CN106778764A - A kind of auxiliary first osteocomma conjugation methods based on color of image feature extraction - Google Patents
A kind of auxiliary first osteocomma conjugation methods based on color of image feature extraction Download PDFInfo
- Publication number
- CN106778764A CN106778764A CN201611028723.8A CN201611028723A CN106778764A CN 106778764 A CN106778764 A CN 106778764A CN 201611028723 A CN201611028723 A CN 201611028723A CN 106778764 A CN106778764 A CN 106778764A
- Authority
- CN
- China
- Prior art keywords
- color
- image
- histogram
- osteocomma
- auxiliary
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
The invention discloses a kind of auxiliary first osteocomma conjugation methods based on color of image feature extraction, including:From hsv color spatial model, RGB data is converted to corresponding HSV numerical value in Essential colour;The division of H, S, V component is carried out, the channel image for storing H, S, V component plane is respectively created;The RGB color of original image is converted into hsv color space and is split;Color histogram is shown after creating histogram and assignment according to HS components;The comparing of color histogram is carried out using OpenCV image libraries.The present invention extracts first bone color histogram as color characteristic by the use of computer, and using histogrammic EMD distances, coefficient correlation, card side occurs simultaneously, and Pasteur carries out last characteristic matching apart from this 5 kinds of methods;Can apply in first bone conjugation procedure, as the quick lookup before being conjugated, reduce the complexity in conjugation procedure, lifting is conjugated speed and efficiency, but also can use for reference in other image indexing systems.
Description
Technical field
The invention belongs to inscriptions on bones or tortoise shells research field, more particularly to a kind of auxiliary first osteocomma based on color of image feature extraction
Conjugation methods.
Background technology
The conjugated of first bone is a very cumbersome process, in conjugation procedure, it is necessary to which each side to first bone is carried out carefully
Analysis and considering can just make a decision, this will expend the time and efforts of numerous studies personnel.
In the conjugation procedure of the inscriptions on bones or tortoise shells, the matching of first bone chip can be carried out according to the different attribute of first bone.Wherein one
Item attribute, is exactly the color characteristic of first bone piece youngster.That is, the same color of the different fragments of first bone is divided in overall
Difference is little, even identical in cloth.But, it is completely inadequate to rely solely on the color of first bone and be conjugated.According to
The meaning of the matching technique carried out according to first bone color of image feature is:From substantial amounts of first bone chip, using computer picture
Identification technology and mode-matching technique, quickly match in this similar first bone of the color characteristic of first bone.In color
Carrying out in similar first bone chip again is artificial or computer is conjugated, and the conjugated speed of its first bone and efficiency must have been carried
It is high.
In sum, the conjugated work of existing first bone is more complicated, speed is conjugated and efficiency is low.
The content of the invention
It is an object of the invention to provide a kind of auxiliary first osteocomma conjugation methods based on color of image feature extraction, it is intended to
Solve the conjugated work of existing first bone more complicated, speed and the low problem of efficiency is conjugated.
The present invention is achieved in that a kind of auxiliary first osteocomma conjugation methods based on color of image feature extraction include:
Step one, from hsv color spatial model, RGB data is converted to corresponding HSV numerical value in Essential colour;
Step 2, the division for carrying out H, S, V component, are respectively created the channel image for storing H, S, V component plane;
Step 3, the RGB color of original image is converted to hsv color space and is split;
Step 4, histogram and assignment are created according to HS components after color histogram is shown;
Step 5, the comparing that color histogram is carried out using OpenCV image libraries.
Further, EMD distances, the coefficient correlation of color histogram is respectively adopted, card side occurs simultaneously, and Pasteur's distance carries out face
The comparing of Color Histogram.
Further, using one image of locking, convert the method for an other image to carry out the matching of color histogram,
By Optimum Matching and worst matching, the similarity value of each matching process is extrapolated.
The present invention extracts first bone color histogram as color characteristic using computer, and using histogrammic EMD apart from,
Coefficient correlation, card side occurs simultaneously, and Pasteur carries out last characteristic matching apart from this 5 kinds of methods;Can apply in first bone conjugation procedure
In, as the quick lookup before being conjugated, the complexity in conjugation procedure being reduced, lifting is conjugated speed and efficiency, but also can be with
Use for reference in other image indexing systems.
Brief description of the drawings
Fig. 1 is the auxiliary first osteocomma conjugation methods flow based on color of image feature extraction provided in an embodiment of the present invention
Figure.
Specific embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to embodiments, to the present invention
It is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to
Limit the present invention.
Below in conjunction with the accompanying drawings and specific embodiment is further described to application principle of the invention.
Refer to Fig. 1:
A kind of auxiliary first osteocomma conjugation methods based on color of image feature extraction, including:
S101, from hsv color spatial model, RGB data is converted to corresponding HSV numerical value in Essential colour;
S102, the division for carrying out H, S, V component, are respectively created the channel image for storing H, S, V component plane;
S103, the RGB color of original image is converted to hsv color space and is split;
S104, histogram and assignment are created according to HS components after color histogram is shown;
S105, the comparing that color histogram is carried out using OpenCV image libraries.
The embodiment of the present invention has selected hsv color spatial model, RGB data is converted to corresponding HSV numbers in Essential colour
Value.By basic experimental calculation, the scope of table 1 is drawn:
The hsv color spatial model numerical value of table 1 is changed
The scope of HSV can be drawn by table 1:
H:Minimum value 0, maximum 180;
S:Minimum value 0, maximum 255;
V:Minimum value 0, maximum 255.
1) division of H, S, V component
Because colour brightness V need not be considered, for simplicity, for shade of color H, nH_Level=18 is divided into
Part, color saturation S is divided into nS_level=8 parts, and the width of the small rectangle of each color of color histogram is set to
RectW=8 pixels are wide.
The overall width of the color histogram for then finally obtaining:NHistWidth=(nH_Level*nS_level) * rectW;
2) it is respectively created the channel image for storing H, S, V component plane:(pSrcImg is the pointer of original image)
H plane, components:PHplaneImg=cvCreateImage (cvGetSize (pSrcImg), 8,1)
S plane, components:PSplaneImg=cvCreateImage (cvGetSize (pSrcImg), 8,1)
V plane, components:PVplaneImg=cvCreateImage (cvGetSize (pSrcImg), 8,1)
3) RGB color of original image is converted into hsv color space and is split
Color space conversion:CvCvtColor (pSrcImg, pHsvImg, CV_BGR2HSV)
Color space is split:CvSplit (pHsvImg, pHplaneImg, pSplaneImg, pVplaneImg, 0)
4) histogram and assignment are created according to HS components:
NHS_HistLevel={ 18,8 }, fpHS_ranges are the scope of HS.
Color histogram:PHist=cvCreateHist (2, nHS_HistLevel, CV_HIST_ARRAY, fpHS_
Ranges, 1) pHSplanes={ pHplaneImg, pSplaneImg }
Fill Color histogram:CvCalcHist (pHSplanes, pHist, 0,0)
5) color histogram is shown.
The embodiment of the present invention utilizes OpenCV image libraries, and EMD distances, the coefficient correlation of color histogram, card is respectively adopted
Side, occurs simultaneously, and Pasteur's distance carries out the comparing of color histogram, and specific computational methods are as follows:
The calculating of EMD distances can be called such as minor function in OpenCV image libraries:
Float cvCalcEMD2 (CvArr*signature1, CvArr*signature2, int distance_type)
Parameter 1,2 represents two color histogram characteristic vectors respectively, it is actually used before, done the conversion of matrix;
Distance_type represents comparison criterion, has CV_DIST_L1, CV_DIST_L2, and CV_DIST_C to be respectively mark
Accurate criterion.In application program, we use CV_DIST_L2 comparison criterions.
Comparing two dense histogrammic methods in OpenCV image libraries can call such as minor function:
Double cvCompareHist (CvHistogram*hist1, CvHistogram*hist2, int method);
Parameter 1,2 represents two dense histograms;
Parameter 3 represents comparative approach:Method method numerical value is as follows:
1)CV_COMP_CORREL:Calculate two coefficient correlations of color histogram;
2)CV_COMP_CHISQR:Calculate two card side's coefficients of color histogram;
3)CV_COMP_INTERSECT:Calculate two common factors of color histogram;
4)CV_COMP_BHATTACHARYYA:Calculate two Pasteur's distances of color histogram.
For multiple matching strategies of color histogram, using one image of locking, the method for converting an other image
To be tested.
There are 2 identical images, it can be deduced that five kinds of Optimum Matchings of comparative approach.
There is the very big image of 2 difference, it can be deduced that five kinds of worst matchings of comparative approach.
By Optimum Matching and worst matching, the similarity value of each matching process can be deduced.Such as table 2 below.
The histogram comparative approach experiment test of table 2
Can be drawn the following conclusions by table 2:
1) the EMD distances between two color histograms are closer to 0, and the similitude between color histogram is stronger;Conversely,
EMD distances are bigger, then similarity is weaker.Therefore show that EMD two first bone images of the distance less than or equal to 1.5 can tentatively be judged as phase
Seemingly.
2) coefficient correlation between two color histograms is closer to 1, and the similitude between color histogram is stronger;Conversely,
Closer to 0, then similarity is weaker for coefficient correlation.Its number range (0,1].Therefore draw two of coefficient correlation more than or equal to 0.5
First bone image can tentatively be judged as similar.
3) the card side's coefficient between two color histograms is closer to 0, and the similitude between color histogram is stronger;Conversely,
Card side's coefficient is bigger, then similarity is weaker.Its number range [0 ,+inf).Therefore obtain card release side's coefficient and be less than or equal to 50000
Two first bone images can tentatively be judged as similar.
4) the common factor coefficient between two color histograms is bigger, and the similitude between color histogram is stronger;Conversely, occuring simultaneously
Coefficient is smaller, then similarity is weaker.Its number range [0 ,+inf).Therefore draw two of common factor numerical value more than or equal to 50000
First bone image can tentatively be judged as similar.
Pasteur's distance values between two color histograms are closer to 0, and the similitude between color histogram is stronger;Instead
It, closer to 1, then similarity is weaker for Pasteur's distance values.Its number range [0,1).Therefore show that Pasteur's distance is less than or equal to
0.5 two first bone images can tentatively be judged as similar.
It with the experience of the person of joining together is support that manually conjugated first bone chip is, with the in kind or book of rubbings according to the feature of first bone, break
Whether trace, radian, color and luster are identical, character stroke, brill, chisel, the goodness of fit of the trace that burns and fore-telling at font style, oracle inscriptions of the Shang Dynasty content, mark of break
Million direction etc. differentiates first bone is conjugated.The letters such as unearthed time, place, the hole position of the first bone for treating conjugated are should be noted when joining together
Breath.According to nearest statistics, the unearthed quantity up to 150,000 of first bone, be manually conjugated to be differentiated one by one, very multiple
It is miscellaneous and need to take considerable time, even need some months sometimes, so that time of several years, and be possible to finally be conjugated.
The present invention extracts first bone color histogram as color characteristic using computer, and using histogrammic EMD apart from,
Coefficient correlation, card side occurs simultaneously, and Pasteur carries out last characteristic matching apart from this 5 kinds of methods;Can apply in first bone conjugation procedure
In, as the quick lookup before being conjugated, only need the time of several seconds to can be carried out preliminary screening, select and meet matching characteristic
Picture;With the artificial complexity for being conjugated and comparing and reduce in conjugation procedure, it is not necessary to which labor costs' plenty of time consults largely
Book information, it is not necessary to differentiated to each picture by naked eyes such that it is able to which lifting is conjugated speed and efficiency, but also can
To use for reference in other image indexing systems.
Presently preferred embodiments of the present invention is the foregoing is only, is not intended to limit the invention, it is all in essence of the invention
Any modification, equivalent and improvement made within god and principle etc., should be included within the scope of the present invention.
Claims (3)
1. a kind of auxiliary first osteocomma conjugation methods based on color of image feature extraction, it is characterised in that described based on image face
The auxiliary first osteocomma conjugation methods that color characteristic is extracted include:
Step one, from hsv color spatial model, RGB data is converted to corresponding HSV numerical value in Essential colour;
Step 2, the division for carrying out H, S, V component, are respectively created the channel image for storing H, S, V component plane;
Step 3, the RGB color of original image is converted to hsv color space and is split;
Step 4, histogram and assignment are created according to HS components after color histogram is shown;
Step 5, the comparing that color histogram is carried out using OpenCV image libraries.
2. the auxiliary first osteocomma conjugation methods of color of image feature extraction are based on as claimed in claim 1, it is characterised in that respectively
The comparing of color histogram is carried out using the EMD distances of color histogram, coefficient correlation, card side, common factor, Pasteur's distance.
3. the auxiliary first osteocomma conjugation methods of color of image feature extraction are based on as claimed in claim 2, it is characterised in that used
One image of locking, converts the method for an other image to carry out the matching of color histogram, by Optimum Matching and worst
Match somebody with somebody, extrapolate the similarity value of each matching process.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611028723.8A CN106778764A (en) | 2016-11-18 | 2016-11-18 | A kind of auxiliary first osteocomma conjugation methods based on color of image feature extraction |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611028723.8A CN106778764A (en) | 2016-11-18 | 2016-11-18 | A kind of auxiliary first osteocomma conjugation methods based on color of image feature extraction |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106778764A true CN106778764A (en) | 2017-05-31 |
Family
ID=58970336
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611028723.8A Pending CN106778764A (en) | 2016-11-18 | 2016-11-18 | A kind of auxiliary first osteocomma conjugation methods based on color of image feature extraction |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106778764A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107358079A (en) * | 2017-06-16 | 2017-11-17 | 微梦创科网络科技(中国)有限公司 | Real-time face identifies login validation method and system |
CN110598030A (en) * | 2019-09-26 | 2019-12-20 | 西南大学 | Oracle bone rubbing classification method based on local CNN framework |
CN112837334A (en) * | 2021-04-02 | 2021-05-25 | 河南大学 | Automatic conjugation method of Chinese character and sketch image |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101551823A (en) * | 2009-04-20 | 2009-10-07 | 浙江师范大学 | Comprehensive multi-feature image retrieval method |
CN102087742A (en) * | 2011-01-26 | 2011-06-08 | 王爱民 | Tortoise shell fragment conjugating method based on image processing |
CN102622420A (en) * | 2012-02-22 | 2012-08-01 | 哈尔滨工程大学 | Trademark image retrieval method based on color features and shape contexts |
CN104331447A (en) * | 2014-10-29 | 2015-02-04 | 邱桃荣 | Cloth color card image retrieval method |
CN105243667A (en) * | 2015-10-13 | 2016-01-13 | 中国科学院自动化研究所 | Target re-identification method based on local feature fusion |
-
2016
- 2016-11-18 CN CN201611028723.8A patent/CN106778764A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101551823A (en) * | 2009-04-20 | 2009-10-07 | 浙江师范大学 | Comprehensive multi-feature image retrieval method |
CN102087742A (en) * | 2011-01-26 | 2011-06-08 | 王爱民 | Tortoise shell fragment conjugating method based on image processing |
CN102622420A (en) * | 2012-02-22 | 2012-08-01 | 哈尔滨工程大学 | Trademark image retrieval method based on color features and shape contexts |
CN104331447A (en) * | 2014-10-29 | 2015-02-04 | 邱桃荣 | Cloth color card image retrieval method |
CN105243667A (en) * | 2015-10-13 | 2016-01-13 | 中国科学院自动化研究所 | Target re-identification method based on local feature fusion |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107358079A (en) * | 2017-06-16 | 2017-11-17 | 微梦创科网络科技(中国)有限公司 | Real-time face identifies login validation method and system |
CN110598030A (en) * | 2019-09-26 | 2019-12-20 | 西南大学 | Oracle bone rubbing classification method based on local CNN framework |
CN110598030B (en) * | 2019-09-26 | 2022-05-17 | 西南大学 | Oracle bone rubbing classification method based on local CNN framework |
CN112837334A (en) * | 2021-04-02 | 2021-05-25 | 河南大学 | Automatic conjugation method of Chinese character and sketch image |
CN112837334B (en) * | 2021-04-02 | 2022-07-05 | 河南大学 | Automatic conjugation method of Chinese character and sketch image |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105450411B (en) | The method, apparatus and system of authentication are carried out using card feature | |
Lin | Face detection in complicated backgrounds and different illumination conditions by using YCbCr color space and neural network | |
CN105320248B (en) | Aerial gesture input method and device | |
Dhavalikar et al. | Face detection and facial expression recognition system | |
CN110084135B (en) | Face recognition method, device, computer equipment and storage medium | |
CN102542275B (en) | Automatic identification method for identification photo background and system thereof | |
CN105844242A (en) | Method for detecting skin color in image | |
CN109740572A (en) | A kind of human face in-vivo detection method based on partial color textural characteristics | |
Lakshmi et al. | Segmentation algorithm for multiple face detection for color images with skin tone regions | |
Atharifard et al. | Robust component-based face detection using color feature | |
CN106778764A (en) | A kind of auxiliary first osteocomma conjugation methods based on color of image feature extraction | |
CN104484652A (en) | Method for fingerprint recognition | |
CN109377494A (en) | A kind of semantic segmentation method and apparatus for image | |
CN106909884A (en) | A kind of hand region detection method and device based on hierarchy and deformable part sub-model | |
CN110598574A (en) | Intelligent face monitoring and identifying method and system | |
Youlian et al. | Face detection method using template feature and skin color feature in rgb color space | |
Sharif et al. | Real time face detection | |
Nugroho et al. | Negative content filtering for video application | |
Liang et al. | Applying Image Processing Technology to Face Recognition. | |
CN107798325B (en) | Card recognition method and apparatus, computer storage medium | |
CN112164057A (en) | Qualified label detection method, storage medium and electronic equipment | |
Ghouzali et al. | A skin detection algorithm based on discrete cosine transform and generalized Gaussian density | |
Ghimire et al. | A lighting insensitive face detection method on color images | |
Devi et al. | Active facial tracking | |
CN109598737A (en) | A kind of image border recognition methods and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170531 |