CN110286139B - Method for distinguishing big data composite characteristics of paint film of ancient lacquerware - Google Patents
Method for distinguishing big data composite characteristics of paint film of ancient lacquerware Download PDFInfo
- Publication number
- CN110286139B CN110286139B CN201910480897.5A CN201910480897A CN110286139B CN 110286139 B CN110286139 B CN 110286139B CN 201910480897 A CN201910480897 A CN 201910480897A CN 110286139 B CN110286139 B CN 110286139B
- Authority
- CN
- China
- Prior art keywords
- paint film
- sample
- data
- image
- calculating
- 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
- 239000003973 paint Substances 0.000 title claims abstract description 175
- 238000000034 method Methods 0.000 title claims abstract description 64
- 239000002131 composite material Substances 0.000 title claims abstract description 15
- 239000004922 lacquer Substances 0.000 claims abstract description 66
- 238000005033 Fourier transform infrared spectroscopy Methods 0.000 claims abstract description 36
- 230000008569 process Effects 0.000 claims abstract description 27
- 238000001069 Raman spectroscopy Methods 0.000 claims abstract description 17
- 238000000605 extraction Methods 0.000 claims abstract description 12
- 238000003909 pattern recognition Methods 0.000 claims abstract description 8
- 230000002159 abnormal effect Effects 0.000 claims abstract description 7
- 239000000126 substance Substances 0.000 claims abstract description 7
- 238000001228 spectrum Methods 0.000 claims abstract description 4
- 238000002441 X-ray diffraction Methods 0.000 claims description 33
- 238000001514 detection method Methods 0.000 claims description 33
- 239000013598 vector Substances 0.000 claims description 33
- 239000011159 matrix material Substances 0.000 claims description 28
- 230000002776 aggregation Effects 0.000 claims description 15
- 238000004220 aggregation Methods 0.000 claims description 15
- 238000005259 measurement Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 9
- 238000013139 quantization Methods 0.000 claims description 8
- 238000010183 spectrum analysis Methods 0.000 claims description 7
- 239000006185 dispersion Substances 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 238000012512 characterization method Methods 0.000 claims description 4
- RMTXUPIIESNLPW-UHFFFAOYSA-N 1,2-dihydroxy-3-(pentadeca-8,11-dienyl)benzene Natural products CCCC=CCC=CCCCCCCCC1=CC=CC(O)=C1O RMTXUPIIESNLPW-UHFFFAOYSA-N 0.000 claims description 3
- QARRXYBJLBIVAK-UEMSJJPVSA-N 3-[(8e,11e)-pentadeca-8,11-dienyl]benzene-1,2-diol;3-[(8e,11e)-pentadeca-8,11,14-trienyl]benzene-1,2-diol;3-[(8e,11e,13e)-pentadeca-8,11,13-trienyl]benzene-1,2-diol;3-[(e)-pentadec-8-enyl]benzene-1,2-diol;3-pentadecylbenzene-1,2-diol Chemical compound CCCCCCCCCCCCCCCC1=CC=CC(O)=C1O.CCCCCC\C=C\CCCCCCCC1=CC=CC(O)=C1O.CCC\C=C\C\C=C\CCCCCCCC1=CC=CC(O)=C1O.C\C=C\C=C\C\C=C\CCCCCCCC1=CC=CC(O)=C1O.OC1=CC=CC(CCCCCCC\C=C\C\C=C\CC=C)=C1O QARRXYBJLBIVAK-UEMSJJPVSA-N 0.000 claims description 3
- IYROWZYPEIMDDN-UHFFFAOYSA-N 3-n-pentadec-8,11,13-trienyl catechol Natural products CC=CC=CCC=CCCCCCCCC1=CC=CC(O)=C1O IYROWZYPEIMDDN-UHFFFAOYSA-N 0.000 claims description 3
- 238000010521 absorption reaction Methods 0.000 claims description 3
- 230000002146 bilateral effect Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 3
- 229910052802 copper Inorganic materials 0.000 claims description 3
- RTZKZFJDLAIYFH-UHFFFAOYSA-N ether Chemical group CCOCC RTZKZFJDLAIYFH-UHFFFAOYSA-N 0.000 claims description 3
- 125000002887 hydroxy group Chemical group [H]O* 0.000 claims description 3
- 238000003706 image smoothing Methods 0.000 claims description 3
- DQTMTQZSOJMZSF-UHFFFAOYSA-N urushiol Natural products CCCCCCCCCCCCCCCC1=CC=CC(O)=C1O DQTMTQZSOJMZSF-UHFFFAOYSA-N 0.000 claims description 3
- 238000012850 discrimination method Methods 0.000 claims description 2
- 239000000523 sample Substances 0.000 description 91
- 238000001237 Raman spectrum Methods 0.000 description 22
- 238000004458 analytical method Methods 0.000 description 12
- 238000004422 calculation algorithm Methods 0.000 description 9
- 230000008901 benefit Effects 0.000 description 4
- 238000006116 polymerization reaction Methods 0.000 description 4
- 238000007689 inspection Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 241000220442 Cajanus Species 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000006698 induction Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000000634 powder X-ray diffraction Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 239000013074 reference sample Substances 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 206010067484 Adverse reaction Diseases 0.000 description 1
- 230000006838 adverse reaction Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002329 infrared spectrum Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000007591 painting process Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
-
- 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
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/65—Raman scattering
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
- G01N23/20—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by using diffraction of the radiation by the materials, e.g. for investigating crystal structure; by using scattering of the radiation by the materials, e.g. for investigating non-crystalline materials; by using reflection of the radiation by the materials
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
- G01N23/20—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by using diffraction of the radiation by the materials, e.g. for investigating crystal structure; by using scattering of the radiation by the materials, e.g. for investigating non-crystalline materials; by using reflection of the radiation by the materials
- G01N23/20091—Measuring the energy-dispersion spectrum [EDS] of diffracted radiation
-
- 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
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N2021/3595—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using FTIR
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2223/00—Investigating materials by wave or particle radiation
- G01N2223/05—Investigating materials by wave or particle radiation by diffraction, scatter or reflection
- G01N2223/056—Investigating materials by wave or particle radiation by diffraction, scatter or reflection diffraction
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2223/00—Investigating materials by wave or particle radiation
- G01N2223/10—Different kinds of radiation or particles
- G01N2223/101—Different kinds of radiation or particles electromagnetic radiation
- G01N2223/1016—X-ray
Abstract
The invention discloses a distinguishing method based on big data composite characteristics of a paint film of an antique paint device, which comprises the following steps: performing data characteristic extraction on a paint film certified sample of the existing ancient lacquerware, wherein the data characteristic extraction comprises chemical component characteristics, microscopic image characteristics, scanning electron microscope X-ray energy spectrum characteristics, Fourier transform infrared spectrometer characteristics and Raman spectrometer characteristics; establishing a sample library and a pattern recognition classifier of the ancient lacquer paint film according to the extracted data characteristics, and respectively performing label calibration on the existing ancient lacquer paint film genuine product samples; and acquiring data characteristics of the unknown lacquerware lacquer film to be detected and identified, comparing the data characteristics with parameters in the sample library to acquire an optimal matching result, if the matching is successful, determining the type, process and year prediction result of the sample to be detected, and otherwise, prompting abnormal information by the classifier to finish the identification of the lacquerware lacquer film. The invention provides a scientific method for the true and false deduction, the year identification and the process identification of cultural relics of the lacquer film of the lacquerware.
Description
Technical Field
The invention relates to the field of cultural relic discrimination, chemical composition and spectral analysis and image pattern recognition, in particular to a method for discriminating big data composite characteristics of a paint film of an antique paint device.
Background
With the application of modern detection instruments and big data and pattern recognition technologies, scientific detection and discrimination by using modern instrument detection means and combining a computer big data processing method in the fields of cultural relic component analysis, age, process identification and authenticity identification become research hotspots. The cultural relic identification and discrimination method based on pattern recognition can obtain scientific and accurate decision results, and has a plurality of advantages compared with manual identification.
From the current research and application, the analysis of the chemical components of the lacquer paint film by using the instrument means such as X-ray detection, Raman spectrum analysis, Fourier infrared spectrum detection and the like is a more common detection method, has been widely applied in the field of single-piece ancient lacquer paint film analysis, but the defects are also more remarkable and mainly expressed as follows:
1. the method adopts an instrument for analysis, focuses on analyzing the internal composition of the paint film of the lacquerware, generally detects single samples one by one, and lacks of induction storage and system analysis of batch detection data;
2. the component content of instrument detection data is judged, and compared with a standard substance spectral image, although accurate judgment can be made within an error tolerance range, the actual components of a paint film of a real paint device often have various elements mixed, and the detected data often cannot reflect the real component and content of the material;
3. generally, the process judgment, the age judgment and the classification judgment of paint films of lacquerwares depend on observation and experience, and data support is lacked, so that the judgment result is not objective enough.
Disclosure of Invention
The invention aims to provide a scientific and effective method for accurately classifying paint films of ancient lacquerware and providing a decision-making method.
The method for distinguishing the big data composite characteristics of the paint film based on the antique paint device comprises the following steps:
performing data characteristic extraction on a paint film certified sample of the existing ancient lacquerware, wherein the data characteristic extraction comprises chemical component characteristics, microscopic image characteristics, scanning electron microscope X-ray energy spectrum characteristics, Fourier transform infrared spectrometer characteristics and Raman spectrometer characteristics;
establishing a sample library and a pattern recognition classifier of the ancient lacquer paint film according to the extracted data characteristics, and respectively performing label calibration on the existing ancient lacquer paint film genuine product samples;
and acquiring data characteristics of the unknown lacquerware lacquer film to be detected and identified, comparing the data characteristics with parameters in the sample library to acquire an optimal matching result, if the matching is successful, determining the type, process and year prediction result of the sample to be detected, and otherwise, prompting abnormal information by the classifier to finish the identification of the lacquerware lacquer film.
According to the technical scheme, the data feature extraction method specifically comprises the following steps:
s1, respectively obtaining microscopic images of the front surface, the back surface and the cross section of the paint film sample, and establishing microscopic characterization parameters; carrying out image smoothing on the microscopic image of the paint film, filtering image noise, and eliminating the influence of tiny change between adjacent pixel points of the paint film image on the image;
s2, calculating color aggregation vector characteristics of the microscopic image;
s3, carrying out color space quantization on the microscopic image of the paint film, and respectively quantizing RGB color components of the image aiming at the microscopic image of the color paint film;
s4, carrying out gray level processing on the microscopic image of the paint film, and calculating a gray level co-occurrence matrix of image gray level pixels after normalization;
s5, calculating the EDS characteristics of the paint film according to the X-ray energy dispersion spectrum analysis data of the paint film sample;
s6, calculating the XRD characteristics of the paint film sample according to the back X-ray diffraction analysis data of the paint film sample;
s7, calculating the RS characteristic of the paint film sample according to the laser confocal Raman spectrometer data of the paint film sample;
and S8, calculating the FTIR characteristics of the paint film sample according to the Fourier transform infrared spectrometer data of the paint film sample.
In the above technical solution, step S1 includes:
s11, filtering the microscopic image of the paint film by using a bilateral filter, filtering the noise containing motes and shadows on the surface, and outputting a smooth preprocessed image;
and S12, extracting the edge and the texture line characteristics of the image.
In the above technical solution, the step S2 includes:
s21, extracting color aggregation vectors from the smoothed image, classifying pixel points into aggregation and non-aggregation according to the image communication region, and writing the pixel points into label marks respectively;
and S22, calculating a color aggregation vector for the aggregation area, and storing the feature marks.
In the above technical solution, step S3 includes:
s31, performing 10-level quantization processing on the color image of the paint film according to the color channel components;
and S32, normalizing and storing the quantized component values into a one-dimensional feature vector.
In the above technical solution, step S4 includes:
s41, dividing the paint film image into pixel blocks with the size of a 3 x 3 pixel neighborhood from left to right and from top to bottom in sequence, and sliding in 2 x 2 sub-blocks to obtain 4 sub-neighborhood blocks, wherein each sub-neighborhood comprises 4 pixels, and each sub-neighborhood takes the middle pixel point of the original 3 x 3 pixel neighborhood as a reference pixel;
s42, scanning 4 pixel points in the sub-neighborhood by taking the upper left corner point as a base point, wherein 7 scanning paths are shared;
s43, calculating a co-occurrence matrix of the sub-neighborhood images by 7 scanning paths, wherein the pixel distance is 1, and the included angles between the paths and the transverse axis are 0 degree, 45 degrees, 90 degrees and 135 degrees respectively;
and S44, calculating the sum of squares of the gray level co-occurrence matrix element values according to the co-occurrence matrix to obtain the energy attribute.
S45, calculating the difference degree between the primitive pixel points and the field pixel points, and constructing the contrast attribute of the definition of the image and the depth degree of the texture groove;
s45, calculating and measuring the similarity of the space gray level co-occurrence matrix elements in the row or column direction, and obtaining the correlation attribute of the joint probability of the occurrence of the designated pixel pair;
and S46, calculating the roughness of the image texture to obtain the uniformity attribute.
S47, calculating the entropy of the sub-neighborhood;
and S48, storing the parameters of contrast, correlation, entropy, evenness and energy statistics as feature vectors.
In the above technical solution, step S5 includes:
s51, acquiring X-ray energy dispersion spectrum analysis data of the paint film sample, and sequentially extracting the contents of C, O, Hg, S, Al, Si, Fe, Ca, Cu, K, Cl, Cr, Mg and Na elements;
s52, storing the content of each element as a one-dimensional vector, and adding a feature tag;
in the above technical solution, step S6 includes:
s61, acquiring XRD characteristics of the paint film sample according to the back X-ray diffraction analysis data of the paint film sample;
s62, calculating HgS and SiO in XRD characteristics of paint film2The peak position of (d);
s62, storing the peak positions of the corresponding elements of the standard X-ray diffraction standard card PDF;
and S63, calculating and matching the peak position of the sample with the peak position of the PDF element, and acquiring the XRD element component in the sample.
In the above technical solution, step S7 includes:
s71, obtaining the RS characteristics of the sample according to the laser confocal Raman spectrometer data of the paint film sample;
and S72, respectively referring to the laser Raman characteristic peak parameters of HgS and C, matching with the RS characteristic peak of the sample, and deducing whether the sample contains the element.
10. The method for discriminating the composite characteristic based on the big data of the paint film of the antique paint device as claimed in claim 2, wherein the step S8 comprises the following steps:
s81, acquiring FTIR characteristics of the paint film sample according to the data of the Fourier transform infrared spectrometer;
s82, comparing infrared absorption peaks of hydroxyl, ether bond and urushiol with the FTIR characteristics, and determining whether the component is contained therein.
According to the technical scheme, when the sample library is established, the method specifically comprises the following steps:
storing the extracted data characteristics as a characteristic space database, and adding the category, lacquerware process, age, region and number parameters of the lacquer film sample into a data label to be used as a standard template library of the characteristic data;
performing network distributed storage on the standard template library, and storing the images of the front, the cross section and the back of the microscope of the sample and the images of the EDS, XRD, RS and FTIR characteristic curves in corresponding paths;
providing a detection input interface through a network server;
the method specifically comprises the following steps of comparing the data characteristics of the unknown lacquerware lacquer film to be detected and identified with the parameters in the sample library:
inputting microscopic images, EDS, XRD, RS and FTIR detection data of the unknown lacquer film to be detected and identified through a network interface;
calculating the microscopic image co-occurrence matrix, matching with the parameters of each known sample in the template library to obtain the similarity sigma of the matching resulti；
EDS, XRD, RS and FTIR detection data of the paint film of the unknown paint container are matched with characteristic parameters of corresponding samples in the standard template library to obtain matching similarity ki、λi、μi、νi；
For similarity sigmaiAnd matching similarity ki、λi、μi、νiWeighting and sorting;
outputting a matching result, and setting a threshold value as a category, process and age prediction result of a sample to be tested of a paint film of the lacquerware; and if the set threshold value is exceeded, the classifier prompts abnormal information to finish the identification of the lacquer film of the lacquer.
The invention has the following beneficial effects: the invention collects the data of the existing instruments by using a big data processing method, so that the current processing mode of the detection data aiming at a single paint film sample of the ancient lacquerware is converted into a data clustering method for classifying the lacquerware paint films with the same process, components, year and category, and the matching and the identification of the sample lacquerware paint film to be detected are realized by using a characteristic template matching algorithm.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flowchart of an overall algorithm according to an embodiment of the present invention;
FIG. 2 is a co-occurrence matrix feature extraction algorithm used by the present invention.
FIG. 3 is a similarity metric structure for a combination of features used by the present invention.
Fig. 4 is an example of a microscopic image of a paint film of an ancient painter (a paint-coated wooden box in the middle and late war nations with mounds unearthed) used in the present invention.
Fig. 5(1) is an example of the front inspection data of EDS sample (tuser, warring) red paint film of ancient lacquer film used in the present invention.
FIG. 5(2) an example of the data for the front inspection of the EDS sample of ancient lacquerware paint film (Chase, warring) black paint film used in the present invention.
FIG. 6(1) is an example of X-ray powder diffraction analysis test data for the back of paint film from XRD samples of ancient lacquer paint films (Cajanus javanica, late in the war) used in the present invention.
Fig. 6(2) is an example of cross-sectional mapping data for an ancient lacquer XRD sample (jatroda, late middle warring) used in the present invention.
Fig. 7(1) is an example of raman data for a black paint film from an ancient lacquer RS sample (lacquer wooden box, late middle warring) used in the present invention.
Fig. 7(2) is an example of raman detection data for an ancient lacquer RS sample (lacquer wooden box, late middle warring) of a red lacquer film used in the present invention.
Fig. 7(3) is an example of raman data for a yellow paint film from an ancient lacquer RS sample (lacquer wooden box, late middle warring) used in the present invention.
FIG. 8(1) is an exemplary FTIR sample (ear cup, late in the warring) of a ancient lacquer film used in the present invention.
FIG. 8(2) is an exemplary FTIR sample (ear cup, late in the warring) black paint film test data for an ancient lacquer film used in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The invention provides an ancient lacquer ware paint film process identification method based on modern instrument data clustering, which is used for analyzing modern instrument data of an ancient unearthed lacquer paint film, extracting the modern instrument detection data of an ancient lacquer ware paint film sample, extracting the multi-dimensional characteristics of the ancient lacquer ware paint film sample, matching the characteristics of the lacquer painting process, the times and the like of the ancient lacquer ware paint film by using a machine learning algorithm, establishing a detection data characteristic database of the ancient lacquer ware paint film by big data machine learning, and providing a scientific method for performing true and false deduction, chronological identification and process identification of the ancient lacquer ware paint film.
The method comprises the steps of firstly, detecting and obtaining collected data of a sample by using an instrument, carrying out feature extraction on the collected data, establishing a training sample library, carrying out feature clustering and machine learning on feature vector data of a paint film of a sample lacquering device by using a pattern recognition clustering method on the basis of a training set, and establishing a reference sample library. And for the newly sampled sample to be detected, pattern matching is carried out on the sample, the matching result of the sample in the training feature library is judged, and the authenticity identification of the cultural relic is realized by comparing the identification result.
The main innovation of the invention is as follows: 1. and carrying out big data storage and processing on originally isolated single sample data, and carrying out induction classification on detection results of samples of the same type and the same process and age to obtain comprehensive characteristic parameters of the samples. 2. Considering that the appearance microscopic data of the paint film of the lacquerware is the visual observation result of identification and process analysis, a symbiotic matrix is introduced into the characteristic parameters to describe the texture appearance of the microscopic image, so that the identified pattern library is richer and more accurate; 3. on the basis of data of sample detection, a pattern sample library of paint films of the lacquerware is established, approximate clustering of samples of the same type is realized by adding sample labels, and a decision-making basis is provided for accurate classification of a sample to be detected.
The method comprises the steps of firstly, storing big data of paint film data of the antique paint device, and extracting EDS (scanning electron microscope energy spectrometer) data characteristics of the paint film data, wherein the EDS data characteristics comprise chemical composition characteristics, microscopic image characteristics, scanning electron microscope X-ray energy spectrum characteristics, Fourier transform infrared spectrometer characteristics and Raman spectrometer characteristics.
Specifically, the feature extraction mainly comprises the following steps:
s1, extracting microscopic characterization features, respectively acquiring front, back and cross-section microscopic images of a paint film sample, and establishing characterization parameters of the front, back and cross-section microscopic images; performing image smoothing on the paint film image, filtering image noise, and eliminating the influence of small change between adjacent pixel points of the paint film image on the image;
s2, calculating color polymerization vector characteristics of the microscopic paint film image;
and S3, performing color space quantization on the paint film image, and performing color quantization on the color space according to the 10-level quantization interval. The RGB color components of the image are quantized separately for the color paint film image.
S4, performing gray scale processing on the paint film image, and calculating a gray scale co-occurrence matrix of image gray scale pixels after normalization;
s5, calculating the EDS characteristics of the paint film according to the X-ray energy dispersion spectrum analysis data of the paint film sample; as shown in fig. 5(1), 5(2), an exemplary graph of the front inspection data of the used ancient lacquer film EDS sample (chan, warring) red lacquer film and black lacquer film.
S6, calculating XRD characteristics according to the back X-ray diffraction analysis data of the paint film sample, as shown in figure 6(1), which is an example of the back X-ray powder diffraction analysis detection data of the paint film XRD sample (Cajanus arguta, late middle warring) of the ancient lacquer paint film used by the invention. Fig. 6(2) is an example of cross-sectional mapping data of an ancient lacquer XRD sample (jatroda, late middle warring).
S7, calculating the RS characteristic of the paint film sample according to the laser confocal Raman spectrometer data of the paint film sample; as shown in fig. 7(1), 7(2), examples of raman detection data of the ancient lacquer RS samples (lacquer wooden box, late middle warring) of black and red lacquer films used in the present invention are shown.
S8, calculating the FTIR characteristics of the paint film sample according to the Fourier transform infrared spectrometer data of the paint film sample; as shown in fig. 8(1), 8(2), there are examples of the test data for the ancient lacquerware paint film FTIR samples (ear cup, late middle warring) red paint film and black paint film used in the present invention.
Secondly, after the extracted data features, the following steps are carried out:
and S9, establishing a pattern recognition classifier of the ancient lacquer paint film by integrating the characteristic parameters of the S1-S9, and respectively performing label calibration on the existing ancient lacquer paint film genuine product samples.
S10, comparing the characteristic parameters of S1-S10 with the parameters in the sample library respectively to obtain the most matching result, if the matching is successful, determining the type, process and age prediction result of the sample to be detected, otherwise, prompting abnormal information by a classifier, and finishing the identification of the lacquer film of the lacquer.
Wherein step S1 includes:
s11, filtering the paint film microscopic image by using a bilateral filter, filtering noises such as motes and shadows on the surface, and outputting a smooth preprocessed image;
and S12, extracting the edge and the texture line characteristics of the image.
Step S2 includes:
s21, extracting color polymerization vectors from the smoothed paint film image, classifying pixel points into polymerization and non-polymerization according to an image communication area, and writing the pixel points into label marks respectively;
and S22, calculating a color aggregation vector for the aggregation area, and storing the feature marks.
Step S3 includes:
s31, performing 10-level quantization processing on the paint film color image according to the color channel components;
s32, normalizing and storing the quantized component values of the paint film image into one-dimensional characteristic vectors;
step S4 includes:
s41, dividing the paint film image into pixel blocks with the size of a 3 x 3 pixel neighborhood from left to right and from top to bottom in sequence, and sliding in 2 x 2 sub-blocks to obtain 4 sub-neighborhood blocks, wherein each sub-neighborhood comprises 4 pixels, and each sub-neighborhood takes the middle pixel point of the original 3 x 3 pixel neighborhood as a reference pixel;
s42, scanning 4 pixel points in the sub-neighborhood by taking the upper left corner point as a base point, wherein 7 scanning paths are shared;
s43, calculating a co-occurrence matrix of the sub-neighborhood images by 7 scanning paths, wherein the pixel distance is 1, and the included angles between the paths and the transverse axis are 0 degree, 45 degrees, 90 degrees and 135 degrees respectively;
and S44, calculating the sum of squares of the gray level co-occurrence matrix element values according to the co-occurrence matrix to obtain the energy attribute.
S45, calculating the difference degree between the primitive pixel points and the field pixel points, and constructing the contrast attribute of the definition of the image and the depth degree of the texture groove;
s45, calculating and measuring the similarity of the space gray level co-occurrence matrix elements in the row or column direction, and obtaining the correlation attribute of the joint probability of the occurrence of the designated pixel pair;
and S46, calculating the roughness of the image texture to obtain the uniformity attribute.
S47, calculating the entropy of the sub-neighborhood;
s48, storing the contrast, correlation, entropy, uniformity and energy statistical parameters as feature vectors;
step S5 includes:
s51, acquiring X-ray energy dispersion spectrum analysis data of the paint film sample, and sequentially extracting the contents of C, O, Hg, S, Al, Si, Fe, Ca, Cu, K, Cl, Cr, Mg and Na in the EDS characteristics;
s52, storing the content of each element as a one-dimensional vector, and adding a feature tag;
step S6 includes:
s61, acquiring XRD characteristics of the paint film sample according to the back X-ray diffraction analysis data of the paint film sample;
s62, calculating HgS and SiO in XRD characteristics of paint film2The peak position of (d);
s62, storing the peak positions of the corresponding elements of the standard X-ray diffraction standard card PDF;
and S63, calculating and matching the peak position of the sample with the peak position of the PDF element, and acquiring the XRD element component in the sample.
Step S7 includes:
s71, obtaining the RS characteristics of the sample according to the laser confocal Raman spectrometer data of the paint film sample;
and S72, respectively referring to the laser Raman characteristic peak parameters of HgS and C, matching with the RS characteristic peak of the sample, and deducing whether the sample contains the element.
Step S8 includes:
s81, acquiring FTIR characteristics of the paint film sample according to the data of the Fourier transform infrared spectrometer;
s82, comparing infrared absorption peaks of hydroxyl, ether bond and urushiol with the FTIR characteristics, and determining whether the component is contained therein.
Step S9 includes:
s91, for the paint film of the sample antique paint device, respectively adopting a calculation method of S1-S9 to obtain characteristic parameters of the paint film, storing the characteristic parameters as a characteristic space database, and adding the type, the process, the year, the region and the number parameters of the paint film sample in a data label as a standard template library of the characteristic data;
s92, performing network distributed storage on the standard template library, and storing the microscopic front, cross section and back images of the sample and EDS, XRD, RS and FTIR characteristic curve images in corresponding paths;
and S93, providing a detection input interface through the network server, and reading the microscopic image, EDS, XRD, RS and FTIR detection data of the paint film of the position lacquerware to be detected.
Step S10 includes:
s101, inputting microscopic images, EDS, XRD, RS and FTIR detection data of the unknown lacquer film to be detected and identified through a network interface;
s102, calculating the microscopic image co-occurrence matrix, matching the microscopic image co-occurrence matrix with the parameters of all known samples in a template library, and obtaining the similarity sigma of the matching resulti；
S103, matching EDS, XRD, RS and FTIR detection data of the paint film of the unknown lacquerware with characteristic parameters of corresponding samples in a standard template library to obtain matching similarity ki、λi、μi、νi；
S104, degree of similarity sigmai、κi、λi、μi、νiWeighting and sorting;
s105, outputting a matching result, and setting a threshold value as a category, process and age prediction result of a sample to be tested of a paint film of the lacquerware; and if the set threshold value is exceeded, the classifier prompts abnormal information to finish the identification of the lacquer film of the lacquer.
Calculating a pixel symbiotic matrix of a microscopic image of a paint film sample of the lacquerware to obtain a texture characteristic vector, and specifically comprising the following steps:
calculating angular second order distances on microscopic images
Calculating angular second order contrast on microscopic images
Computing angular second order distance correlation on microscopic images
Calculating angular second-order variance on microscopic image
Calculating angular second-order inverse variance on microscopic image
Calculating an angular second order distance average on a microscopic image
Calculating angular second-order variance on microscopic image
Computing angular second order entropy on microscopic images
Computing angular second order entropy on microscopic images
Calculating angular second-order variance on microscopic image
f10＝px-y
Computing an angular second-order difference entropy on a microscopic image
Computing angular second order distance correlation information measure on microscopic image
The gray level co-occurrence matrix is calculated on the microscopic image by adopting the following calculation steps:
(1) second moment of angle (energy)
(2) Contrast ratio (moment of inertia)
(3) Entropy of the entropy
(4) Moment of adverse reaction (local stationarity)
Calculating the gray level co-occurrence matrix characteristics:
the flow of the feature extraction algorithm of the gray level co-occurrence matrix is shown in fig. 2.
Firstly, dividing the brightness of an image into L gray levels with the size of R rows and C columns, then respectively calculating 5 texture characteristic quantities of a second moment, entropy, contrast, inverse difference moment and sum variance of 4 direction co-occurrence matrixes, and finally forming texture characteristic vectors by using the 5 characteristic quantities.
When the EDS (Energy Dispersive Spectrometer), XRD (X-ray diffraction), RS (Raman spectrum), FTIR (Fourier Transform Infrared Spectrometer) sample detection data are stored as the feature vectors, the method specifically includes:
the EDS data is subjected to normalization processing and then stored as a one-dimensional feature vector;
normalizing the XRD data, and storing the normalized XRD data as a one-dimensional characteristic vector;
the RS data is subjected to normalization processing and then stored as a one-dimensional feature vector;
the FTIR data is normalized and then stored as a one-dimensional feature vector.
Then, carrying out mode storage on the texture co-occurrence matrix of the microscopic image of the sample and one-dimensional vectors of EDS, XRD, RS and FTIR sample detection data; writing corresponding sample numbers, age and process marks in the characteristic label field of the database, and outputting the corresponding sample numbers, age and process marks as matching results of the samples;
the method for distinguishing big data composite characteristics of the paint film of the antique paint device, disclosed by the embodiment of the invention, as shown in figure 1, comprises the following steps:
initializing the model parameters, including loading the detection data template of the ancient lacquerware lacquer film, and initializing according to the label characteristics of the ancient lacquerware lacquer film.
And acquiring a texture co-occurrence matrix vector and an instrument detection data vector of a paint film sample to be detected, and matching the texture co-occurrence matrix vector and the instrument detection data vector with a pattern vector in a template library. The characteristic matching process adopts the Minkowski distance between the sample to be detected and the template base:
wherein x1，x2Respectively, the characteristic vectors of the sample to be tested and the template library, d12Is a measure of similarity between the two.
In order to improve the matching precision of the detection data, a similarity measurement structure of multi-feature combination is adopted, similarity weighting is carried out on multiple features of a lacquer film of the lacquer ware, and finally a measurement distance is obtained.
Wherein the similarity weighting process is illustrated in figure 3.
The similarity weighting process specifically comprises the following steps:
1. inputting paint film samples to be classified, dividing the characteristics of the paint film samples into n sub-class characteristics, and correspondingly taking n sub-class characteristics from m class classification templates to obtain n groups of characteristic combinations;
2. and calculating the distance measurement between the corresponding sample to be classified and the n types of characteristics of the class template, and then summing the n distance measurements by using a weighting method to obtain a matching result of the weighted subclass template.
3. And optimizing the measurement results of the m types of templates, and acquiring the template result with the highest matching degree as the recognition result.
The similarity weighting measurement can avoid the problem of characteristic sample deviation caused by using a single characteristic measurement, so that the matching result is more reliable. And according to the weighting result, obtaining matching similarity, respectively calculating the similarity measurement of the multi-feature combination for the reference sample in the template, sequencing the results, and outputting the sequencing result by using a threshold value to obtain the identification result of the components and the process of the paint film of the sample lacquerware to be tested.
In summary, the invention has the advantages that:
1. the method is characterized in that big data storage and clustering are carried out on the data of a paint film sample of the ancient lacquerware, the classified aggregation characteristics are obtained, data support is provided for the correlation characteristic analysis of the paint film of the ancient lacquerware, and the method is a great improvement on the existing paint film analysis method.
2. The method adopts a big data pattern matching algorithm in the component and process identification of the paint film of the lacquerware, performs multi-feature coincidence weighting on data of various instruments, improves the reliability and scientificity of a final identification result, is a great improvement on the existing method for identifying the paint film of the ancient lacquerware by adopting an empirical observation method, and ensures that the identification method is more scientific and effective according to detection data and a computer intelligent matching algorithm.
3. The establishment of the ancient lacquerware paint film mode sample database provides a data base for later-stage scientific research, so that later-stage identification and analysis can be established on the data base, and the template database can be continuously expanded, so that the matching algorithm is continuously enhanced.
The invention is a comprehensive data application of the existing unearthed and ancient lacquerware lacquer film, belongs to the characteristic processing of applying a big data algorithm to the ancient lacquerware lacquer film for the first time in the field of cultural relics archaeology, and has novelty and strong practical value on application prospect and social benefit.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.
Claims (8)
1. A discrimination method based on big data composite characteristics of a paint film of an antique paint device is characterized by comprising the following steps:
performing data characteristic extraction on a paint film certified sample of the existing ancient lacquerware, wherein the data characteristic extraction comprises chemical component characteristics, microscopic image characteristics, scanning electron microscope X-ray energy spectrum characteristics, Fourier transform infrared spectrometer characteristics and Raman spectrometer characteristics;
establishing a sample library and a pattern recognition classifier of the ancient lacquer paint film according to the extracted data characteristics, and respectively performing label calibration on the existing ancient lacquer paint film genuine product samples;
acquiring data characteristics of an unknown lacquerware lacquer film to be detected and identified, comparing the data characteristics with parameters in a sample library to acquire an optimal matching result, if the matching is successful, determining the type, process and year prediction result of the sample to be detected, otherwise, prompting abnormal information by a classifier to finish the identification of the lacquerware lacquer film;
the data feature extraction specifically comprises the following steps:
s1, respectively obtaining microscopic images of the front surface, the back surface and the cross section of the paint film sample, and establishing microscopic characterization parameters; carrying out image smoothing on the microscopic image of the paint film, filtering image noise, and eliminating the influence of tiny change between adjacent pixel points of the paint film image on the image;
s2, calculating color aggregation vector characteristics of the microscopic image;
s3, carrying out color space quantization on the microscopic image of the paint film, and respectively quantizing RGB color components of the image aiming at the microscopic image of the color paint film;
s4, carrying out gray level processing on the microscopic image of the paint film, and calculating a gray level co-occurrence matrix of image gray level pixels after normalization;
s5, calculating the EDS characteristics of the paint film according to the X-ray energy dispersion spectrum analysis data of the paint film sample;
s6, calculating the XRD characteristics of the paint film sample according to the back X-ray diffraction analysis data of the paint film sample;
s7, calculating the RS characteristic of the paint film sample according to the laser confocal Raman spectrometer data of the paint film sample;
s8, calculating the FTIR characteristics of the paint film sample according to the Fourier transform infrared spectrometer data of the paint film sample;
wherein step S4 includes:
s41, dividing the paint film image into pixel blocks with the size of a 3 x 3 pixel neighborhood from left to right and from top to bottom in sequence, and sliding in 2 x 2 sub-blocks to obtain 4 sub-neighborhood blocks, wherein each sub-neighborhood comprises 4 pixels, and each sub-neighborhood takes the middle pixel point of the original 3 x 3 pixel neighborhood as a reference pixel;
s42, scanning 4 pixel points in the sub-neighborhood by taking the upper left corner point as a base point, wherein 7 scanning paths are shared;
s43, calculating a co-occurrence matrix of the sub-neighborhood images by 7 scanning paths, wherein the pixel distance is 1, and the included angles between the paths and the transverse axis are 0 degree, 45 degrees, 90 degrees and 135 degrees respectively;
s44, calculating the sum of squares of the gray level co-occurrence matrix element values according to the co-occurrence matrix to obtain an energy attribute;
s45, calculating the difference degree between the primitive pixel points and the field pixel points, and constructing the contrast attribute of the definition of the image and the depth degree of the texture groove;
s45, calculating and measuring the similarity of the space gray level co-occurrence matrix elements in the row or column direction, and obtaining the correlation attribute of the joint probability of the occurrence of the designated pixel pair;
s46, calculating the roughness of the image texture to obtain the uniformity attribute;
s47, calculating the entropy of the sub-neighborhood;
s48, storing the contrast, correlation, entropy, uniformity and energy statistical parameters as feature vectors;
wherein, the characteristic matching process adopts the Minkowski distance between the sample to be detected and the template library; specifically, a similarity measurement structure with multi-feature combination is adopted, similarity weighting is carried out on multiple features of a paint film of a lacquerware, and a measurement distance is obtained finally;
the similarity weighting process specifically comprises the following steps:
inputting paint film samples to be classified, dividing the characteristics of the paint film samples into n sub-class characteristics, and correspondingly taking n sub-class characteristics from m class classification templates to obtain n groups of characteristic combinations;
calculating the distance measurement between the corresponding sample to be classified and the n types of characteristics of the class template, and then using a weighting method to sum the n distance measurements to obtain a matching result of the weighted subclass template;
and optimizing the measurement results of the m types of templates, and acquiring the template result with the highest matching degree as the recognition result.
2. The method for discriminating the composite characteristics based on the big data of the paint film of the antique paint machine as claimed in claim 1, wherein the step S1 comprises:
s11, filtering the microscopic image of the paint film by using a bilateral filter, filtering the noise containing motes and shadows on the surface, and outputting a smooth preprocessed image;
and S12, extracting the edge and the texture line characteristics of the image.
3. The method for discriminating the composite characteristic based on the big data of the paint film of the antique paint machine as claimed in claim 1, wherein the step S2 comprises:
s21, extracting color aggregation vectors from the smoothed image, classifying pixel points into aggregation and non-aggregation according to the image communication region, and writing the pixel points into label marks respectively;
and S22, calculating a color aggregation vector for the aggregation area, and storing the feature marks.
4. The method for discriminating the composite characteristic based on the big data of the paint film of the antique paint device as claimed in claim 1, wherein the step S3 comprises the following steps:
s31, performing 10-level quantization processing on the color image of the paint film according to the color channel components;
and S32, normalizing and storing the quantized component values into a one-dimensional feature vector.
5. The method for discriminating the composite characteristic based on the big data of the paint film of the antique paint device as claimed in claim 1, wherein the step S5 comprises the following steps:
s51, acquiring X-ray energy dispersion spectrum analysis data of the paint film sample, and sequentially extracting the contents of C, O, Hg, S, Al, Si, Fe, Ca, Cu, K, Cl, Cr, Mg and Na elements;
and S52, storing the content of each element as a one-dimensional vector, and adding a feature tag.
6. The method for discriminating the composite characteristic based on the big data of the paint film of the antique paint device as claimed in claim 1, wherein the step S6 comprises the following steps:
s61, acquiring XRD characteristics of the paint film sample according to the back X-ray diffraction analysis data of the paint film sample;
s62, calculating HgS and SiO in XRD characteristics of paint film2The peak position of (d);
s62, storing the peak positions of the corresponding elements of the standard X-ray diffraction standard card PDF;
and S63, calculating and matching the peak position of the sample with the peak position of the PDF element, and acquiring the XRD element component in the sample.
7. The method for discriminating the composite characteristic based on the big data of the paint film of the antique paint device as claimed in claim 1, wherein the step S7 comprises the following steps:
s71, obtaining the RS characteristics of the sample according to the laser confocal Raman spectrometer data of the paint film sample;
s72, matching the laser Raman characteristic peak parameters of HgS and C with the RS characteristic peak of the sample, and deducing whether the sample contains the element;
step S8 includes:
s81, acquiring FTIR characteristics of the paint film sample according to the data of the Fourier transform infrared spectrometer;
s82, comparing infrared absorption peaks of hydroxyl, ether bond and urushiol with the FTIR characteristics, and determining whether the component is contained therein.
8. The method for distinguishing paint film big data composite characteristics based on the antique paint device as claimed in claim 1, wherein when the sample library is established, the method specifically comprises the following steps:
storing the extracted data characteristics as a characteristic space database, and adding the category, lacquerware process, age, region and number parameters of the lacquer film sample into a data label to be used as a standard template library of the characteristic data;
performing network distributed storage on the standard template library, and storing the images of the front, the cross section and the back of the microscope of the sample and the images of the EDS, XRD, RS and FTIR characteristic curves in corresponding paths;
providing a detection input interface through a network server;
the method specifically comprises the following steps of comparing the data characteristics of the unknown lacquerware lacquer film to be detected and identified with the parameters in the sample library:
inputting microscopic images, EDS, XRD, RS and FTIR detection data of the unknown lacquer film to be detected and identified through a network interface;
calculating the microscopic imageThe co-occurrence matrix is matched with the parameters of all known samples in the template library to obtain the similarity sigma of the matching resulti；
EDS, XRD, RS and FTIR detection data of the paint film of the unknown paint container are matched with characteristic parameters of corresponding samples in the standard template library to obtain matching similarity ki、λi、μi、νi；
For similarity sigmaiAnd matching similarity ki、λi、μi、νiWeighting and sorting;
outputting a matching result, and setting a threshold value as a category, process and age prediction result of a sample to be tested of a paint film of the lacquerware; and if the set threshold value is exceeded, the classifier prompts abnormal information to finish the identification of the lacquer film of the lacquer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910480897.5A CN110286139B (en) | 2019-06-04 | 2019-06-04 | Method for distinguishing big data composite characteristics of paint film of ancient lacquerware |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910480897.5A CN110286139B (en) | 2019-06-04 | 2019-06-04 | Method for distinguishing big data composite characteristics of paint film of ancient lacquerware |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110286139A CN110286139A (en) | 2019-09-27 |
| CN110286139B true CN110286139B (en) | 2022-03-15 |
Family
ID=68003333
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910480897.5A Active CN110286139B (en) | 2019-06-04 | 2019-06-04 | Method for distinguishing big data composite characteristics of paint film of ancient lacquerware |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110286139B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111122634A (en) * | 2019-12-25 | 2020-05-08 | 同济大学 | Method for identifying nano plastic particles in aqueous solution based on scanning electron microscope-Raman technology |
| CN111242226A (en) * | 2020-01-16 | 2020-06-05 | 广州运达智能科技有限公司 | Subway train image fault detection method based on correlation coefficient |
| CN114067106B (en) * | 2022-01-12 | 2022-04-15 | 西南交通大学 | Inter-frame contrast-based pantograph deformation detection method and equipment and storage medium |
| CN117093882B (en) * | 2023-10-20 | 2023-12-26 | 南通赛可特电子有限公司 | Optimization method and system for copper plating waste liquid treatment process |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013160243A1 (en) * | 2012-04-24 | 2013-10-31 | Basf Coatings Gmbh | Hdr images as a basis for effect characterisation |
| CN106124508A (en) * | 2016-07-01 | 2016-11-16 | 宁夏六维辩证文物鉴定研究所 | A kind of verification retrieval system and authentication method |
| CN107064196A (en) * | 2017-04-28 | 2017-08-18 | 洛阳理工学院 | A kind of authentication method of bronze relic |
| CN107092699A (en) * | 2017-04-28 | 2017-08-25 | 洛阳理工学院 | A kind of verification retrieval method based on database |
| CN107101993A (en) * | 2017-04-28 | 2017-08-29 | 洛阳理工学院 | A kind of authentication method of ceramic-like historical relic |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101059425A (en) * | 2007-05-29 | 2007-10-24 | 浙江大学 | Method and device for identifying different variety green tea based on multiple spectrum image texture analysis |
| CN107966420B (en) * | 2016-10-19 | 2020-09-18 | 中国石油化工股份有限公司 | Method for predicting crude oil property by near infrared spectrum |
| CN107992901A (en) * | 2017-12-18 | 2018-05-04 | 武汉大学 | A kind of borehole radar image rock stratum sorting technique based on textural characteristics |
-
2019
- 2019-06-04 CN CN201910480897.5A patent/CN110286139B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013160243A1 (en) * | 2012-04-24 | 2013-10-31 | Basf Coatings Gmbh | Hdr images as a basis for effect characterisation |
| CN106124508A (en) * | 2016-07-01 | 2016-11-16 | 宁夏六维辩证文物鉴定研究所 | A kind of verification retrieval system and authentication method |
| CN107064196A (en) * | 2017-04-28 | 2017-08-18 | 洛阳理工学院 | A kind of authentication method of bronze relic |
| CN107092699A (en) * | 2017-04-28 | 2017-08-25 | 洛阳理工学院 | A kind of verification retrieval method based on database |
| CN107101993A (en) * | 2017-04-28 | 2017-08-29 | 洛阳理工学院 | A kind of authentication method of ceramic-like historical relic |
Non-Patent Citations (4)
| Title |
|---|
| Study of historical Chinese lacquer culture and technology – Analysis of Chinese Qin-Han dynasty lacquerware;Meesook Sung et al.;《Journal of Cultural Heritage》;20160601;第21卷;889-893 * |
| 六件宋代温州漆器成分结构与工艺剖析;李晓远等;《文物保护与考古科学》;20180831;第30卷(第4期);44-52 * |
| 战国早期皮甲胄漆膜检测与研究;李澜;《中国生漆》;20170930;第36卷(第3期);第33-36页 * |
| 无损检测技术在漆器分析中的应用;任红艳等;《中国生漆》;20170331;第36卷(第1期);43-47 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110286139A (en) | 2019-09-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110286139B (en) | Method for distinguishing big data composite characteristics of paint film of ancient lacquerware | |
| CN106295124B (en) | The method of a variety of image detecting technique comprehensive analysis gene subgraph likelihood probability amounts | |
| CN106599854B (en) | Automatic facial expression recognition method based on multi-feature fusion | |
| CN115294113A (en) | Wood veneer quality detection method | |
| Li et al. | Soft measurement of wood defects based on LDA feature fusion and compressed sensor images | |
| CN110210448B (en) | Intelligent face skin aging degree identification and evaluation method | |
| CN115187548A (en) | Mechanical part defect detection method based on artificial intelligence | |
| CN114998323B (en) | Deformed steel bar abnormity determination method based on attention mechanism | |
| CN110991389B (en) | Matching method for judging appearance of target pedestrian in non-overlapping camera view angles | |
| CN112766227B (en) | Hyperspectral remote sensing image classification method, device, equipment and storage medium | |
| CN103984940B (en) | A kind of method based on high-spectral data identification hematization | |
| CN115994907B (en) | Intelligent processing system and method for comprehensive information of food detection mechanism | |
| CN111401426A (en) | Small sample hyperspectral image classification method based on pseudo label learning | |
| CN113222062A (en) | Method, device and computer readable medium for tobacco leaf classification | |
| CN110766657B (en) | Laser interference image quality evaluation method | |
| CN109271544B (en) | Method and device for automatically selecting painter representatives | |
| JPH0573663A (en) | Recognition method for picture of three-dimensional object | |
| CN117372332A (en) | Fabric flaw detection method based on improved YOLOv7 model | |
| CN114863189B (en) | Intelligent image identification method based on big data | |
| CN115713634A (en) | Color collocation evaluation method combining similarity measurement and visual perception | |
| CN113792666A (en) | Concrete classification method and system based on scanning electron microscope images | |
| CN112257544A (en) | High-resolution remote sensing image post-earthquake building detection method based on sparse dictionary | |
| Roux et al. | Multiscale anisotropic texture unsupervised clustering for photographic paper | |
| Le et al. | Systematic design of object shape matched wavelet filter banks for defect detection | |
| CN111008653B (en) | Clustering optimization method for mixed pigment information unmixing |
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 |