CN101408937A - Method and apparatus for locating character row - Google Patents
Method and apparatus for locating character row Download PDFInfo
- Publication number
- CN101408937A CN101408937A CNA2008102171998A CN200810217199A CN101408937A CN 101408937 A CN101408937 A CN 101408937A CN A2008102171998 A CNA2008102171998 A CN A2008102171998A CN 200810217199 A CN200810217199 A CN 200810217199A CN 101408937 A CN101408937 A CN 101408937A
- Authority
- CN
- China
- Prior art keywords
- candidate characters
- angle
- inclination
- information
- intercharacter
- 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.)
- Granted
Links
Images
Landscapes
- Character Input (AREA)
Abstract
The invention discloses a method for locating a character row and a device thereof. The method comprises the following steps: obtaining information of candidate characters according to image data to be processed; obtaining the information of an intercharacter gradient according to the information of the candidate characters, wherein, the information of the intercharacter gradient is used for marking the position relation between the candidate characters and the neighborhood candidate characters thereof; and determining a bounding box of the candidate characters rows according to the information of the intercharacter gradient to finish the location. The intercharacter gradient information is obtained according to the candidate character information in location by the technical proposal in an embodiment of the invention, thus accurately locating the character rows with certain gradient on the premise of ensuring better anti-noise capability.
Description
The technology neighborhood
The present invention relates to the Flame Image Process neighborhood, relate in particular to the method and the device of a kind of character row location.
Background technology
In automatic character recognition technologies, with the character of needs identification accurately the location be the difficult point and the key point of whole recognition technology.A complete character recognition process, as process to the character recognition in the magnetic code character row of bill, generally include: treat identification character zone (as the magnetic code character row) binaryzation, character zone to be identified is further located, according to positioning result character zone to be identified is cut out, the character zone that cuts out is carried out feature extraction and identification, as shown in Figure 1.And the precision of locating follow-up cutting, feature extraction and identification has fundamental influence.
Locator meams in the character recognition in traditional magnetic code character row mainly is divided into two kinds of space projection method and space clustering methods.Other is that the character in the magnetic code character row is arranged in rows, and has only delegation or a few row usually, so positioning range has obvious regional feature with the positioning area of common character recognition.And the space projection method is utilized the magnetic code character row bigger characteristics of projection components in the horizontal direction just, chooses appropriate threshold value and determines magnetic code character row locus in vertical direction; The space clustering method is investigated the dense degree of candidate character center on vertical y direction of principal axis based on connected component labeling, locatees magnetic code character row position with this.
But, in realizing process of the present invention,, the magnetic code character row detection of tilting there is certain robustness though the inventor finds that space projection method locating speed is fast, but its anti-noise ability is not enough, especially occurs causing possibly under the situation of black-tape or black patch the location failure in the region of search; Though and the space clustering method has good anti-noise ability, for the magnetic code character row that occurs to a certain degree tilting, because the center of each character is bigger in offset from vertical, so detect poor effect.
Summary of the invention
Technical matters to be solved by this invention is, the method and the device of a kind of character row location is provided.Can guarantee to have better anti noise ability shortly, accurately the location has the to a certain degree character row of inclination simultaneously.
For this reason, on the one hand, embodiments of the invention provide the method for a kind of character row location, and method comprises: obtain information of candidate characters according to pending view data; Obtain intercharacter angle of inclination information according to described information of candidate characters, the position that described intercharacter angle of inclination information is used to identify the neighborhood candidate characters of candidate characters and described candidate characters concerns; Determine the bounding box of described candidate characters according to described intercharacter angle of inclination information, to finish the location.
On the other hand, embodiments of the invention provide the device of a kind of character row location, comprising: the character information acquiring unit is used for obtaining information of candidate characters according to pending view data; The degree of tilt acquiring unit is used for obtaining intercharacter angle of inclination information according to described information of candidate characters, and the position that described intercharacter angle of inclination information is used to identify the neighborhood candidate characters of candidate characters and described candidate characters concerns; Positioning unit is used for determining according to described intercharacter angle of inclination information the bounding box of described candidate characters, to finish the location.
In the technical scheme that the embodiment of the invention adopted, when positioning, also obtain intercharacter angle of inclination information, thereby can guarantee under the prerequisite of anti-noise power preferably, accurately the character row of location with certain degree of tilt according to information of candidate characters.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, to do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for this neighborhood those of ordinary skill, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the schematic flow sheet of existing automatic character recognition technologies;
Fig. 2 is an idiographic flow synoptic diagram of the method for the character row location in the embodiment of the invention;
Fig. 3 is another idiographic flow synoptic diagram of the method for the character row location in the embodiment of the invention;
Fig. 4 is the image synoptic diagram that need position;
Fig. 5 is the result schematic diagram of location;
Fig. 6 is a concrete synoptic diagram of forming of the device of the character row location in the embodiment of the invention;
Fig. 7 is a concrete synoptic diagram of forming of the degree of tilt acquiring unit among Fig. 6;
Fig. 8 is a concrete synoptic diagram of forming of the positioning unit among Fig. 6;
Fig. 9 is another concrete synoptic diagram of forming of the positioning unit among Fig. 6.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, this neighborhood those of ordinary skill all belongs to the scope of protection of the invention not making the every other embodiment that is obtained under the creative work prerequisite.
Below with reference to accompanying drawing embodiments of the invention are described.Referring to Fig. 2, be an idiographic flow synoptic diagram of the method for the character row in embodiment of the invention location.This flow process comprises:
201, obtain information of candidate characters according to pending view data.Specifically can comprise, described pending view data is carried out binaryzation, obtain binary map; Described binary map is carried out the connected domain analysis obtain information of candidate characters,, qualified connected component as candidate characters, and is calculated the centre coordinate of this candidate characters as binary map is carried out connected component labeling.Can pending image not carried out binaryzation yet, but directly on gray-scale map or cromogram, utilize information acquisition information of candidate characters such as gray scale or color.
Wherein, described candidate characters can comprise a plurality of, as, after the candidate characters number that obtains arrives some, determine to position and to carry out subsequent step, otherwise change the zone that obtains view data, obtain view data again.
202, obtain intercharacter angle of inclination information according to described information of candidate characters, the position that described intercharacter angle of inclination information is used to identify the neighborhood candidate characters of candidate characters and described candidate characters concerns.Specifically can comprise:
A, obtain the positional information of candidate characters in the described information of candidate characters, as the centre coordinate of candidate characters;
B, obtain to meet the positional information of other candidate characters of the neighborhood range of condition of described candidate characters according to the positional information of described candidate characters, calculate other contiguous candidate characters in the certain space distance range of this coordinate as centre coordinate according to above-mentioned candidate characters.Promptly this territory is the neighborhood scope based on the spatial measure distance, as based on the ring-type neighborhood scope of Euclidean distance or based on the neighborhood scope of city block distance or based on other the neighborhood scope of distance calculating method.
C, obtain described intercharacter angle of inclination information according to the positional information of described candidate characters and the positional information of described other candidate characters.
203, determine the bounding box of the character row of described candidate characters according to described intercharacter angle of inclination information, to finish the location.Wherein, intercharacter angle of inclination information comprises the intercharacter angle of inclination; Bounding box can be a rectangle frame.
Specifically can be to determine the angle of inclination according to the method for statistics, as: obtain with as described at least one relevant intercharacter angle of inclination of candidate characters; Span to each described intercharacter angle of inclination quantizes; Statistics obtains to fall into the number at the intercharacter angle of inclination in each quantized interval; The described number that obtains each quantized interval according to statistics is determined the character row angle of inclination; Determine the bounding box of candidate characters according to described character row angle of inclination, i.e. the border of determining the character row of candidate characters according to above-mentioned definite described character row angle of inclination and pairing character space positional information thereof.
Also can be to determine the angle of inclination according to the method for cluster, as: obtain with as described at least one relevant intercharacter angle of inclination of candidate characters; Carry out cluster according to each described intercharacter angle of inclination, the classification center that obtains with cluster is as definite character row angle of inclination; Determine the bounding box of the character row of candidate characters according to the described character row angle of inclination of determining.
As shown in Figure 3, for being applied to magnetic code character row to bill when positioning, the idiographic flow synoptic diagram of the method for the character row location in the embodiment of the invention.This flow process comprises:
301, choose subregion in the whole bill image as the zone of fixed-position searching, then the view data in this zone is carried out binaryzation, obtain binary map.Usually the magnetic code character provisional capital in the bill is below bill, therefore, can choose in the bill image below part (as below 1/5) zone as the region of search.
302, the binary map that obtained in the last step is carried out the connected domain analysis, carry out connected component labeling, with position and the size that writes down each connected component according to analysis result.As, can carry out the connected domain analysis according to fast area labeling algorithm.
303, through in the binary map of connected component labeling, according to predetermined wide height (or wide high scope) choose meet the wide high condition of magnetic code character connected component as candidate characters.As, determine that according to some prior imformations predetermined wide height or wide high scope determine candidate characters, the magnitude range of promptly knowing certain character that will locate in advance should be width and highly in what pixels between what pixels.
304, whether the total quantity of statistics candidate characters if the quantity of candidate characters greater than predetermined quantity, then think magnetic code character row exist, carries out 305 steps greater than predetermined quantity; Otherwise, think that then the magnetic code character row does not exist, withdraw from the location, or withdraw from this location, re-execute step 301, choose next region of search and position, finish until search.
305, calculate the centre coordinate of all candidate characters.
306, according to angle of inclination between the centre coordinate calculating character of candidate characters.As, the angle of the line of the centre coordinate of each candidate characters of calculating connection and the centre coordinate of each candidate characters in its neighborhood scope, wherein, the neighborhood scope can be based on the scope of Euclidean distance, and for example, the centre coordinate of establishing candidate characters is (x
0, y
0), be that the territory based on Euclidean distance at center may be defined as then with it:
RingNeibour (x
0, y
0)={ (x, y) | T1＜f (x, y, x
0, y
0)＜T2} (formula 1)
Wherein, the three to four-fold of the desirable candidate characters mean breadth of T1, T2 gets five to six times of candidate characters mean breadth.Satisfy the candidate characters of above-mentioned neighborhood scope definition, then elect the neighborhood candidate characters as and calculate the angle of inclination.
Wherein, above-mentioned territory also can be based on other definition mode, for example city block distance or the like for example.That is, the range formula in the formula 2 also may be defined as:
Range formula 1:f (x, y, x
0, y
0)=| x-x
0|+| y-y
0|
Range formula 2:f (x, y, x
0, y
0)=Max (| x-x
0|, y-y
0|)
Range formula 3:f (x, y, x
0, y
0)=Min (| x-x
0|, y-y
0|)
……
To each candidate characters, or according to the candidate characters of regular selected part, calculate the intercharacter angle of inclination of each candidate characters in itself and the neighborhood scope respectively.Because aforementioned calculation obtains a plurality of angles of inclination and may be not quite similar, need statistics to obtain one near the angle of inclination of actual value.Specifically can adopt multiple probabilistic statistical method.
Method to statistics is exemplified below: the possible range of determining an angle of inclination, as be defined as [Max, Max], wherein the Max value is 10, this angular range is quantized, the long t value of quantization step is 0.1, adds up the number that falls in the above-mentioned a plurality of angles of inclination that obtain in each quantization step, and determines that the angle of quantization step correspondence of the number maximum at the angle of inclination that falls into is the angle of inclination of magnetic code character row.
As, can aforesaid scope and angle counter of quantization step definition, add up each angle of inclination, if angle value drops in certain rank district of counter, the counting in corresponding rank district increases a unit certainly, and writes down the information (as the numbering of candidate characters) of the pairing candidate characters of this angle value simultaneously, count angle with maximum count unit, if count value less than certain threshold value, thinks that the magnetic code character row does not exist, and withdraws from; Otherwise, think that this angle is magnetic code character row angle of inclination.
Also can adopt the method for cluster to determine the angle of inclination, as, all angle values are carried out cluster, the angle of inclination of other center correspondence of maximum kind that obtains with cluster is as the angle of inclination of magnetic code character row, promptly by choosing a classification in the angle classification that cluster found, and the angle value at this classification center with maximum count as net result.。As carry out cluster in following cluster space: there are two coordinate axis in the cluster space, transverse axis reflection angle value information, the counting of longitudinal axis reflection angle value correspondence.Concrete clustering method can adopt multiple clustering algorithms such as kmeans (k average) cluster, ISODATA cluster.
307, according to the angle of inclination of the character row of determining, utilize the last boundary line and the following boundary line of character row of the candidate characters of this angle value correspondence of least square fitting, finish the location.
In the scheme that the embodiment of the invention provided, spatial disposition characteristics (magnetic code character center substantially point-blank segmentation is closely arranged) according to the magnetic code character row, the candidate characters that utilization calculates the what angle of inclination between the neighborhood character is inferred the angle of inclination of magnetic code character row, thereby locatees the magnetic code character row with certain angle of inclination exactly.Wherein, when determining the angle of inclination of magnetic code character row, the stability that degree of tilt is determined has been improved at the angle of inclination that the mode that adopts all angles to vote is calculated whole magnetic code character row.
Simultaneously, for fear of because adjacent magnetic code character pitch from too near, the deflection situation that can not truly reflect the magnetic code character row, so deflection situation with its ring-type field magnetic code character investigated in each magnetic code character, like this inclined degree that the intercharacter angle of inclination of each that obtains can the whole magnetic code character row of more real reflection.The stability and the accuracy of location have been improved.
As shown in Figure 4 and Figure 5, for adopting the situation that actual items is positioned of scheme in the embodiment of the invention.Image for positioning among Fig. 4.Rectangle frame is the result of location among Fig. 5.As seen situation about tilting for magnetic code character row in this image adopts the scheme in the embodiment of the invention still can position accurately.
Accordingly, the embodiment of the invention also provides the device of a kind of character row location, and as shown in Figure 6, this device comprises:
Character information acquiring unit 2 is used for obtaining information of candidate characters according to pending view data; Degree of tilt acquiring unit 4 is used for obtaining intercharacter angle of inclination information according to described information of candidate characters, and the position that described intercharacter angle of inclination information is used to identify the neighborhood candidate characters of candidate characters and described candidate characters concerns; Positioning unit 6 is used for determining according to described intercharacter angle of inclination information the bounding box of the character row of described candidate characters, to finish the location.
Wherein, as shown in Figure 7, degree of tilt acquiring unit 4 comprises: primary importance information module 40 is used for obtaining the positional information of described information of candidate characters candidate characters; Second place information module 42 is used for obtaining to meet according to the positional information of described candidate characters the positional information of other candidate characters of the neighborhood range of condition of described candidate characters; Inclination information module 44 is used for obtaining described intercharacter angle of inclination information according to the positional information of described candidate characters and the positional information of described other candidate characters.
As shown in Figure 8, positioning unit 6 comprises: acquisition module 60, at least one the intercharacter angle of inclination relevant with described candidate characters that is used for obtaining described intercharacter angle of inclination information; Quantization modules 62 is used for the span at each described intercharacter angle of inclination is quantized; Statistical module 64 is used to add up the number that obtains to fall into the intercharacter angle of inclination in each quantized interval; Angle determination module 66 is used for determining the character row angle of inclination according to the described number of each quantized interval of statistics acquisition; Border determination module 68 is used for the bounding box that the character row of candidate characters is determined at the definite described character row angle of inclination of basis, to finish the location.
Or as shown in Figure 9, positioning unit 6 comprises: acquisition module 60, at least one the intercharacter angle of inclination relevant with described candidate characters that is used for obtaining described intercharacter angle of inclination information; Cluster module 61 is used for carrying out cluster according to each described intercharacter angle of inclination, and other center of maximum kind that obtains with cluster is as the character row angle of inclination; Border determination module 68, be used for according to the described character row angle of inclination of determining determine candidate characters the character row bounding box, to finish the location.
Wherein, consistent in other details of above-mentioned module and unit and the other embodiments of the invention do not done herein and given unnecessary details.
In the scheme that the embodiment of the invention provided, spatial disposition characteristics (magnetic code character center substantially point-blank segmentation is closely arranged) according to the magnetic code character row, the candidate characters that utilization calculates the what angle of inclination between the neighborhood character is inferred the angle of inclination of magnetic code character row, thereby accurate in locating has the magnetic code character row at certain angle of inclination.Wherein, when determining the angle of inclination of magnetic code character row, the stability that degree of tilt is determined has been improved at the angle of inclination that the mode that adopts all angles to vote is calculated whole magnetic code character row.
Simultaneously, for fear of because adjacent magnetic code character pitch from too near, the deflection situation that can not truly reflect the magnetic code character row, so deflection situation with its ring-type field magnetic code character investigated in each magnetic code character, like this inclined degree that the intercharacter angle of inclination of each that obtains can the whole magnetic code character row of more real reflection.The stability and the accuracy of location have been improved.
Through the above description of the embodiments, the technician of this neighborhood can be well understood to each embodiment and can realize by the mode that software adds essential general hardware platform, can certainly pass through hardware.Based on such understanding, the part that technique scheme contributes to prior art in essence in other words can embody with the form of software product, this computer software product can be stored in the computer-readable recording medium, as ROM/RAM, magnetic disc, CD etc., comprise that some instructions are with so that a computer equipment (can be a personal computer, server, perhaps network equipment etc.) carry out the described method of some part of each embodiment or embodiment.
Above-described embodiment does not constitute the qualification to this technical scheme protection domain.Any at above-mentioned embodiment spirit and principle within done modification, be equal to and replace and improvement etc., all should be included within the protection domain of this technical scheme.
Claims (10)
1, the method for a kind of character row location is characterized in that described method comprises:
Obtain information of candidate characters according to pending view data;
Obtain intercharacter angle of inclination information according to described information of candidate characters, the position that described intercharacter angle of inclination information is used to identify the neighborhood candidate characters of candidate characters and described candidate characters concerns;
Determine the bounding box of the character row of described candidate characters according to described intercharacter angle of inclination information, to finish the location.
2, the method for claim 1, it is characterized in that, describedly obtain intercharacter angle of inclination information according to described information of candidate characters, the position that described intercharacter angle of inclination information is used to identify the neighborhood candidate characters of candidate characters and described candidate characters concerns and comprises:
Obtain the positional information of candidate characters in the described information of candidate characters;
Obtain to meet the positional information of other candidate characters of the neighborhood range of condition of described candidate characters according to the positional information of described candidate characters;
Obtain described intercharacter angle of inclination information according to the positional information of described candidate characters and the positional information of described other candidate characters.
3, method as claimed in claim 2 is characterized in that, described neighborhood scope is the neighborhood scope based on the spatial measure distance.
4, the method for claim 1 is characterized in that, comprises the intercharacter angle of inclination in the information of described intercharacter angle of inclination, describedly determines that according to described intercharacter angle of inclination information the bounding box of described candidate characters comprises:
Obtain at least one intercharacter angle of inclination relevant with described candidate characters;
Span to each described intercharacter angle of inclination quantizes;
Statistics obtains to fall into the number at the intercharacter angle of inclination in each quantized interval;
The described number that obtains each quantized interval according to statistics is determined the character row angle of inclination;
Determine the bounding box of the character row of described candidate characters according to described character row angle of inclination.
5, the method for claim 1 is characterized in that, comprises the intercharacter angle of inclination in the information of described intercharacter angle of inclination, describedly determines that according to described intercharacter angle of inclination information the bounding box of described candidate characters comprises:
Obtain at least one intercharacter angle of inclination relevant with described candidate characters;
Carry out cluster according to each described intercharacter angle of inclination, other center of maximum kind that obtains with cluster is as definite character row angle of inclination;
According to described character row angle of inclination determine described candidate characters the bounding box of character row.
6, as each described method in the claim 1 to 5, it is characterized in that, describedly obtain information of candidate characters according to pending view data and comprise:
Described pending view data is carried out the connected domain analysis obtain information of candidate characters.
7, the device of a kind of character row location is characterized in that described device comprises:
The character information acquiring unit is used for obtaining information of candidate characters according to pending view data;
The degree of tilt acquiring unit is used for obtaining intercharacter angle of inclination information according to described information of candidate characters, and the position that described intercharacter angle of inclination information is used to identify the neighborhood candidate characters of candidate characters and described candidate characters concerns;
Positioning unit is used for determining according to described intercharacter angle of inclination information the bounding box of the character row of described candidate characters, to finish the location.
8, device as claimed in claim 7 is characterized in that, described degree of tilt acquiring unit comprises:
The primary importance information module is used for obtaining the positional information of described information of candidate characters candidate characters;
Second place information module is used for obtaining to meet according to the positional information of described candidate characters the positional information of other candidate characters of the neighborhood range of condition of described candidate characters;
The inclination information module is used for obtaining described intercharacter angle of inclination information according to the positional information of described candidate characters and the positional information of described other candidate characters.
9, as claim 7 or 8 described devices, it is characterized in that described positioning unit comprises:
Acquisition module, at least one the intercharacter angle of inclination relevant that is used for obtaining described intercharacter angle of inclination information with described candidate characters;
Quantization modules is used for the span at each described intercharacter angle of inclination is quantized;
Statistical module is used to add up the number that obtains to fall into the intercharacter angle of inclination in each quantized interval;
The angle determination module is used for determining the character row angle of inclination according to the described number of each quantized interval of statistics acquisition;
The border determination module is used for determining according to described character row angle of inclination the bounding box of the character row of candidate characters, to finish the location.
10, as claim 7 or 8 described devices, it is characterized in that described positioning unit comprises:
Acquisition module, at least one the intercharacter angle of inclination relevant that is used for obtaining described intercharacter angle of inclination information with described candidate characters;
The cluster module is used for carrying out cluster according to each described intercharacter angle of inclination, and other center of maximum kind that obtains with cluster is as definite character row angle of inclination;
The border determination module is used for determining according to described character row angle of inclination the bounding box of the character row of candidate characters, to finish the location.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008102171998A CN101408937B (en) | 2008-11-07 | 2008-11-07 | Method and apparatus for locating character row |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008102171998A CN101408937B (en) | 2008-11-07 | 2008-11-07 | Method and apparatus for locating character row |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101408937A true CN101408937A (en) | 2009-04-15 |
| CN101408937B CN101408937B (en) | 2011-12-21 |
Family
ID=40571945
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008102171998A Active CN101408937B (en) | 2008-11-07 | 2008-11-07 | Method and apparatus for locating character row |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101408937B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102779275A (en) * | 2012-07-04 | 2012-11-14 | 广州广电运通金融电子股份有限公司 | Paper characteristic identification method and relative device |
| CN105261110A (en) * | 2015-10-26 | 2016-01-20 | 江苏国光信息产业股份有限公司 | Efficient DSP banknote serial number recognizing method |
| CN106815561A (en) * | 2016-12-22 | 2017-06-09 | 北京五八信息技术有限公司 | Business license printed page analysis method and device |
| CN107092864A (en) * | 2017-03-27 | 2017-08-25 | 成都优译信息技术股份有限公司 | Drawing Reading text method and system based on clustering |
| CN107680108A (en) * | 2017-07-28 | 2018-02-09 | 平安科技（深圳）有限公司 | Inclination value-acquiring method, device, terminal and the storage medium of tilted image |
| CN107918935A (en) * | 2017-12-05 | 2018-04-17 | 深圳市北电仪表有限公司 | The images match cutter device and cutting method of embedded system |
| CN108268868A (en) * | 2017-07-28 | 2018-07-10 | 平安科技（深圳）有限公司 | Inclination value-acquiring method and device, terminal, the storage medium of ID Card Image |
| CN109961068A (en) * | 2017-12-26 | 2019-07-02 | 阿里巴巴集团控股有限公司 | Image recognition, training, searching method and device and equipment, medium |
| CN110375645A (en) * | 2019-07-04 | 2019-10-25 | 成都绝影智能科技有限公司 | A kind of method and apparatus of character size detection |
| CN111199224A (en) * | 2018-11-20 | 2020-05-26 | 中国电信股份有限公司 | Curved character recognition method and device |
| CN111914764A (en) * | 2020-08-05 | 2020-11-10 | 杭州睿琪软件有限公司 | Image processing method, image processing apparatus, electronic device, and storage medium |
| TWI766930B (en) * | 2018-01-11 | 2022-06-11 | 台達電子工業股份有限公司 | Classification-based character skew correction apparatus and method thereof |
| CN115249362A (en) * | 2022-09-20 | 2022-10-28 | 京华信息科技股份有限公司 | OCR table recognition method and system based on connectivity of pixels in stable direction |
-
2008
- 2008-11-07 CN CN2008102171998A patent/CN101408937B/en active Active
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102779275B (en) * | 2012-07-04 | 2015-06-17 | 广州广电运通金融电子股份有限公司 | Paper characteristic identification method and relative device |
| CN102779275A (en) * | 2012-07-04 | 2012-11-14 | 广州广电运通金融电子股份有限公司 | Paper characteristic identification method and relative device |
| CN105261110A (en) * | 2015-10-26 | 2016-01-20 | 江苏国光信息产业股份有限公司 | Efficient DSP banknote serial number recognizing method |
| CN106815561A (en) * | 2016-12-22 | 2017-06-09 | 北京五八信息技术有限公司 | Business license printed page analysis method and device |
| CN107092864B (en) * | 2017-03-27 | 2020-02-21 | 成都优译信息技术股份有限公司 | Drawing text reading method and system based on cluster analysis |
| CN107092864A (en) * | 2017-03-27 | 2017-08-25 | 成都优译信息技术股份有限公司 | Drawing Reading text method and system based on clustering |
| CN107680108A (en) * | 2017-07-28 | 2018-02-09 | 平安科技（深圳）有限公司 | Inclination value-acquiring method, device, terminal and the storage medium of tilted image |
| CN108268868B (en) * | 2017-07-28 | 2020-07-10 | 平安科技（深圳）有限公司 | Method and device for acquiring inclination value of identity card image, terminal and storage medium |
| CN108268868A (en) * | 2017-07-28 | 2018-07-10 | 平安科技（深圳）有限公司 | Inclination value-acquiring method and device, terminal, the storage medium of ID Card Image |
| US11074443B2 (en) | 2017-07-28 | 2021-07-27 | Ping An Technology (Shenzhen) Co., Ltd. | Method and device for acquiring slant value of slant image, terminal and storage medium |
| CN107918935A (en) * | 2017-12-05 | 2018-04-17 | 深圳市北电仪表有限公司 | The images match cutter device and cutting method of embedded system |
| CN109961068A (en) * | 2017-12-26 | 2019-07-02 | 阿里巴巴集团控股有限公司 | Image recognition, training, searching method and device and equipment, medium |
| TWI766930B (en) * | 2018-01-11 | 2022-06-11 | 台達電子工業股份有限公司 | Classification-based character skew correction apparatus and method thereof |
| CN111199224B (en) * | 2018-11-20 | 2023-06-23 | 中国电信股份有限公司 | Method and device for recognizing curved characters |
| CN111199224A (en) * | 2018-11-20 | 2020-05-26 | 中国电信股份有限公司 | Curved character recognition method and device |
| CN110375645A (en) * | 2019-07-04 | 2019-10-25 | 成都绝影智能科技有限公司 | A kind of method and apparatus of character size detection |
| CN111914764A (en) * | 2020-08-05 | 2020-11-10 | 杭州睿琪软件有限公司 | Image processing method, image processing apparatus, electronic device, and storage medium |
| CN111914764B (en) * | 2020-08-05 | 2023-09-15 | 杭州睿琪软件有限公司 | Image processing method, image processing apparatus, electronic device, and storage medium |
| CN115249362B (en) * | 2022-09-20 | 2022-12-27 | 京华信息科技股份有限公司 | OCR table recognition method and system based on connectivity of pixels in stable direction |
| CN115249362A (en) * | 2022-09-20 | 2022-10-28 | 京华信息科技股份有限公司 | OCR table recognition method and system based on connectivity of pixels in stable direction |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101408937B (en) | 2011-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101408937B (en) | Method and apparatus for locating character row | |
| CN106295646B (en) | A kind of registration number character dividing method and device based on deep learning | |
| CN103093181B (en) | A kind of method and apparatus of license plate image location | |
| CN111179152B (en) | Road identification recognition method and device, medium and terminal | |
| CN101615252B (en) | Method for extracting text information from adaptive images | |
| EP3214577A1 (en) | Image text recognition method and device | |
| CN109633676A (en) | A kind of method and system based on the laser radar obstruction detection direction of motion | |
| EP3474191A1 (en) | Method and device for constructing a table including information on a pooling type and testing method and testing device using the same | |
| CN104200188B (en) | Method and system for rapidly positioning position detection patterns of QR code | |
| WO2017041600A1 (en) | Chinese-sensitive code feature pattern detection method and system | |
| CN103207987A (en) | Indicating value identification method of dial instrument | |
| CN111932537A (en) | Object deformation detection method and device, computer equipment and storage medium | |
| CN109598947B (en) | Vehicle identification method and system | |
| CN112013921B (en) | Method, device and system for acquiring water level information based on water level gauge measurement image | |
| CN100487723C (en) | Method for recognizing print form italic character | |
| CN112883820A (en) | Road target 3D detection method and system based on laser radar point cloud | |
| CN111062285A (en) | Parking lot intelligent monitoring and recognizing system based on deep learning algorithm | |
| CN112183301A (en) | Building floor intelligent identification method and device | |
| JP6364565B1 (en) | Analysis device, crack detection processing device, and analysis program | |
| CN108344997B (en) | Road guardrail rapid detection method based on trace point characteristics | |
| CN110322428A (en) | Detect the method, apparatus and electronic equipment of tunnel defect | |
| CN112434583B (en) | Lane transverse deceleration marking line detection method and system, electronic equipment and storage medium | |
| CN109726722B (en) | Character segmentation method and device | |
| CN115375679B (en) | Edge-finding and point-searching positioning method and device for defective chip | |
| CN103235951A (en) | Preliminary positioning method for matrix type two-dimensional bar code |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee | ||
| CP03 | Change of name, title or address |
Address after: 523000, building 6, 310, 311 and 3, 312 South Industrial Road, Songshan hi tech Industrial Development Zone, Guangdong, Dongguan Patentee after: GUANGDONG MICROPATTERN SOFTWARE CO., LTD. Address before: 523808 room 18, building 509, Songshan overseas student Pioneer Park, Dongguan, Guangdong Patentee before: Dongguan MicroPattern Corporation |