraphy and painting (stroke) is very important for the value expression of calligraphy and painting. The object images of painting and calligraphy, such as figures and texture, are mostly embodied by different calligraphy. Moreover, the calligraphy also includes the mind and interest of the painter when creating the painting and calligraphy to express a certain artistic conception. Regardless of the writing and calligraphy, the traditional painting and calligraphy always has some basic techniques, such as hook and hook, a clockwise line, an anticlockwise line, a single pen as hook, a double pen as hook, a left line as hook, and a right line as hook. In addition, the wrinkle method can show the shape, texture and features of landscape painting.
Furthermore, calligraphy is also the core basis of painting and calligraphy authentication, because a painting and calligraphy work is always created by the use of pen and ink, no matter what subject or mood the work represents. The personal style is different from the time interest, and finally the difference is achieved on the writing method of the works. In addition, the density distribution of strokes of each character and each graph in the painting and calligraphy is often within millicentimetres and within thousands of miles, so that the analysis of the painting and calligraphy (stroke) is of great importance to the painting and calligraphy identification.
The traditional pen-based analysis method based on expert identification or full-color camera acquisition mainly utilizes information of pen ink in a visible light wave band to carry out analysis, and human eyes or a full-color camera acquire integral accumulation information of visible light information without spectrum resolution capability, so that the analysis of the pen-based analysis method can only reflect spatial information of a pen-based method, and the authenticity of the spatial information is difficult to judge.
Disclosure of Invention
In order to make up for the defects of the prior art, the invention provides a hyperspectral identification method of a writing and painting tip.
The invention adopts the following technical scheme:
1. a method for identifying the tip of a painting and calligraphy is characterized by comprising the following steps:
step 1, hyperspectral imaging
Performing hyperspectral imaging on the original painting and the painting to be detected in or after circulation by using a hyperspectral imager to obtain hyperspectral imaging data of the painting and the calligraphy;
step 2, hyperspectral image preprocessing
Carrying out registration, correction and calibration on the acquired original painting and calligraphy hyperspectral image data and the painting and calligraphy hyperspectral image data to be detected to obtain preprocessed painting and calligraphy hyperspectral image data;
step 3, hyperspectral image fusion transformation
Performing hyperspectral image fusion transformation on the original preprocessed painting and calligraphy hyperspectral image data and the painting and calligraphy hyperspectral image data to be preprocessed to obtain a painting and calligraphy hyperspectral fusion transformation image;
step 4, extracting and measuring the stroke edge characteristics
Extracting and measuring typical stroke points of the original painting and calligraphy hyperspectral fusion transformation image and the painting and calligraphy hyperspectral fusion transformation image to be detected to obtain the characteristics of the typical stroke points of the painting and calligraphy;
And comparing the stroke features of the original and to-be-detected painting and calligraphy data, and completing stroke identification by calculating the similarity discrimination value of the original and to-be-detected painting and calligraphy data. If the similarity is higher than the threshold value, the point of the painting and calligraphy to be detected is the same as that of the original painting and calligraphy, and the painting and calligraphy are the same; if the similarity is lower than the threshold value, the point of the painting and calligraphy to be detected is different from that of the original painting and calligraphy, and the painting and calligraphy are not the same.
As a further scheme of the invention, the hyperspectral imaging is to obtain visible light-near infrared (450 plus 2500nm) hyperspectral imaging data of the original painting and calligraphy to be detected under the same environment, including illumination, distance, angle and the same imaging instrument parameters.
As a further scheme of the invention, the hyperspectral image preprocessing refers to registration, correction and calibration processing of the original painting and calligraphy spectral image and the painting and calligraphy spectral image to be detected, wherein the registration mainly completes the spatial position alignment of the original painting and calligraphy and painting to be detected, the correction mainly completes the spatial deformation correction of painting and calligraphy data, and the calibration mainly completes the calibration of painting and calligraphy data spectral information.
As a further scheme of the invention, the hyperspectral image fusion transformation refers to the spectral dimensional mathematical transformation of the original painting and calligraphy hyperspectral image and the painting and calligraphy hyperspectral image to be detected, and the method comprises but is not limited to principal component transformation and independent component transformation.
As a further scheme of the invention, the stroke feature extraction and measurement refers to extraction and measurement of hooks, chaps and the like, and specifically refers to completing matching search and feature extraction of the four typical stroke models in the fusion image by constructing horizontal, vertical, left-falling and right-falling models and performing windowed search matching.
As a further scheme of the invention, the distinguishing of the similarity of the writing and drawing tip features refers to calculating a distinguishing value of the correlation between the original painting and the typical tip features of the painting and drawing to be detected by using a spectrum similarity angle and a correlation function, and judging the distinguishing value by using a threshold value, wherein if the distinguishing value is higher than the threshold value, the writing and drawing to be detected and the original painting and drawing tip features are the same and are the same painting and drawing; if the discrimination value is lower than the threshold value, the brush points of the painting and calligraphy to be detected are different from those of the original painting and calligraphy, and the painting and calligraphy are not the same.
The invention has the beneficial effects that: the problem that forgery is difficult to identify when the existing pen tip analysis and identification is carried out by using human eyes or visible light images is solved; the method has the advantages that the hyperspectral imaging technology and the atlas analysis technology are used for obtaining and modeling the stroke point information, and the objective identification of the stroke point of the painting and calligraphy is completed in a more efficient, comprehensive and accurate mode by starting from two aspects of space and spectrum instead of single space information, so that the identification of the whole painting and calligraphy work is better assisted. The method is mainly applied to the identification of the penpoint penmanship of calligraphy, Chinese wash painting, Chinese color painting and foreign oil painting.
Drawings
FIG. 1 is a hyperspectral imaging painting and calligraphy stroke identification process according to the invention.
Detailed Description
The invention is explained in further detail below with reference to the figures and the specific embodiments.
A method for identifying the tip of a painting and calligraphy pen comprises the following steps:
step 1, hyperspectral imaging
Performing hyperspectral imaging on the original painting and the painting to be detected in or after circulation by using a hyperspectral imager, and acquiring visible light-near infrared (450 plus 2500nm) hyperspectral imaging data of the original painting and the painting to be detected in the same environment under the conditions of illumination (halogen lamps), distance (2 meters), angle (right angle) and the same imaging instrument parameters (3nm spectral resolution, 30nm focal length, 1000 x 1000 pixels and the like);
step 2, hyperspectral image preprocessing
And carrying out registration, correction and calibration on the acquired original painting and calligraphy hyperspectral image data and the painting and calligraphy hyperspectral image data to be detected to obtain preprocessed painting and calligraphy hyperspectral image data. Wherein, the registration mainly completes the spatial position alignment of the original painting and calligraphy and the painting to be detected, and adopts a center point threshold characteristic point alignment scheme; the correction mainly completes the space deformation correction of the painting and calligraphy data, namely the field distortion correction and the image scaling treatment; the calibration mainly completes the calibration of the spectrum of the painting and calligraphy data, including white balance and the like;
step 3, hyperspectral image fusion transformation
And performing spectral dimensional hyperspectral image fusion transformation on the original preprocessed painting and calligraphy hyperspectral image data and the painting and calligraphy hyperspectral image data to be preprocessed to obtain a painting and calligraphy hyperspectral fusion transformation image. The principal component transformation is adopted, and a first principal component image is selected as a fusion transformation image;
step 4, extracting and measuring the stroke edge characteristics
And extracting and measuring typical stroke points of the original painting and calligraphy hyperspectral fusion transformation image and the painting and calligraphy hyperspectral fusion transformation image to be detected to obtain the characteristics of the typical stroke points of the painting and calligraphy. Specifically, matching search and feature extraction of the four typical stroke models in the fused image are completed by constructing horizontal, vertical, left-falling and right-falling models and performing windowed search matching, so that matching degree vectors of the four stroke models are obtained;
And comparing the stroke features (matching degree vectors) of the original and to-be-detected painting and calligraphy data, and completing stroke identification by calculating the similarity discrimination value of the original and to-be-detected painting and calligraphy data. Selecting a spectrum similarity angle as a discrimination value calculation function, and selecting a fixed threshold as a threshold processing mode. If the discrimination value is higher than the threshold value, the point of the painting and calligraphy to be detected is the same as that of the original painting and calligraphy, and the painting and calligraphy are the same; if the discrimination value is lower than the threshold value, the brush points of the painting and calligraphy to be detected are different from those of the original painting and calligraphy, and the painting and calligraphy are not the same.
The foregoing is a preferred embodiment of the present invention, and it will be apparent to those skilled in the art that variations, modifications, substitutions and alterations can be made in the embodiment without departing from the principles and spirit of the invention.
Claims (1)
1. A method for identifying the tip of a painting and calligraphy is characterized by comprising the following steps:
step 1, hyperspectral imaging
Performing hyperspectral imaging on the original painting and the painting to be detected in or after circulation by using a hyperspectral imager to obtain hyperspectral imaging data of the painting and the calligraphy;
step 2, hyperspectral image preprocessing
Carrying out registration, correction and calibration on the acquired original painting and calligraphy hyperspectral image data and the painting and calligraphy hyperspectral image data to be detected to obtain preprocessed painting and calligraphy hyperspectral image data;
step 3, hyperspectral image fusion transformation
Performing hyperspectral image fusion transformation on the original preprocessed painting and calligraphy hyperspectral image data and the painting and calligraphy hyperspectral image data to be preprocessed to obtain a painting and calligraphy hyperspectral fusion transformation image;
step 4, extracting and measuring the stroke edge characteristics
Extracting and measuring typical stroke points of the original painting and calligraphy hyperspectral fusion transformation image and the painting and calligraphy hyperspectral fusion transformation image to be detected to obtain the characteristics of the typical stroke points of the painting and calligraphy;
step 5, distinguishing the similarity of the features of the pen nib
Comparing the stroke characteristics of the original and to-be-detected painting and calligraphy data, and completing stroke identification by calculating the similarity discrimination value of the original and to-be-detected painting and calligraphy data, wherein if the similarity is higher than a threshold value, the to-be-detected painting and calligraphy are the same as the original painting and calligraphy; if the similarity is lower than the threshold value, the point of the painting and calligraphy to be detected is different from that of the original painting and calligraphy, and the painting and calligraphy are not the same;
the hyperspectral imaging of the step 1 is to obtain visible light-near infrared 450-plus-2500 nm hyperspectral imaging data of an original painting and calligraphy to be detected under the same environment, including illumination, distance, angle and the same imaging instrument parameters;
the step 2 of hyperspectral image preprocessing refers to registering, correcting and calibrating the original painting and calligraphy spectral image and the painting and calligraphy spectral image to be detected, wherein the registering mainly completes the spatial position alignment of the original painting and calligraphy and the painting and calligraphy to be detected, the correcting mainly completes the spatial deformation correction of painting and calligraphy data, and the calibrating mainly completes the calibration of painting and calligraphy data spectral information;
the step 3 of hyperspectral image fusion transformation refers to the step of performing spectral dimensional transformation on the original painting and calligraphy hyperspectral image and the painting and calligraphy hyperspectral image to be detected, wherein the spectral dimensional transformation comprises principal component transformation and independent component transformation;
the stroke feature extraction and measurement in the step 4 refers to extraction and measurement of hooking, tying and chapping waves, and specifically refers to completing matching search and feature extraction of four typical stroke models in the fused image by constructing horizontal, vertical, left-falling and right-falling models and performing windowed search matching;
the judgment of the similarity of the brush point characteristics in the step 5 is to calculate a correlation judgment value between the original painting and the typical brush point characteristics of the painting to be detected by using a spectrum similarity angle or a correlation function, and to judge a threshold value of the correlation judgment value, wherein the judgment value is higher than the threshold value, so that the brush points of the painting to be detected and the original painting are the same and are the same painting; if the discrimination value is lower than the threshold value, the brush points of the painting and calligraphy to be detected are different from those of the original painting and calligraphy, and the painting and calligraphy are not the same.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711314760.XA CN108152217B (en) | 2017-12-12 | 2017-12-12 | Method for identifying writing and drawing tips |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711314760.XA CN108152217B (en) | 2017-12-12 | 2017-12-12 | Method for identifying writing and drawing tips |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108152217A CN108152217A (en) | 2018-06-12 |
| CN108152217B true CN108152217B (en) | 2020-07-31 |
Family
ID=62466942
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711314760.XA Expired - Fee Related CN108152217B (en) | 2017-12-12 | 2017-12-12 | Method for identifying writing and drawing tips |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108152217B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109543757A (en) * | 2018-11-27 | 2019-03-29 | 陕西文投艺术品光谱科技有限公司 | A kind of painting and calligraphy painting style identification method based on spectral imaging technology and atlas analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102496180A (en) * | 2011-12-15 | 2012-06-13 | 李大锦 | Method for automatically generating wash landscape painting image |
| CN103164699A (en) * | 2013-04-09 | 2013-06-19 | 北京盛世融宝国际艺术品投资有限公司 | Painting and calligraphy work fidelity identification system |
| CN104134150A (en) * | 2014-07-30 | 2014-11-05 | 刘思岩 | Computer assisted Chinese art paper painting and calligraphy identification method |
| CN204255501U (en) * | 2014-10-29 | 2015-04-08 | 北京帝测科技股份有限公司 | A kind of for the Hyperspectral imaging devices that ancient character is drawn and cultural and artistic products is differentiated |
| CN106338487A (en) * | 2016-09-29 | 2017-01-18 | 北京建筑大学 | Calligraphy and painting identification method and device |
-
2017
- 2017-12-12 CN CN201711314760.XA patent/CN108152217B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102496180A (en) * | 2011-12-15 | 2012-06-13 | 李大锦 | Method for automatically generating wash landscape painting image |
| CN103164699A (en) * | 2013-04-09 | 2013-06-19 | 北京盛世融宝国际艺术品投资有限公司 | Painting and calligraphy work fidelity identification system |
| CN104134150A (en) * | 2014-07-30 | 2014-11-05 | 刘思岩 | Computer assisted Chinese art paper painting and calligraphy identification method |
| CN204255501U (en) * | 2014-10-29 | 2015-04-08 | 北京帝测科技股份有限公司 | A kind of for the Hyperspectral imaging devices that ancient character is drawn and cultural and artistic products is differentiated |
| CN106338487A (en) * | 2016-09-29 | 2017-01-18 | 北京建筑大学 | Calligraphy and painting identification method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108152217A (en) | 2018-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110210448B (en) | Intelligent face skin aging degree identification and evaluation method | |
| CN113592845A (en) | Defect detection method and device for battery coating and storage medium | |
| CN105913093A (en) | Template matching method for character recognizing and processing | |
| CN104298992B (en) | A kind of adaptive scale pedestrian recognition methods again based on data-driven | |
| JP2016541052A (en) | Image analysis to certify products | |
| CN106203329B (en) | A method of identity template is established based on eyebrow and carries out identification | |
| CN101533466B (en) | Image processing method for positioning eyes | |
| CN110136130A (en) | A kind of method and device of testing product defect | |
| CN111127417B (en) | Printing defect detection method based on SIFT feature matching and SSD algorithm improvement | |
| CN104766344B (en) | Vehicle checking method based on movement edge extractor | |
| CN112381840B (en) | Method and system for marking vehicle appearance parts in loss assessment video | |
| JP2012527665A (en) | Apparatus and method for identifying the original author of a work of art | |
| CN109507198A (en) | Mask detection system and method based on Fast Fourier Transform (FFT) and linear Gauss | |
| CN115375991A (en) | Strong/weak illumination and fog environment self-adaptive target detection method | |
| CN109472257B (en) | Character layout determining method and device | |
| JP4003465B2 (en) | Specific pattern recognition method, specific pattern recognition program, specific pattern recognition program recording medium, and specific pattern recognition apparatus | |
| Li et al. | Sublingual vein extraction algorithm based on hyperspectral tongue imaging technology | |
| CN108152217B (en) | Method for identifying writing and drawing tips | |
| Zhao et al. | MSC-AD: A Multiscene Unsupervised Anomaly Detection Dataset for Small Defect Detection of Casting Surface | |
| CN111339974B (en) | Method for identifying modern ceramics and ancient ceramics | |
| CN108154116A (en) | A kind of image-recognizing method and system | |
| JP6311461B2 (en) | Gaze analysis system and gaze analysis apparatus | |
| CN109724921B (en) | Painting and calligraphy identification map feature representation method based on spectral imaging technology | |
| CN111046883B (en) | Intelligent assessment method and system based on ancient coin image | |
| Jung et al. | A Deep Learning Approach Based on Image Patch Sets for Art Forgery Detection |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200731 Termination date: 20201212 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |