CN104573656A - License plate color judging method based on connected region information - Google Patents
License plate color judging method based on connected region information Download PDFInfo
- Publication number
- CN104573656A CN104573656A CN201510013056.5A CN201510013056A CN104573656A CN 104573656 A CN104573656 A CN 104573656A CN 201510013056 A CN201510013056 A CN 201510013056A CN 104573656 A CN104573656 A CN 104573656A
- Authority
- CN
- China
- Prior art keywords
- image
- connected region
- license plate
- car plate
- vertical edge
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a license plate color judging method based on connected region information. The license plate color judging method comprises the following steps that dimension normalization processing is conducted on license plate images; effective character zones in the license plate images obtained after the dimension normalization processing is conducted are obtained; contrast enhancement is conducted on the license plate images obtained after the effective character zones are obtained; binary value perpendicular edge images of the license plate images obtained after the contrast enhancement is conducted are obtained; the optimal width of character strokes is obtained based on the binary value perpendicular edge images of the license plate images; morphological operation is conducted, foreground connected region images and background connected region images of the license plates are obtained; the foreground connected region images are optimized; statistics is conducted on gray level average values corresponding to the foreground connected regions and gray level average values of the background connected regions; color types of the license plate are judged. By means of the license plate color judging method, the judgment result is more accurate, and higher robustness is achieved on the interference such as color shift and stains of the license plates.
Description
Technical field
The present invention relates to field of license plate recognition, be specifically related to a kind of car plate color determination methods based on connected region information.
Background technology
Car plate color is a key character of car plate, in our country, the color type of car plate is not unique, mainly contain Four types: wrongly written or mispronounced character of the blue end, yellow end surplus, black matrix wrongly written or mispronounced character, white gravoply, with black engraved characters or The Scarlet Letter, if now to the misjudgment of car plate color, directly can affect the correctness of successive character segmentation, and then affect final Car license recognition rate.Therefore, it is the requisite part of License Plate Character Segmentation that car plate color type judges, the accuracy rate for whole recognition result has important impact.
At present, conventional car plate color determination methods mainly contains following a few class:
(1) based on the car plate color determination methods of coloured image, first these class methods by R, G, B color component information of all pixels in statistics license plate area, according to the difference degree between each component information, and then judge the color type of car plate.These class methods, for color car plate clearly, have and well judge effect, but the colour cast occurring stain, fade or cause due to different light rays when car plate, these class methods there will be serious erroneous judgement, and when there is gray scale car plate, the method directly lost efficacy.
(2) based on the car plate color determination methods of gray level image, these class methods are reduced to two classes car plate color type: the bright word in the dark end and dark word of putting one's cards on the table; First obtain the binary image of car plate, then add up foreground target point proportion, in conjunction with features such as vertical projections, comprehensive descision car plate color.The advantage of these class methods is applied widely, for the car plate of uniform gray level, judges that effect is fine, fading and colour cast for car plate, has certain robustness simultaneously, but for comprising the car plate of strong jamming and stain, due to the impact by binary image quality, the method there will be erroneous judgement.
Summary of the invention
The object of the present invention is to provide a kind of car plate color determination methods based on connected region information, adopt the vertical edge of ticket to obtain character connected region, and then judge car plate color type, judged result is more accurate, for interference such as car plate colour cast and stains, there is stronger robustness.
Technical scheme of the present invention is:
Based on a car plate color determination methods for connected region information, comprise the following steps:
(1) license plate image is carried out size normalized;
(2) the significant character region in the license plate image of size normalized is obtained;
(3) license plate image behind acquisition significant character region is carried out contrast strengthen;
(4) utilize vertical edge detective operators, obtain the two-value vertical edge figure of the license plate image through contrast strengthen;
(5) based on the two-value vertical edge figure of license plate image, the optimal width of character stroke is obtained;
(6) morphology operations is carried out to the license plate image after the optimal width of acquisition character stroke, obtain prospect connected region image and the background connected region image of car plate;
(7) prospect connected region image is optimized;
(8) gray average corresponding to prospect connected region and gray average corresponding to background connected region is added up;
(9) based on gray average corresponding to prospect connected region and gray average corresponding to background connected region, car plate color type is judged.
In step (3), described carries out contrast strengthen by the license plate image after obtaining significant character region, and the following formula of concrete employing realizes:
Wherein, f (x, y) represents the gray-scale value of former figure, and g (x, y) represents the gray-scale value strengthening image, t
minfor former figure minimum gray value adds 10, t
maxfor former figure gray scale maximal value deducts 10.
In step (4), described utilizes vertical edge detective operators, obtains the two-value vertical edge figure of the license plate image through contrast strengthen; Specifically comprise the following steps:
(41) based on edge detection operator formula, the vertical edge characteristic pattern of the license plate image through contrast strengthen is obtained; Described edge detection operator formula is:
(42) utilize maximum kind spacing algorithm, obtain the two-value vertical edge figure of the license plate image through contrast strengthen.
In step (5), the described two-value vertical edge figure based on license plate image, obtains the optimal width of character stroke; Specifically comprise the following steps:
(51) the pending current line of two-value vertical edge figure is entered;
(52) whole trip point is found from left to right;
(53) calculate the distance between current transition point and previous trip point, a rear trip point respectively, and the distance being less than certain threshold value is put into stroke width list;
(54) judge that whether current transition point is last trip point of this row; If so, step (55) is then performed; If not, then enter next trip point, return and perform step (53);
(55) judge that whether current line is last column of image; If so, step (57) is then performed; If not, then step (56) is performed;
(56) enter the next line of two-value vertical edge figure, return and perform step (51)-(54);
(57) based on stroke width list, stroke width histogram is built;
(58) width that selection maximal dimension is corresponding is as the character stroke width n of this license plate image.
In step (6), described carries out morphology operations to the license plate image after the optimal width of acquisition character stroke, obtains prospect connected region image and the background connected region image of car plate; Specifically comprise the following steps:
(61) structure based element template formula (3), carries out two-value vertical edge figure
secondary morphological dilations computing, obtains image image_dilate; Described structural element template type (3) is:
(62) structure based element template formula (4), carries out 2 closing operation of mathematical morphology to image image_dilate, obtains image image_close; Described structural element template type (4) is:
(63) inverse process is carried out to image image_close, background extraction connected region image image_bg;
(64) structure based element template formula (3), carries out image_close
secondary morphological erosion computing, obtains prospect connected region image image_obj.
In step (7), described optimization prospect connected region image; Specifically comprise the following steps:
(71) the current pending row of image image_obj is entered;
(72) utilize following formula, find whole left trip points and left hand edge point;
Wherein, g (x, y) is image intensity value;
(73) based on current transition point, foreground points all in the n-1 neighborhood of right side is retained;
(74) judge that whether current transition point is last trip point of this row; If so, step (75) is then performed; If not, then enter next trip point, return and perform step (73);
(75) judge that whether current line is last column of image image_obj; If so, then export and optimize foreground region image; If not, then step (76) is performed;
(76) enter the next line of image image_obj, return and perform step (71)-(75);
After being finished, according to step (71)-(76), find right hand edge point and corresponding foreground area retention point, superpose based on left hand edge reserve area with based on right hand edge reserve area, obtain final optimization foreground area.
In step (9), the described gray average corresponding based on prospect connected region and gray average corresponding to background connected region, judge car plate color type; The following formula of concrete employing realizes:
Wherein,
the gray average that expression prospect connected region is corresponding,
represent the gray average that background connected region is corresponding, class=1 represents that car plate belongs to the bright word type in the dark end, and class=0 represents that car plate belongs to dark word type of putting one's cards on the table.
The present invention adopts vertical edge and morphology operations to obtain the character connected region of car plate, and the gray scale difference value of prospect and background is judged based on connected region, and then judge car plate color type, judged result is more accurate, and for interference such as car plate colour cast and stains, there is stronger robustness.
Accompanying drawing explanation
Fig. 1 is method flow diagram of the present invention;
Fig. 2 obtains the optimum character duration process flow diagram of car plate;
Fig. 3 is optimization prospect connected region process flow diagram;
Fig. 4 chooses significant character area schematic;
Fig. 5 is size normalized image;
Fig. 6 chooses significant character area image;
Fig. 7 is image enhaucament figure;
Fig. 8 is two-value vertical edge figure;
Fig. 9 is background connected region image;
Figure 10 is prospect connected region image;
Figure 11 is the background connected region image optimized.
Embodiment
The present invention is further illustrated below in conjunction with the drawings and specific embodiments.
As shown in Figure 1, specific embodiment of the invention step is as follows:
S1, carry out size normalization to license plate image, eliminate the impact that different size causes, effect as shown in Figure 5.
S2, choose significant character region, eliminate the interference in the regions such as car plate frame, as shown in Figure 4, select the 0.6w*0.4h region of picture centre, effect as shown in Figure 6 for specific practice.
S3, employing formula (1), carry out contrast strengthen to license plate image, effect is as Fig. 7;
Wherein, f (x, y) represents the gray-scale value of former figure, and g (x, y) represents the gray-scale value strengthening image, t
minfor former figure minimum gray value adds 10, t
maxfor former figure gray scale maximal value deducts 10.
S4, use vertical edge detective operators, obtain the two-value vertical edge figure of license plate image, as shown in Figure 8, concrete steps are as follows for effect:
S41, based on edge detection operator formula (2), obtain vertical edge characteristic pattern;
S42, based on maximum kind spacing algorithm, obtain two-value vertical edge figure.
S5, as shown in Figure 2, obtain the optimal width n of stroke, concrete steps are as follows:
S51, enter current pending row;
S52, find all trip points from left to right;
S53, respectively calculating current transition point and the distance between previous trip point and a rear trip point, select less distance to put into stroke width list;
S54, judge last trip point of current transition point whether this row; If so, then S55 is performed; If not, then enter next trip point, return and perform step S53;
S55, judge that current line is last column of image; If so, then step S57 is performed; If not, then step S56 is performed;
S56, enter next line, continue to perform step S51 to step S54;
S57, structure stroke width histogram;
S58, select width corresponding to maximal dimension as the stroke width n of this license plate image.
S6, carry out morphology operations, obtain prospect and the background connected region of car plate respectively, concrete steps are as follows:
S61, structure based element template formula (3), carry out two-value vertical edge figure
secondary morphological dilations computing, image is now denoted as image_dilate;
S62, structure based element template formula (4), carry out 2 closing operation of mathematical morphology to image_dilate, image is now denoted as image_close;
S63, carry out inverse process to image_close, background extraction connected region image image_bg, effect as shown in Figure 9;
S64, structure based element template formula (3), carry out image_close
secondary morphological erosion computing, image now and prospect connected region image image_obj, effect is as Figure 10.
S7, as shown in Figure 3, optimize prospect connected region image, effect is as Figure 11, and concrete steps are as follows:
S71, enter the current pending row of image_obj;
S72, employing formula (5), find all left trip points and left hand edge point;
Wherein, g (x, y) is image intensity value.
S73, based on current transition point, retain foreground points all in the n-1 neighborhood of right side;
S74, judge last trip point of current transition point whether this row; If so, then step S75 is performed; If not, then enter next trip point, return and perform step S73;
S75, judge that current line is last column of image; If so, then export and optimize foreground region image; If not, then step S76 is performed;
S76, enter next line, continue to perform step S71 to step S75;
After being finished, according to step S71 to step S76, find right hand edge point and corresponding foreground area retention point, superpose based on left hand edge reserve area with based on right hand edge reserve area, obtain final optimization foreground area.
The gray average that S8, statistics foreground area are corresponding
the gray average corresponding with background area
S9, employing formula (6), judge car plate color type.In the present embodiment, foreground area average is 147.5, and background area average is 210.64, and therefore car plate belongs to dark word type of putting one's cards on the table, and car plate actual type is yellow end surplus.
Wherein, class=1 represents that car plate belongs to the bright word type in the dark end, and class=0 represents that car plate belongs to dark word type of putting one's cards on the table.
The above embodiment is only be described the preferred embodiment of the present invention; not scope of the present invention is limited; under not departing from the present invention and designing the prerequisite of spirit; the various distortion that those of ordinary skill in the art make technical scheme of the present invention and improvement, all should fall in protection domain that claims of the present invention determine.
Claims (7)
1., based on a car plate color determination methods for connected region information, it is characterized in that, comprise the following steps:
(1) license plate image is carried out size normalized;
(2) the significant character region in the license plate image of size normalized is obtained;
(3) license plate image behind acquisition significant character region is carried out contrast strengthen;
(4) utilize vertical edge detective operators, obtain the two-value vertical edge figure of the license plate image through contrast strengthen;
(5) based on the two-value vertical edge figure of license plate image, the optimal width of character stroke is obtained;
(6) morphology operations is carried out to the license plate image after the optimal width of acquisition character stroke, obtain prospect connected region image and the background connected region image of car plate;
(7) prospect connected region image is optimized;
(8) gray average corresponding to prospect connected region and gray average corresponding to background connected region is added up;
(9) based on gray average corresponding to prospect connected region and gray average corresponding to background connected region, car plate color type is judged.
2. a kind of car plate color determination methods based on connected region information according to claim 1, it is characterized in that: in step (3), described carries out contrast strengthen by the license plate image after obtaining significant character region, and the following formula of concrete employing realizes:
Wherein, f (x, y) represents the gray-scale value of former figure, and g (x, y) represents the gray-scale value strengthening image, t
minfor former figure minimum gray value adds 10, t
maxfor former figure gray scale maximal value deducts 10.
3. a kind of car plate color determination methods based on connected region information according to claim 1, it is characterized in that: in step (4), described utilizes vertical edge detective operators, obtains the two-value vertical edge figure of the license plate image through contrast strengthen; Specifically comprise the following steps:
(41) based on edge detection operator formula, the vertical edge characteristic pattern of the license plate image through contrast strengthen is obtained; Described edge detection operator formula is:
(42) utilize maximum kind spacing algorithm, obtain the two-value vertical edge figure of the license plate image through contrast strengthen.
4. a kind of car plate color determination methods based on connected region information according to claim 1, is characterized in that: in step (5), the described two-value vertical edge figure based on license plate image, obtains the optimal width of character stroke; Specifically comprise the following steps:
(51) the pending current line of two-value vertical edge figure is entered;
(52) whole trip point is found from left to right;
(53) calculate the distance between current transition point and previous trip point, a rear trip point respectively, and the distance being less than certain threshold value is put into stroke width list;
(54) judge that whether current transition point is last trip point of this row; If so, step (55) is then performed; If not, then enter next trip point, return and perform step (53);
(55) judge that whether current line is last column of image; If so, step (57) is then performed; If not, then step (56) is performed;
(56) enter the next line of two-value vertical edge figure, return and perform step (51)-(54);
(57) based on stroke width list, stroke width histogram is built;
(58) width that selection maximal dimension is corresponding is as the character stroke width n of this license plate image.
5. a kind of car plate color determination methods based on connected region information according to claim 1, it is characterized in that: in step (6), described carries out morphology operations to the license plate image after the optimal width of acquisition character stroke, obtains prospect connected region image and the background connected region image of car plate; Specifically comprise the following steps:
(61) structure based element template formula (3), carries out two-value vertical edge figure
secondary morphological dilations computing, obtains image image_dilate; Described structural element template type (3) is:
(62) structure based element template formula (4), carries out 2 closing operation of mathematical morphology to image image_dilate, obtains image image_close; Described structural element template type (4) is:
(63) inverse process is carried out to image image_close, background extraction connected region image image_bg;
(64) structure based element template formula (3), carries out image_close
secondary morphological erosion computing, obtains prospect connected region image image_obj.
6. a kind of car plate color determination methods based on connected region information according to claim 5, is characterized in that: in step (7), described optimization prospect connected region image; Specifically comprise the following steps:
(71) the current pending row of image image_obj is entered;
(72) utilize following formula, find whole left trip points and left hand edge point;
Wherein, g (x, y) is image intensity value;
(73) based on current transition point, foreground points all in the n-1 neighborhood of right side is retained;
(74) judge that whether current transition point is last trip point of this row; If so, step (75) is then performed; If not, then enter next trip point, return and perform step (73);
(75) judge that whether current line is last column of image image_obj; If so, then export and optimize foreground region image; If not, then step (76) is performed;
(76) enter the next line of image image_obj, return and perform step (71)-(75);
After being finished, according to step (71)-(76), find right hand edge point and corresponding foreground area retention point, superpose based on left hand edge reserve area with based on right hand edge reserve area, obtain final optimization foreground area.
7. a kind of car plate color determination methods based on connected region information according to claim 1, it is characterized in that: in step (9), the described gray average corresponding based on prospect connected region and gray average corresponding to background connected region, judge car plate color type; The following formula of concrete employing realizes:
Wherein,
the gray average that expression prospect connected region is corresponding,
represent the gray average that background connected region is corresponding, class=1 represents that car plate belongs to the bright word type in the dark end, and class=0 represents that car plate belongs to dark word type of putting one's cards on the table.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510013056.5A CN104573656B (en) | 2015-01-09 | 2015-01-09 | A kind of car plate color determination methods based on connected region information |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510013056.5A CN104573656B (en) | 2015-01-09 | 2015-01-09 | A kind of car plate color determination methods based on connected region information |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104573656A true CN104573656A (en) | 2015-04-29 |
CN104573656B CN104573656B (en) | 2018-02-02 |
Family
ID=53089678
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510013056.5A Active CN104573656B (en) | 2015-01-09 | 2015-01-09 | A kind of car plate color determination methods based on connected region information |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104573656B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104966107A (en) * | 2015-07-10 | 2015-10-07 | 安徽清新互联信息科技有限公司 | Credit card card-number identification method based on machine learning |
CN107844761A (en) * | 2017-10-25 | 2018-03-27 | 海信集团有限公司 | The detection method and device of traffic sign |
CN108205675A (en) * | 2016-12-20 | 2018-06-26 | 浙江宇视科技有限公司 | The processing method and equipment of a kind of license plate image |
CN111860539A (en) * | 2020-07-20 | 2020-10-30 | 济南博观智能科技有限公司 | License plate color recognition method, device and medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040234105A1 (en) * | 2003-05-19 | 2004-11-25 | Quen-Zong Wu | Method of automatic vehicle licence plate recognition by color discrimination and reorganization |
CN101408942A (en) * | 2008-04-17 | 2009-04-15 | 浙江师范大学 | Method for locating license plate under a complicated background |
CN101872416A (en) * | 2010-05-06 | 2010-10-27 | 复旦大学 | Vehicle license plate recognition method and system of road image |
CN102214290A (en) * | 2010-04-12 | 2011-10-12 | 无锡科利德斯科技有限公司 | License plate positioning method and license plate positioning template training method |
CN102254152A (en) * | 2011-06-17 | 2011-11-23 | 东南大学 | License plate location method based on color change points and color density |
-
2015
- 2015-01-09 CN CN201510013056.5A patent/CN104573656B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040234105A1 (en) * | 2003-05-19 | 2004-11-25 | Quen-Zong Wu | Method of automatic vehicle licence plate recognition by color discrimination and reorganization |
CN101408942A (en) * | 2008-04-17 | 2009-04-15 | 浙江师范大学 | Method for locating license plate under a complicated background |
CN102214290A (en) * | 2010-04-12 | 2011-10-12 | 无锡科利德斯科技有限公司 | License plate positioning method and license plate positioning template training method |
CN101872416A (en) * | 2010-05-06 | 2010-10-27 | 复旦大学 | Vehicle license plate recognition method and system of road image |
CN102254152A (en) * | 2011-06-17 | 2011-11-23 | 东南大学 | License plate location method based on color change points and color density |
Non-Patent Citations (2)
Title |
---|
丛露微等: "一种基于笔画特征的车牌前景背景分割新方法", 《电子测试》 * |
吴德会等: "基于车牌特征颜色相似度的定位方法", 《公路交通科技》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104966107A (en) * | 2015-07-10 | 2015-10-07 | 安徽清新互联信息科技有限公司 | Credit card card-number identification method based on machine learning |
CN108205675A (en) * | 2016-12-20 | 2018-06-26 | 浙江宇视科技有限公司 | The processing method and equipment of a kind of license plate image |
CN107844761A (en) * | 2017-10-25 | 2018-03-27 | 海信集团有限公司 | The detection method and device of traffic sign |
CN107844761B (en) * | 2017-10-25 | 2021-08-10 | 海信集团有限公司 | Traffic sign detection method and device |
CN111860539A (en) * | 2020-07-20 | 2020-10-30 | 济南博观智能科技有限公司 | License plate color recognition method, device and medium |
CN111860539B (en) * | 2020-07-20 | 2024-05-10 | 济南博观智能科技有限公司 | License plate color recognition method, device and medium |
Also Published As
Publication number | Publication date |
---|---|
CN104573656B (en) | 2018-02-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103136845B (en) | A kind of Renminbi false distinguishing method based on crown word number characteristics of image | |
CN101398894B (en) | Automobile license plate automatic recognition method and implementing device thereof | |
CN102968646B (en) | A kind of detection method of license plate based on machine learning | |
CN103824066B (en) | A kind of licence plate recognition method based on video flowing | |
CN107273896A (en) | A kind of car plate detection recognition methods based on image recognition | |
CN103268489B (en) | Automotive number plate recognition methods based on sliding window search | |
CN107729899A (en) | A kind of number-plate number recognition methods and device | |
CN106096610A (en) | A kind of file and picture binary coding method based on support vector machine | |
CN103136528B (en) | A kind of licence plate recognition method based on dual edge detection | |
CN106096601A (en) | The method and system of character types in a kind of automatic detection bill | |
CN102254152B (en) | License plate location method based on color change points and color density | |
CN106650553A (en) | License plate recognition method and system | |
CN107103317A (en) | Fuzzy license plate image recognition algorithm based on image co-registration and blind deconvolution | |
CN104573656A (en) | License plate color judging method based on connected region information | |
CN103824091B (en) | A kind of licence plate recognition method for intelligent transportation system | |
CN102880863B (en) | Method for positioning license number and face of driver on basis of deformable part model | |
CN104200561A (en) | Method for identifying character-stained RMB based on texture features | |
CN102800148A (en) | RMB sequence number identification method | |
CN104751142A (en) | Natural scene text detection algorithm based on stroke features | |
CN102043950A (en) | Vehicle outline recognition method based on canny operator and marginal point statistic | |
CN102024144A (en) | Container number identification method | |
CN107832762A (en) | A kind of License Plate based on multi-feature fusion and recognition methods | |
CN104598905A (en) | License plate positioning method and device | |
CN104751466A (en) | Deform able object tracking algorithm based on visual salience and system thereof | |
CN107346580A (en) | Billing information recognition methods and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |