CN102855632B - A kind of number variable localization method for bill printing on-line checkingi - Google Patents
A kind of number variable localization method for bill printing on-line checkingi Download PDFInfo
- Publication number
- CN102855632B CN102855632B CN201210309550.2A CN201210309550A CN102855632B CN 102855632 B CN102855632 B CN 102855632B CN 201210309550 A CN201210309550 A CN 201210309550A CN 102855632 B CN102855632 B CN 102855632B
- Authority
- CN
- China
- Prior art keywords
- image
- bill
- color
- octree
- leaf node
- 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
Abstract
The present invention relates to a kind of number variable localization method for bill printing on-line checkingi, it comprises step S01: gather number colouring information in bill printing, and preserve; Step S02: online acquisition bill printed matter image, the image of collection must comprise a complete graph text information of bill; Step S03: the color quantization method combined based on octree structure quantizes and denoising the image gathered; Step S04: process in step S03 is obtained image and carries out binaryzation, and image is projected from horizontal direction and vertical direction, according to number feature, data analysis is carried out to this, and obtain final number area coordinate.Method of the present invention does not need standard picture masterplate, avoids a large amount of wastes caused because obtaining actual standard bill printing image, and not only too much complicated calculations is not brought in process into, ensure that arithmetic speed but also number positioning precision can be made to be improved.
Description
Technical field
The invention belongs to printing technology, be specifically related to a kind of number variable localization method for bill printing on-line checkingi.
Background technology
Bill printing often comprises variable information and fixed information, and in printing process, general point two independent process carry out, and namely first print fixed information, then print variable information.For bill printing, variable information is some note numbers mainly, and this is also bill printing and important ingredient thereof, and note number is also the important means of bill anti-counterfeit, and the quality of number printing quality is the main factor determining invoice printing quality.
Quality testing at present for bill printing has developed into the on-line checkingi mode based on machine vision by manual detection.And for on-line checkingi mode, the location of number variable is the key affecting Detection results, if locate inaccurate, identifying just cannot be carried out.
The existing localization method for bill printing on-line checkingi mainly contains following two kinds of methods:
(1) pattern matching method: the method is by obtaining the number information in note number machine, and online acquisition bill printing image carries out matching primitives, until matching degree meets when imposing a condition, then locates successfully.The method is based on image model matching algorithm, therefore calculated amount is larger, real-time location efficiency is lower, by the number that obtains in note number machine, in picture specification, there is some difference with the picture specification of actual acquisition simultaneously, also can affect matching effect to a certain extent, requirements at the higher level are proposed to the matching condition set by rule of thumb.
(2) coordinate pixel projection method: the method is by being converted to gray level image by collection image, and set threshold value and carry out binary conversion treatment, then the image processed is carried out pixel projection respectively on ordinate and horizontal ordinate, according to the result obtained that projects, and carry out data analysis in conjunction with the feature of note number, thus obtain locating information.The method, when carrying out pixel projection, is easily subject to the interference of other pattern in bill, thus can affect the precision of positioning analysis.
Summary of the invention
The object of the invention is the deficiency existed for prior art, propose a kind of number variable localization method for bill printing on-line checkingi.Due in bill printing, printing color is general more single, this feature that this method is printed for bill, adopt the coding method of octree structure, image is quantized and denoising, and obtain the locating area of number variable on this basis, contemplated by the invention the requirement of on-line checkingi for locating speed and precision two aspect.
The present invention adopts following scheme to realize: a kind of number variable localization method for bill printing on-line checkingi, is characterized in that comprising the following steps:
Step S01: gather number colouring information in bill printing, and preserve;
Step S02: online acquisition bill printed matter image, the image of collection must comprise a complete graph text information of bill;
Step S03: the color quantization method combined based on octree structure quantizes and denoising the image gathered;
Step S04: process in step S03 is obtained image and carries out binaryzation, and image is projected from horizontal direction and vertical direction, according to number feature, data analysis is carried out to this, and obtain final number area coordinate.
In an embodiment of the present invention, gathering number colouring information in bill printing in described step S01 is the COLOR COMPOSITION THROUGH DISTRIBUTION by manual observation bill, in hand-kept bill color species number K, K be greater than 1 integer, utilize the colouring information of number in CCD camera collection bill.
In an embodiment of the present invention, described colouring information adopts RBG color-values to represent, is designated as
, wherein
,
with
represent R, G, B color component value of note number standard colors respectively.
In an embodiment of the present invention, the color quantization method first based on octree structure in described step S03 is carried out coding to the color of image gathered and is set up Octree, position prior in bill images rgb value is placed on the upper strata of this Octree.
In an embodiment of the present invention, image to be quantized and the concrete steps of denoising comprise:
Step S001: scan whole Octree, judges color Octree leaf node quantity N, N be greater than 1 integer, the number of colors 3 that the needs that are greater than in advance setting retain if leaf segment is counted, if i.e. N>3, then carries out merger to leaf node;
Step S002: according to the rule of described coding, carries out being back-calculated to obtain rgb value corresponding to 3 leaf nodes
, wherein
, calculated by following formula:
In above formula
represent the value of chromatism corresponding to i-th leaf node between color and standard colors in Octree, get wherein minimum
corresponding rgb value, and other leaf node is directly deleted, set up the color palette quantized and after denoising, what this palette mainly comprised is the coding that in image, each pixel rgb value is corresponding in Octree;
Step S003: according to described color palette, and according to the B-rule of the rule of described coding, original image color is quantized and denoising.
In an embodiment of the present invention, first search the degree of depth maximum leaf node during described step S001 operation, pay the utmost attention to the leaf node that merger frequency is minimum when the degree of depth is identical.
In an embodiment of the present invention, the concrete steps of described merger are: the father node first reading this leaf node, then delete all child nodes of father node, and the color information that child node stores is passed to father node, finally this father node is set to leaf node.
In an embodiment of the present invention, described step S002 is by CCD camera online acquisition bill printed matter image.
The present invention compared with prior art tool has the following advantages: the present invention does not need standard picture masterplate, avoids a large amount of wastes caused because obtaining actual standard bill printing image.In addition not only, in the present invention, image quantization and denoising process are the methods combining pixel projection on the basis of octree quantization algorithm, do not bring too much complicated calculations into, ensure that arithmetic speed but also number positioning precision can be made to be improved.
Accompanying drawing explanation
Fig. 1 is the system flow schematic diagram of the embodiment of the present invention.
Fig. 2 is a kind of number variable localization method realization flow figure for bill printing on-line checkingi.
Fig. 3 is the image color coding rule schematic diagram based on Octree.
Fig. 4 is based on octree quantization denoising method implementing procedure figure.
Embodiment
Below in conjunction with drawings and Examples, the present invention will be further described.
As shown in Figure 1, the present embodiment provides a kind of number variable localization method for bill printing on-line checkingi, it is characterized in that comprising the following steps:
Step S01: gather number colouring information in bill printing, and preserve;
Step S02: online acquisition bill printed matter image, the image of collection must comprise a complete graph text information of bill;
Step S03: the color quantization method combined based on octree structure quantizes and denoising the image gathered;
Step S04: process in step S03 is obtained image and carries out binaryzation, and image is projected from horizontal direction and vertical direction, according to number feature, data analysis is carried out to this, and obtain final number area coordinate.
Concrete, refer to Fig. 2, Fig. 2 is the comparatively concrete schematic flow sheet of the present embodiment.The present embodiment comprises the following steps:
(1) by the COLOR COMPOSITION THROUGH DISTRIBUTION of manual observation bill, color species number K in hand-kept bill, utilizes the colouring information of number in CCD camera collection bill, and this colouring information adopts RBG color-values to represent, is designated as
, wherein
,
with
represent R, G, B color component value of note number standard colors respectively;
(2) by CCD camera online acquisition bill printed matter image, and store with RGB color format, the image of collection must comprise the graph text information of a complete bill surface;
(3) utilize the color quantization method based on octree structure to encode to the color of image gathered, its implementing procedure as shown in Figure 3, specifically describes as follows:
After setting quantizes, image needs the number of colors retained to be 3, and the degree of depth setting Octree is 4.The RGB color value of the image collected in step (2) is distributed in the Octree of stratiform.The each node of Octree has 8 child nodes, the degree of depth be 4 Octree maximumly can to have
individual leaf node, for bill printed matter surface COLOR COMPOSITION THROUGH DISTRIBUTION enough, in Octree, each leaf node is to the path of root node, is determined by the rgb value of color.The binary coding of the RGB component of color directly provides the position of color in octree nodes.At top layer, the 7th (most significant digit) of getting R, G, B component value is combined, and forms the index value of 3, and the 7th of R, G, B component lays respectively on 2,1,0 of index value.Along with the increase of the number of plies, the figure place of R, G, B component of use reduces.Specifically as shown in Figure 2, R, G, B color component is respectively 118(01110110), 47 (00101111), the pixel of 209 (11010001), after octree encoding, the index of the index of the 0th level of child nodes to be the index of the 1,1st level of child nodes be the 5,2nd level of child nodes is 6.Position prior in bill images rgb value is placed on the comparatively upper strata of Octree by this coding method.
(4) according to the colouring information gathered in step (1)
, the Octree generated in integrating step (2), quantizes and denoising image.Concrete steps are as follows:
1. scan whole Octree, judge color Octree leaf node quantity N, the number of colors 3 that the needs that are greater than in advance setting retain if leaf segment is counted, if i.e. N>3, then merger is carried out to leaf node.First search the degree of depth maximum leaf node during operation, pay the utmost attention to the leaf node that merger frequency is minimum when the degree of depth is identical.The concrete operations of merger are: the father node first reading this leaf node, then delete all child nodes of father node, and the color information that child node stores is passed to father node, finally this father node are set to leaf node;
2. the end of scan, it is 3 certainly that the leaf segment now retained is counted, and by the coding rule shown in Fig. 2, carries out being back-calculated to obtain rgb value corresponding to this 3 leaf nodes
, wherein
, calculated by following formula.
3. get wherein minimum
corresponding rgb value, and other leaf node is directly deleted, set up the color palette quantized and after denoising, what this palette mainly comprised is the coding that in image, each pixel rgb value is corresponding in Octree;
4. according to the color palette that obtains above, and according to the B-rule of the coding rule proposed in 1., original image color is quantized and denoising.
(5) a kind of color is only comprised owing to processing through step (4) in the image that obtains, therefore quick binary conversion treatment can be carried out to image, and image is projected from horizontal direction and vertical direction, according to number feature, data analysis is carried out to this, and obtain final number field coordinate.
The foregoing is only preferred embodiment of the present invention, all equalizations done according to the present patent application the scope of the claims change and modify, and all should belong to covering scope of the present invention.
Claims (3)
1., for a number variable localization method for bill printing on-line checkingi, it is characterized in that comprising the following steps:
Step S01: gather number colouring information in bill printing, and preserve;
Step S02: online acquisition bill printed matter image, the image of collection must comprise the complete graph text information of a bill;
Step S03: the color quantization method combined based on octree structure quantizes and denoising the image gathered;
Step S04: carry out binaryzation by processing the image obtained in step S03, and image is projected from horizontal direction and vertical direction, according to number feature, data analysis is carried out to described projection, and obtain final number area coordinate;
Wherein, gathering number colouring information in bill printing in described step S01 is the COLOR COMPOSITION THROUGH DISTRIBUTION by manual observation bill, in hand-kept bill color species number K, K be greater than 1 integer, utilize the colouring information of number in CCD camera collection bill; Described colouring information adopts RBG color-values to represent, is designated as
, wherein
,
with
represent R, G, B color component value of note number standard colors respectively;
Color quantization method first based on octree structure in described step S03 is carried out coding to the color of image gathered and is set up Octree, position prior in bill images rgb value is placed on the upper strata of this Octree;
In addition, image to be quantized and the concrete steps of denoising comprise:
Step S001: scan whole Octree, judges color Octree leaf node quantity N, N be greater than 1 integer, the number of colors 3 that the needs that are greater than in advance setting retain if leaf segment is counted, if i.e. N>3, then carries out merger to leaf node;
Step S002: according to the rule of described coding, carries out being back-calculated to obtain rgb value corresponding to 3 leaf nodes
, wherein
, calculated by following formula:
In above formula
represent the value of chromatism corresponding to i-th leaf node between color and standard colors in Octree, get wherein minimum
corresponding rgb value, and other leaf node is directly deleted, set up the color palette quantized and after denoising, what this palette comprised is the coding that in image, each pixel rgb value is corresponding in Octree;
Step S003: according to described color palette, and according to the B-rule of the rule of described coding, original image color is quantized and denoising;
Wherein, described step S02 is by CCD camera online acquisition bill printed matter image.
2. the number variable localization method for bill printing on-line checkingi according to claim 1, it is characterized in that: first search the degree of depth maximum leaf node during described step S001 operation, pay the utmost attention to the leaf node that merger frequency is minimum when the degree of depth is identical.
3. the number variable localization method for bill printing on-line checkingi according to claim 1, it is characterized in that: the concrete steps of described merger are: the father node first reading this leaf node, then all child nodes of father node are deleted, and the color information that child node stores is passed to father node, finally this father node is set to leaf node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210309550.2A CN102855632B (en) | 2012-08-28 | 2012-08-28 | A kind of number variable localization method for bill printing on-line checkingi |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210309550.2A CN102855632B (en) | 2012-08-28 | 2012-08-28 | A kind of number variable localization method for bill printing on-line checkingi |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102855632A CN102855632A (en) | 2013-01-02 |
| CN102855632B true CN102855632B (en) | 2015-07-29 |
Family
ID=47402194
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210309550.2A Active CN102855632B (en) | 2012-08-28 | 2012-08-28 | A kind of number variable localization method for bill printing on-line checkingi |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102855632B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117291208B (en) * | 2023-11-24 | 2024-01-23 | 四川数盾科技有限公司 | Two-dimensional code extraction method and system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567764A (en) * | 2012-01-13 | 2012-07-11 | 中国工商银行股份有限公司 | Bill certificate and system for improving electronic image recognition efficiency |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20100074708A (en) * | 2008-12-24 | 2010-07-02 | 노틸러스효성 주식회사 | Connecting structure for bills check sensing cap of an automated teller machine |
-
2012
- 2012-08-28 CN CN201210309550.2A patent/CN102855632B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567764A (en) * | 2012-01-13 | 2012-07-11 | 中国工商银行股份有限公司 | Bill certificate and system for improving electronic image recognition efficiency |
Non-Patent Citations (2)
| Title |
|---|
| 数码相机的彩色图像预处理过程;曲兴华等;《光学精密工程》;20090115;第17卷(第1期);第207-212页 * |
| 票据印刷号码自动识别技术研究;贾彦金;《万方数据库》;20090304;第1、6-14页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102855632A (en) | 2013-01-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107424143B (en) | Mining belt conveyor coal quantity detection method based on binocular stereoscopic vision depth perception | |
| CN104392228B (en) | Unmanned plane image object class detection method based on conditional random field models | |
| CN113409382B (en) | Method and device for measuring damaged area of vehicle | |
| CN108323204A (en) | A kind of method and intelligent terminal of detection face flaw point | |
| CN101915573B (en) | Positioning measurement method based on key point detection of marker | |
| CN102542275B (en) | Automatic identification method for identification photo background and system thereof | |
| CN103248906B (en) | Method and system for acquiring depth map of binocular stereo video sequence | |
| CN108198230A (en) | A kind of crop and fruit three-dimensional point cloud extraction system based on image at random | |
| CN109978871B (en) | Fiber bundle screening method integrating probability type and determination type fiber bundle tracking | |
| CN103914699A (en) | Automatic lip gloss image enhancement method based on color space | |
| JP5470607B2 (en) | Computer system and method for aligning images / charts | |
| CN107578376A (en) | The fork division of distinguished point based cluster four and the image split-joint method of local transformation matrix | |
| CN110110646A (en) | A kind of images of gestures extraction method of key frame based on deep learning | |
| CN104846054A (en) | Method for automatically detecting moulds in leucorrhea based on morphological characteristics | |
| CN110727819B (en) | Method for retrieving scale-adaptive pathological full-section image database | |
| CN107545570B (en) | A kind of half with reference to figure reconstructed image quality evaluation method | |
| CN110909640A (en) | Method and device for determining water level line, storage medium and electronic device | |
| CN101246595A (en) | Multi-view point data splitting method of optical three-dimensional scanning system | |
| CN107330873B (en) | Stereo image quality objective evaluation method based on multi-scale binocular fusion and local feature extraction | |
| CN104182970A (en) | Souvenir photo portrait position recommendation method based on photography composition rule | |
| CN108564631A (en) | Car light light guide acetes chinensis method, apparatus and computer readable storage medium | |
| CN107705356A (en) | Image processing method and device | |
| CN109285183A (en) | A kind of multimode video image method for registering based on moving region image definition | |
| CN112561782B (en) | Method for improving reality degree of simulation picture of offshore scene | |
| CN102855632B (en) | A kind of number variable localization method for bill printing on-line checkingi |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C53 | Correction of patent for invention or patent application | ||
| CB02 | Change of applicant information |
Address after: Jinshan Development Zone of Cangshan District of Fuzhou City, Fujian province 350007 Gold Road No. 136 Applicant after: Hongbo Co., Ltd. Address before: Jinshan Development Zone of Cangshan District of Fuzhou City, Fujian province 350007 Gold Road No. 136 Applicant before: Fujian Hongbo Printing Co.,Ltd. |
|
| COR | Change of bibliographic data |
Free format text: CORRECT: APPLICANT; FROM: FUJIAN HONGBO PRINTING CO., LTD. TO: HONGBO CO., LTD. |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |