CN102024148A - Method for identifying green mark of taxi - Google Patents
Method for identifying green mark of taxi Download PDFInfo
- Publication number
- CN102024148A CN102024148A CN 201110003065 CN201110003065A CN102024148A CN 102024148 A CN102024148 A CN 102024148A CN 201110003065 CN201110003065 CN 201110003065 CN 201110003065 A CN201110003065 A CN 201110003065A CN 102024148 A CN102024148 A CN 102024148A
- Authority
- CN
- China
- Prior art keywords
- green mark
- green
- mark
- area
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention relates to a method for identifying a green mark of the taxi, relating to the technical field of traffic control computer image process and mode identification, and comprising a camera control module, a movement detecting vehicle snapshot module, a number plate locating module, a green mark locating module and a green mark identifying module, wherein the camera control module is based on an intelligent vehicle information acquisition system. The location rule of the green mark locating module is based on color features and area features of a detecting region, and the color features are judged in an HSV (Hue, Saturation, Value) model space. The identification of the green mark identifying module comprises two stages: carrying out the morphology conversion according to an extracted selecting region of the green mark to obtain a more complete and regular processing region and calculating two circularity form factors of the region according to the fine process result of the first stage, integrating the result of the two circularity form factors to obtain the final identification according to the judgment rule, and outputting the identification result.
Description
Technical field
The present invention relates to Computer Image Processing and mode identification technology, especially utilize Flame Image Process and mode identification technology taxi to be carried out a kind of taxi green mark recognition methods that detects automatically.
Background technology
In the Chinese Urbanization's process, Traffic Development is rapid, and taxi replenishes as the useful of public transit system, is bringing into play important effect in resident trip.Simultaneously, the problem of management for taxi also shows especially out day by day.For example do not park in accordance with regulations, do not observe operating time and location regulation, not according to indicating on-board and off-board etc.These acts of violating regulations are not only brought confusion to communications and transportation, and exist the hidden danger of the security of the lives and property that jeopardizes the passenger at any time.Therefore, strengthening the standardized administration for taxi, is the instant task of present urban traffic control.
For the management of taxi, in the prior art according to the different multiple monitoring means that need.At present based on vehicle license plate, car mark, the vehicle identification management of vision, it is low to have cost, preserves the vehicle pictures information completely, and therefore anti-interference multiple advantage such as strong, has obtained using widely in field of traffic control.For the management of taxi, equally also can use vision technique to come the distinctive mark of taxi is discerned, and cooperate ripe license plate recognition technology, both combine, and reach the accurate identification to taxi.For example the taxi in Shenzhen is for the ease of unified management, position above car plate increases has installed a semi-circular green mark, so, after capturing vehicle pictures, can carry out image recognition at this feature, recognition result enters background manage software with the car plate identifying information, system can be automatically to whether being the taxi judgement of classifying, form corresponding ladder of management again.Therefore, whether the car plate top has specific green mark for identification, and whether how accurate identification just becomes differentiation is the important means of taxi, also is innovation of the present invention place.Based on the taxi green mark recognition technology of video is a gordian technique in the intelligent vehicle information acquisition system, is to utilize Flame Image Process and mode identification technology taxi to be carried out one of computer processing system that detects automatically.
According to literature search, not temporary transient at present identification according to the distinctive mark of taxi, and judge whether be the relevant report of taxi, the enforcement of this method can improve according to the different characteristic of the taxi of different cities, to improve the management level with the taxi field that strengthens each city.
Summary of the invention
The objective of the invention is to disclose a kind of based on the video method of the special green mark of Real time identification taxi automatically, in the hope of being applied in the highway traffic control systems such as existing electronic police, public security bayonet socket, for the management of traffic information collection and the behavior of standard taxi provides more strong support.
The technology solution of realization the present invention's purpose is as described below: the recognition methods of a kind of taxi green mark, comprise based on the video camera control module 1 in the intelligent vehicle information acquisition system, motion detection vehicle snapshot module 2, car plate locating module 3, green mark locating module 4, green mark identification module 5; Green mark locating module 4 comprises the steps:
4.1 determine surveyed area according to the car plate position; 4.2 surveyed area color conversion; 4.3 according to the rule detection green pixel; 4.4 the non-green pixel of green is carried out binaryzation; 4.5 green target is carried out closing operation of mathematical morphology; 4.6 green connected component in the surveyed area; 4.7 write down largest connected bulk area and position; 4.8 judge whether locate success by rule; Green mark identification module 5 comprises the steps: cavity, 5.1 horizontal direction fill area; 5.2 the zone is symmetrical mapping downwards; 5.3 five morphology dilation operations; 5.415 * 15 mean filters; 5.5 optimal threshold binaryzation; 5.6 zone boundary line drawing; 5.7 calculate the spherical property of circularity form parameter F S; 5.8 judge whether be green mark by rule.
Whether whether above-mentioned 4.8 locate the image that successfully is meant according to binaryzation by the rule judgement exists green mark, and the calculating of green mark location, and its step is as follows:
The first step: to be 255 zone carry out morphologic basic closed operation operation with 3 * 3 structural element to the value of image; Closed operation is the morphology operations of first expansion post-etching;
Second step: adopting the connection characteristic of 8 neighborhoods, is that 255 pixel is carried out determining of connected component to value; And obtain outer section rectangle of each connected component;
The 3rd step: calculate the area that each connected component cuts rectangle outward, the position of the connected component of record area maximum and the outer rectangular area that cuts;
The 4th step: decision rule, largest connected external section rectangular area＞green mark region area to be detected/4 are then judged to have green mark, enter the landmark identification step; Otherwise, judge this vehicle redgreen sign, end process.
Above-mentioned green mark locating module 4 is meant that the main foundation of the locating rule of green mark is the color characteristic and the area features in zone to be detected; Described color characteristic is to finish judgement in the HSV model space.
The identification step of above-mentioned green mark identification module 5 is divided into two-layer, and ground floor is meant according to the favored area for the treatment of of the green mark of extracting and carries out certain morphological transformation, obtains pending zone more complete and rule; The second layer is meant the fine processing result according to ground floor, calculates 2 circularity form factors in this zone, and the result of comprehensive 2 circularity form factors according to decision rule, obtains final identification, and the output recognition result.
The present invention comprised generally a kind of automatically, captured in real time moving vehicle and discern the moving image treatment technology of vehicle head distinctive mark.Its concrete scheme is: at first obtain real time video image by video camera system, handle by realtime graphic then and judged whether that vehicle passes through, secondly from image sequence, be partitioned into moving vehicle according to motion feature, then, utilize color and shape facility that green mark is carried out identification decision at last by the Probability Area of vehicle license plate characteristic information location green mark.
Compared with prior art, main innovate point and beneficial effect are on technical scheme in the present invention:
1, propose and realized a kind of detection and recognition methods of taxi green mark, the green mark discrimination reaches 98%, and recognition correct rate reaches 95%, and recognition time was less than 0.5 second; This technology belongs to initiative, and can be applied to the taxi landmark identification system of similar sign.
2, utilization by the identification to green mark, has solved the classification problem of taxi and general vehicle based on the mode of video; Realized intelligent management effectively to taxi.
3, the present invention program need practice highway section and the place that taxi is monitored, and has obtained effect in the supervision and management to taxi, for example Shenzhen gridding traffic control system in the system of deploying to ensure effective monitoring and control of illegal activities.
Description of drawings
Fig. 1 is an embodiment synoptic diagram implementing flow process required for the present invention.
Fig. 2 is green mark locating module 4 program overall procedure synoptic diagram of the present invention.
Fig. 3 is green mark identification module 5 program overall procedure synoptic diagram of the present invention.
Fig. 4～Figure 14 of the present inventionly handles and result's embodiment (figure sheet mode) picture.
Embodiment
Provide embodiment referring to accompanying drawing the present invention is done concrete further specifying.
As can be seen from Figure 1, realization of the present invention is mainly participated in by the five functional module: video camera control module 1, motion detection vehicle snapshot module 2, car plate locating module 3, green mark locating module 4, green mark identification module 5.The the 1st and the 2nd module wherein adopts us existing patented technology to realize, " based on the camera shutter and the gain Comprehensive Control Technology of car plate luminance contrast " that its technology has been applied for referring to us, " based on the method for accurately recognizing high speed mobile vehicle mark of video " two patents.The 3rd car plate locating module adopts the car plate location technology based on texture of ripe at present and widespread use.Its ultimate principle is the video camera control module according to the video that obtains to video camera---capture card carries out interlock control, is met the video image of green mark identification requirement.Motion detection vehicle snapshot module is finished the collection of parity field, and carries out the piece motion calculation according to the parity field image, obtains the kinematic parameter that needs.Detect direction of vehicle movement according to the motion tracking algorithm, capture a complete headstock picture.The car plate locating module detects the car plate position according to the texture variations feature of license plate area continuation character.The present invention focuses on green mark locating module 4, the work of green mark identification module 5.Green mark locating module 4 utilizes the car plate positional information on picture, determine the general location at green mark possibility place.According to the green component information of HSI color space, handle the size of calculating green area by closing operation of mathematical morphology, whether judge has green mark to exist.Its result of determination is divided into the redgreen sign and may has two kinds of green marks, and may there be the accurate position of green mark in output.Green mark identification module 5 is the positional informations that transmit according to locating module, handles based on morphology technology, utilizes the method for shape analysis, a plurality of form factors of zoning, and whether synthetic determination is the accurate recognition result of green mark finally.
Referring to Fig. 2 Fig. 3 as can be known, green mark locating module 4 comprises the steps:
4.1 determine surveyed area according to the car plate position; 4.2 surveyed area color conversion; 4.3 according to the rule detection green pixel; 4.4 the non-green pixel of green is carried out binaryzation; 4.5 green target is carried out closing operation of mathematical morphology; 4.6 green connected component in the surveyed area; 4.7 write down largest connected bulk area and position; 4.8 judge whether locate success by rule.The condition that the location of green mark mainly relies on is its color characteristic and area features.Above the car plate position, the zone to be detected of selected green mark.Select the method in zone to be detected as follows: car plate position (A, B) (position of the car plate upper left corner in image 4.1), the width W of car plate, the height H of car plate; The position in zone to be detected is (A-2H, B) (position of the regional upper left corner to be detected in image 4.1), the width W in zone to be detected, the height 2H in zone to be detected.If A-2H exceeds image range, then this image does not carry out the location identification of green mark.
Color characteristic is to finish judgement in the HSV model space.Coloured image is as follows from the conversion formula that the original RGB model space is transformed into the HSV model space.Wherein R is the original image red component, and G is the original image green component, and B is the original image blue component; H representative color tone pitch, S represents intensity value, the V representative luminance value.R, G, the span of B is 0-255.
V＝(R+G+B)/3 (3)
When S was 0, corresponding colourless, H was nonsensical, and define H this moment is 0.
Determine according to the HSV value of pixel whether this pixel is that green concrete rule and the step of putting is as follows:
The first step: the rgb value of pixel by formula is converted to the HSV value in (1) (2) (3);
Second step: if step 3 is changeed in 40＜V＜160; Otherwise change step 5;
The 3rd step: if step 4 is changeed in 90＜H＜150; Otherwise change step 5;
The 4th step: judge that this pixel is for green;
The 5th step: judge the non-green of this pixel;
Green mark area image to be detected is carried out binary conversion treatment, mark all to be judged to be green pixel be 255, all pixels that are judged to be non-green of mark are 0.
Whether the following basis image of binaryzation exists green mark, and the calculating of green mark location, and concrete calculation procedure is as follows.
The first step: being 255 zone to the value of binary image carries out morphologic basic closed operation operation with 3 * 3 structural element.Closed operation is the morphology operations of first expansion post-etching;
Second step: adopt the connection characteristic of 8 neighborhoods, be that 255 pixel is carried out determining of connected component to value, and obtain outer section rectangle of each connected component;
The 3rd step: calculate the area that each connected component cuts rectangle outward, the position of the connected component of record area maximum and the outer rectangular area that cuts;
The 4th step: decision rule, largest connected external section rectangular area＞green mark region area to be detected/4 are then judged to have green mark, enter the landmark identification step; Otherwise, judge this vehicle redgreen sign, end process.
Green mark identification module 5 comprises the steps: cavity, 5.1 horizontal direction fill area; 5.2 the zone is symmetrical mapping downwards; 5.3 five morphology dilation operations; 5.4 15 * 15 mean filters; 5.5 optimal threshold binaryzation; 5.6 zone boundary line drawing; 5.7 calculate the spherical property of circularity form parameter F S; 5.8 judge whether be green mark by rule.
The identification step of green mark identification module 5 is divided into two-layer, and ground floor is meant according to the favored area for the treatment of of the green mark of extracting and carries out certain morphological transformation, obtains pending zone more complete and rule; The second layer is meant the fine processing result according to ground floor, calculates 2 circularity form factors in this zone, and the result of comprehensive 2 circularity form factors according to decision rule, obtains final identification, and the output recognition result.
The identification step of ground floor is as follows:
The 1st step: the cavity in the fill area, form complete connected component, in the horizontal direction, if pixel A (i, j)=255, pixel B (k, j)=255, the full assignment of pixel between 2 of A, the B is 255 so;
The 2nd step: the zone is carried out symmetrical mapping, the surveyed area that obtains enlarging downwards;
The 3rd step: the surveyed area that enlarges is carried out morphologic dilation operation with 3 * 3 structural element; This dilation operation carries out 5 times continuously, removes obvious irregular part substantially by the zone after handling;
The 4th step: the back image carries out 15 * 15 mean filter, fuzzy region edge to expanding;
The 5th step: the image behind the mean filter is carried out binaryzation with traditional optimal threshold algorithm (big Tianjin method), obtain regional binary image to be detected again; After handling, the border in zone is level and smooth substantially, is convenient to the accurate calculating of follow-up form factor.
The recognition methods and the decision rule of the second layer are as follows:
2 circularity form factors selecting are: circularity form parameter F, spherical property S.
F＝B
2/4πA
Wherein, B is the girth of connected region, and A is the area of connected region.The girth of connected region refers to the length of the boundary line of whole connected region.With 8 neighborhood boundary lines of 4 neighborhood Rule Extraction connected regions, even there is a point different in 4 neighborhoods of a pixel in the image with it, then this pixel is the border, otherwise is exactly interior pixels.During the boundary line length computation, be designated as 1 for the length distance of the path point of independent direction, the length distance of the point on diagonal line is designated as 1.414.The area of connected region is the number sum of this area pixel point.The judgment threshold of circularity form parameter F is 1.25.
S＝r
1/r
2
R wherein
1Represent the connected region inscribed circle radius; r
2Represent connected region circumcircle radius.The center of circle of two circles is all on the center of gravity of connected region.The minimum distance of gained connected region boundary line, front and center of gravity is an inscribed circle radius; Maximum distance is the circumcircle radius.The judgment threshold of spherical property parameter S is 0.75.(X, Y) computing formula is as follows for the center of gravity C of connected region.
X=∑ i/A (i is a connected region pixel horizontal ordinate)
Y=∑ j/A (j is a connected region pixel ordinate)
Comprehensive 2 circularity form factors obtain decision rule: if F≤1.25 and S 〉=0.75 judges that so this connected region is a green mark; Otherwise be non-green mark.Export result of determination at last.
Fig. 4～Figure 14 handles and result's synoptic diagram picture of the present invention to scheme sheet mode.Its region area A=13066, boundary line length B=431, regional barycenter C=(99,77), inscribed circle radius=60, circumcircle radius=69.6, circularity form parameter F=1.13, spherical property S=0.86.
Fig. 4～Figure 14 is in order to further specify the embodiment that this method provides, promptly as stated above picture is carried out the embodiment of treatment step and effect thereof, comprise the determining of zone to be checked, green pixel point extracts the result, closing operation of mathematical morphology result, maximum UNICOM body and cut the Graphic State that rectangle, zone are filled result, downward symmetrical mapping result, 5 morphology dilation operation fruits, 15*15 mean filter result, optimal threshold binaryzation result, regional edge boundary line, regional barycenter inscribed circle radius circumcircle radius outward.Need to prove that therefore, processing procedure is to be undertaken by image pixel because this method is to belong to technical field of image processing, result is all to present in the mode of picture.
Entire work process comprises by the video camera control technology, obtains round-the-clock video image clearly, guarantees that the green mark horizontal resolution is greater than 80 pixels; The utilization motion detection technique is accurately captured vehicle, and the candid photograph rate reaches 99%; The location car plate, and be partitioned into the zone that need carry out green mark identification; Utilize colouring information bonded area size location green mark; Based on morphology technology, green area is handled; Utilize the method for shape analysis, 2 shapes of zoning are because of 4 sons, and synthetic determination is accurately discerned green mark.Discrimination reaches 98%, and recognition correct rate reaches 95%, and recognition time was less than 0.5 second.
Claims (4)
1. taxi green mark recognition methods, comprise based on the video camera control module (1) in the intelligent vehicle information acquisition system, motion detection vehicle snapshot module (2), car plate locating module (3), green mark locating module (4), green mark identification module (5); It is characterized in that described green mark locating module (4) comprises the steps:
(4.1) determine surveyed area according to the car plate position; (4.2) surveyed area color conversion; (4.3) according to the rule detection green pixel; (4.4) the non-green pixel of green is carried out binaryzation; (4.5) green target is carried out closing operation of mathematical morphology; (4.6) green connected component in the surveyed area; (4.7) largest connected bulk area of record and position; (4.8) judge whether locate success by rule; Described green mark identification module 5 comprises the steps: cavity, (5.1) horizontal direction fill area; (5.2) the downward symmetrical mapping in zone; (5.3) 5 morphology dilation operations; (5.4) 15 * 15 mean filters; (5.5) optimal threshold binaryzation; (5.6) zone boundary line drawing; (5.7) calculate the spherical property of circularity form parameter F S; (5.8) judge whether be green mark by rule.
2. a kind of taxi green mark according to claim 1 recognition methods, it is characterized in that, describedly judge by rule whether locate success (4.8) is meant whether the image according to binaryzation exists green mark, and the calculating of green mark location, its step is as follows:
The first step: to be 255 zone carry out morphologic basic closed operation operation with 3 * 3 structural element to the value of image; Closed operation is the morphology operations of first expansion post-etching;
Second step: adopting the connection characteristic of 8 neighborhoods, is that 255 pixel is carried out determining of connected component to value; And obtain outer section rectangle of each connected component;
The 3rd step: calculate the area that each connected component cuts rectangle outward, the position of the connected component of record area maximum and the outer rectangular area that cuts;
The 4th step: decision rule, largest connected external section rectangular area＞green mark region area to be detected/4 are then judged to have green mark, enter the landmark identification step; Otherwise, judge this vehicle redgreen sign, end process.
3. a kind of taxi green mark according to claim 1 recognition methods is characterized in that, described green mark locating module 4 is meant that the locating rule of green mark is color characteristic and the area features according to zone to be detected; Described color characteristic is to finish judgement in the HSV model space.
4. a kind of taxi green mark according to claim 1 recognition methods, it is characterized in that, the identification step of described green mark identification module 5 is divided into two-layer, ground floor is meant according to the favored area for the treatment of of the green mark of extracting and carries out certain morphological transformation, obtains pending zone more complete and rule; The second layer is meant the fine processing result according to ground floor, calculates 2 circularity form factors in this zone, and the result of comprehensive 2 circularity form factors according to decision rule, obtains final identification, and the output recognition result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110003065 CN102024148A (en) | 2011-01-07 | 2011-01-07 | Method for identifying green mark of taxi |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110003065 CN102024148A (en) | 2011-01-07 | 2011-01-07 | Method for identifying green mark of taxi |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102024148A true CN102024148A (en) | 2011-04-20 |
Family
ID=43865429
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110003065 Pending CN102024148A (en) | 2011-01-07 | 2011-01-07 | Method for identifying green mark of taxi |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102024148A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129701A (en) * | 2011-05-03 | 2011-07-20 | 广东省电力设计研究院 | Method and system for processing core photo |
| CN102819738A (en) * | 2012-07-24 | 2012-12-12 | 东南大学 | License plate identification system based on OpenMP multithreading framework |
| CN103065142A (en) * | 2012-12-30 | 2013-04-24 | 信帧电子技术（北京）有限公司 | Automobile logo division method and device |
| CN103065139A (en) * | 2012-12-30 | 2013-04-24 | 信帧电子技术（北京）有限公司 | Location method and device for automobile logos |
| CN103680148A (en) * | 2013-12-19 | 2014-03-26 | 中国科学院自动化研究所 | Method for identifying taxis |
| CN104164843A (en) * | 2014-08-26 | 2014-11-26 | 无锡市恒通智能交通设施有限公司 | Intelligent traffic sign board with filter circuit |
| CN105448095A (en) * | 2014-06-03 | 2016-03-30 | 杭州海康威视数字技术股份有限公司 | Yellow label car detection method and device |
| CN105975970A (en) * | 2016-05-30 | 2016-09-28 | 北京智芯原动科技有限公司 | SVF-based license plate positioning method and apparatus |
| CN109753945A (en) * | 2019-01-16 | 2019-05-14 | 高翔 | Target subject recognition methods, device, storage medium and electronic equipment |
| CN110705441A (en) * | 2019-09-27 | 2020-01-17 | 四川长虹电器股份有限公司 | Pedestrian crossing line image post-processing method and system |
| CN111401369A (en) * | 2020-03-26 | 2020-07-10 | 上海眼控科技股份有限公司 | Vehicle body danger sign detection method, device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080101656A1 (en) * | 2006-10-30 | 2008-05-01 | Thomas Henry Barnes | Method and apparatus for managing parking lots |
| CN101196980A (en) * | 2006-12-25 | 2008-06-11 | 四川川大智胜软件股份有限公司 | Method for accurately recognizing high speed mobile vehicle mark based on video |
| US7409092B2 (en) * | 2002-06-20 | 2008-08-05 | Hrl Laboratories, Llc | Method and apparatus for the surveillance of objects in images |
| CN101630361A (en) * | 2008-12-30 | 2010-01-20 | 北京邮电大学 | Plate number, body color and mark identification-based equipment and plate number, body color and mark identification-based method for identifying fake plate vehicles |
-
2011
- 2011-01-07 CN CN 201110003065 patent/CN102024148A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7409092B2 (en) * | 2002-06-20 | 2008-08-05 | Hrl Laboratories, Llc | Method and apparatus for the surveillance of objects in images |
| US20080101656A1 (en) * | 2006-10-30 | 2008-05-01 | Thomas Henry Barnes | Method and apparatus for managing parking lots |
| CN101196980A (en) * | 2006-12-25 | 2008-06-11 | 四川川大智胜软件股份有限公司 | Method for accurately recognizing high speed mobile vehicle mark based on video |
| CN101630361A (en) * | 2008-12-30 | 2010-01-20 | 北京邮电大学 | Plate number, body color and mark identification-based equipment and plate number, body color and mark identification-based method for identifying fake plate vehicles |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129701A (en) * | 2011-05-03 | 2011-07-20 | 广东省电力设计研究院 | Method and system for processing core photo |
| CN102819738A (en) * | 2012-07-24 | 2012-12-12 | 东南大学 | License plate identification system based on OpenMP multithreading framework |
| CN103065142A (en) * | 2012-12-30 | 2013-04-24 | 信帧电子技术（北京）有限公司 | Automobile logo division method and device |
| CN103065139A (en) * | 2012-12-30 | 2013-04-24 | 信帧电子技术（北京）有限公司 | Location method and device for automobile logos |
| CN103680148A (en) * | 2013-12-19 | 2014-03-26 | 中国科学院自动化研究所 | Method for identifying taxis |
| CN103680148B (en) * | 2013-12-19 | 2016-05-25 | 中国科学院自动化研究所 | A kind of taxi recognition methods |
| CN105448095A (en) * | 2014-06-03 | 2016-03-30 | 杭州海康威视数字技术股份有限公司 | Yellow label car detection method and device |
| CN105448095B (en) * | 2014-06-03 | 2017-11-17 | 杭州海康威视数字技术股份有限公司 | Method and apparatus are surveyed in a kind of yellow mark car test |
| CN104164843A (en) * | 2014-08-26 | 2014-11-26 | 无锡市恒通智能交通设施有限公司 | Intelligent traffic sign board with filter circuit |
| CN105975970A (en) * | 2016-05-30 | 2016-09-28 | 北京智芯原动科技有限公司 | SVF-based license plate positioning method and apparatus |
| CN109753945A (en) * | 2019-01-16 | 2019-05-14 | 高翔 | Target subject recognition methods, device, storage medium and electronic equipment |
| CN109753945B (en) * | 2019-01-16 | 2021-07-13 | 高翔 | Target subject identification method and device, storage medium and electronic equipment |
| CN110705441A (en) * | 2019-09-27 | 2020-01-17 | 四川长虹电器股份有限公司 | Pedestrian crossing line image post-processing method and system |
| CN111401369A (en) * | 2020-03-26 | 2020-07-10 | 上海眼控科技股份有限公司 | Vehicle body danger sign detection method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102024148A (en) | Method for identifying green mark of taxi | |
| CN102722704B (en) | Method and system for recognizing vehicle license plate by integrating video dynamic tracking | |
| CN106652445B (en) | A kind of road traffic accident method of discrimination and device | |
| CN107665603A (en) | A kind of real-time detection method for judging parking stall and taking | |
| US9704060B2 (en) | Method for detecting traffic violation | |
| CN103824066B (en) | A kind of licence plate recognition method based on video flowing | |
| CN110501018B (en) | Traffic sign information acquisition method for high-precision map production | |
| CN103824452B (en) | A kind of peccancy parking detector based on panoramic vision of lightweight | |
| CN102364496B (en) | Method and system for identifying automobile license plates automatically based on image analysis | |
| CN106373426A (en) | Computer vision-based parking space and illegal lane occupying parking monitoring method | |
| US20130243343A1 (en) | Method and device for people group detection | |
| CN102867417B (en) | Taxi anti-forgery system and taxi anti-forgery method | |
| CN109299674B (en) | Tunnel illegal lane change detection method based on car lamp | |
| CN104036262B (en) | A kind of method and system of LPR car plates screening identification | |
| CN105913685A (en) | Video surveillance-based carport recognition and intelligent guide method | |
| CN101692313A (en) | Portable vehicle recognition device base on embedded platform | |
| KR102122859B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| CN103500322A (en) | Automatic lane line identification method based on low-altitude aerial images | |
| CN113822285A (en) | Vehicle illegal parking identification method for complex application scene | |
| CN107578012A (en) | A kind of drive assist system based on clustering algorithm selection sensitizing range | |
| CN103164958B (en) | Method and system for vehicle monitoring | |
| CN108154146A (en) | A kind of car tracing method based on image identification | |
| CN105740836B (en) | A kind of illegal detection method for occupying Emergency Vehicle Lane | |
| CN104978567A (en) | Vehicle detection method based on scenario classification | |
| CN112885108B (en) | Vehicle change detection method and system on parking space based on deep learning algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110420 |