CN102819728A - Traffic sign detection method based on classification template matching - Google Patents
Traffic sign detection method based on classification template matching Download PDFInfo
- Publication number
- CN102819728A CN102819728A CN2012102457741A CN201210245774A CN102819728A CN 102819728 A CN102819728 A CN 102819728A CN 2012102457741 A CN2012102457741 A CN 2012102457741A CN 201210245774 A CN201210245774 A CN 201210245774A CN 102819728 A CN102819728 A CN 102819728A
- Authority
- CN
- China
- Prior art keywords
- image
- traffic sign
- template
- classification
- sign detection
- 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
Landscapes
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to the technical field of machine vision and image processing and in particular relates to a traffic sign detection method based on classification template matching. The problems that the conventional traffic sign identification method is complex in operation process, low in real-time property and high in light and image quality requirements are solved. The method comprises the following steps of: segmenting areas containing traffic signs in a photographed image according to different color areas, namely classifying the traffic signs according to colors; (2) screening communicating areas in the image through the shape and area characteristics after the color classification, and positioning the sign area; and (3) identifying through a template matching method. The method has high operability and is insensitive to the light change, and when a source image is blurred, a high identification accuracy rate can be guaranteed, and the method is high in real-time property and can be used for an automatic traffic sign detection device on a vehicle.
Description
Technical field
The invention belongs to machine vision and technical field of image processing, be specifically related to a kind of method for traffic sign detection based on the classification model coupling.
Background technology
Along with urban transport problems is serious day by day, intelligent transport technology becomes current research focus and development trend.Intelligent transport technology is that the safety traffic problem of traffic jam issue and vehicle provides huge help.Automatic detection recognition technology for traffic sign is a gordian technique of intelligent transportation field.
Current method for traffic sign detection mainly contains through the identification of classification and the identification through the conventional template matching process.Characteristics such as classifying identification method color, shape combine the neural network scheduling algorithm to segment step by step, finally can confirm it is which kind of sign; Traditional template matching method is that subimage is moved and calculate correlativity to find out the maximum zone of correlativity, perhaps in all templates, searching for to find immediate template after the mark region Primary Location in original image.It is better relatively that the method for simple use classification is carried out the Traffic Sign Recognition real-time; But to analyze the nuance of various traffic sign shape colors through the method that classification is discerned; Classify layer by layer according to general character and individual character; Operating process is very complicated, need a large amount of debugging, and recognition effect is often unsatisfactory.Traditional template matching method calculated amount is bigger, and real-time is relatively poor, and is often higher to hardware requirement in actual traffic sign automatic identification equipment.
In addition, current traffic sign recognition method is had relatively high expectations to color and light, when the two changes, can produce considerable influence to detecting identification.When the movement velocity of the relative traffic sign of camera was very fast, image can thicken, and use traditional recognition methods accuracy rate also can reduce greatly this moment.
Summary of the invention
Technical matters to be solved by this invention is existing traffic sign recognition method complicated operating process, real-time is relatively poor and light and picture quality are required height, and a kind of method for traffic sign detection based on the classification model coupling is provided.
The technical scheme that the present invention adopted is:
A kind of method for traffic sign detection based on the classification model coupling comprises the steps:
(1) zone of containing traffic sign in the photographic images is cut apart by regions of different colours, be about to traffic sign and press color classification;
(2) through shape or area features the connected region in the image is screened behind the color classification, locate this mark region;
(3) discern through the method for template matches.
Aforesaid a kind of method for traffic sign detection based on the classification model coupling, wherein: said step (3) specifically is divided into:
(3.1) use dynamic method for normalizing sign image to be normalized to the target image of m * n: to connect rectangle in the middle mark region of construction step (2), and carry out binary conversion treatment through bimodal method; Utilize bianry image further to confirm the boundary rectangle of mark region internal symbol.
(3.2) template matches is calculated.
Aforesaid a kind of method for traffic sign detection based on classification model coupling, wherein: with the image normalization in the symbol circumscribed rectangular region described in the step (3.1) is the image with the template same size; During step (3.2) coupling; With all templates in the respective color classification respectively with normalization after image carry out matching operation, establishing picture size to be matched is J * K, w (s; T) and f (s; T) be respectively certain pixel value of template image and image to be matched, (x, expression formula y) does the response results c of then final related operation
According to actual conditions set a c (x, threshold value y), if be higher than this threshold value then this image be the traffic sign of corresponding kind.
Aforesaid a kind of method for traffic sign detection based on the classification model coupling; Wherein: the method for normalizing of said step (3.1) is: the wide and height of establishing source images is respectively x, y; The characteristic image resolution of extracting is m * n; Then will source images be divided into m * n zone, it is following that row is cut apart computing formula:
A. as if x%m=0, then each regional every row of source images should contain x/m pixel
B. as if x%m ≠ 0, the every row in preceding x%m zone contains x/m+1 point; The every row in back m-x%m zone contains the point of x/m;
Column split is calculated and row is cut apart calculating in like manner, detects the number of the black pixel point in each zone, if the stain number thinks then that greater than certain fixing empirical value this zone is a black, the target image corresponding point are changed to stain.
Aforesaid a kind of method for traffic sign detection based on the classification model coupling, wherein: when said step (1) is carried out color classification, adopt following steps:
The rgb image is converted into the hsv image, establishes that (r, g b) are red, the green and blue coordinate of a color respectively; If max is equivalent to r, the maximum among g and the b; If min equals r, the reckling among g and the b; Find in the HSV space (h, s, v) value, h ∈ [0,360) be the hue angle of angle, and s, v ∈ [0,1] is saturation degree and brightness, computing formula is:
v＝max
。
Aforesaid a kind of method for traffic sign detection based on the classification model coupling, wherein: shape facility comprises girth, circularity, inscribed circle radius or complex shape property described in the step (2).
Aforesaid a kind of method for traffic sign detection based on the classification model coupling; Wherein: also comprise the step of extracting template according to the actual traffic sign image: aim at the traffic sign that will extract template with camera; Carry out step (1)～(3) and in image, catch said traffic sign; Observe and confirm to catch the corresponding two-value array of normalized image that obtains in the correct back output step (3), this two-value array file is template.
The invention has the beneficial effects as follows:
(1) the present invention proposes a kind of classification of location earlier; The traffic sign recognition method of identification again, minimizing at double the template matches in expensive source calculate, through classification and location step by step; Pending zone is dwindled; Thereby significantly reduced interfere information, this time domain acceptance of the bid will and background have very high contrast, have improved real-time and accuracy; And this method has stronger operability, and is insensitive to the light variation, when source images is fuzzy, still can guarantee higher recognition accuracy, and real-time is better, can be used for the traffic sign automatic detection device on the vehicle.
(2) the present invention is cut apart through color the sign in the image is classified; Screen the location through characteristics such as shape area; Only in subclass, search for when coupling is calculated, and use less template, these improve the calculated amount of the template matches that has reduced expensive source greatly; Than ordinary traffic sign higher real-time and accuracy are arranged, be suitable for real-time demanding equipment and use.
(3) can set color, brightness and the saturation degree scope of broad at minute time-like; During the location with the shape condition establish a little less than; Much more as far as possible must keep traffic sign information, leave final accurate identification work for template matches and carry out, can adapt to the variation of various light like this; Use bimodal method binaryzation when confirming the symbol area in the traffic sign in addition, adapt to various light and change.Two improvement above comprehensive make the present invention lower to light and picture quality requirement than commonsense method, the environment that equipment in being fit to move and light often change.
(4) improved conventional method for normalizing, to the source images ranks value different with target image, can the dynamic calculation source images how the zoning is the most reasonable.
(5) use a cover simple and reliable template extraction method, easy operating can or be revised ATL according to the different situation increase and decrease.
Description of drawings
Fig. 1 is a kind of classification, location and an identification step block diagram based on the traffic sign in the method for traffic sign detection of classification model coupling provided by the invention;
Fig. 2 is the selected example of mark region;
Fig. 3 is speed limit 40 sign template examples;
Fig. 4 is a right-hand rotation Warning Mark template example.
Embodiment
Below in conjunction with accompanying drawing and embodiment a kind of method for traffic sign detection based on the classification model coupling provided by the invention is introduced:
As shown in Figure 1, a kind of method for traffic sign detection based on the classification model coupling comprises the steps:
(1) regions of different colours of area-of-interest is cut apart, traffic sign is pressed color classification.
Also possibly comprise information such as highway landscape owing to not only comprise traffic sign usually in the photographic images, but effective discernible traffic sign always appears at some zones of photographic images, therefore to the image setting area-of-interest; And sign commonly used can be divided into red prohibitory sign, blue Warning Mark, yellow warning notice and black designation (as removing speed(-)limit sign) etc. by color, can carry out the classification of the first step through color successively.
When carrying out color classification, existing model below normal employing the: YUV model, YIQ model, Lab model etc.For better classification, the present invention preferably uses a kind of fast conversion method.The color of traffic sign is distinguished more obvious in the HSV space, therefore will the rgb image be converted into the hsv image, establishes that (r, g are respectively red, the green and blue coordinates of a color b), and their value is the real number between 0 to 1.If max is equivalent to r, the maximum among g and the b.If min equals the reckling in these values.Find in the HSV space (h, s, v) value, the h ∈ here [0,360) be the hue angle of angle, and s, v ∈ [0,1] is saturation degree and brightness, computing formula is:
v＝max
H is a tone, every kind of all corresponding h value of color; S is a saturation degree, and the gay colours saturation degree is high, and the white saturation degree is near zero; V is brightness, and this value and light intensity relation is big.Pre-service is for the zone of each color of reservation as much as possible, for versicolor h, s, the v interval of all setting broad of cutting apart.
(2) location of traffic sign
Through characteristics such as shape and areas the connected region in the image is screened the location traffic sign behind the color classification.As shown in Figure 1, red and black split image is then searched for the ring shape zone, and blueness is cut apart the circular fill area of picture search, yellow image search delta-shaped region.
The present invention uses the element marking method to carry out the search of connected region, and travels through the zone that all obtain, and judges whether it possibly is traffic sign according to provincial characteristics, if then draw this mark region of boundary rectangle location of this connected region.
Can use following one or more characteristic when confirming shape:
A. area S: area is one of essential characteristic of rendering image, and the area S of image can represent with the number of pixels that is comprised in the same marked region.
B. perimeter L
The perimeter L of image is with representing apart from sum between adjacent two pixels of outer boundary on the image.
C. circularity R
o, inscribed circle radius r and complex shape property e
Circularity R
oBe used for describing the scenery shape near circular degree, its computing formula does
S is a graphics area in the formula, and L is the figure girth, R
oThe scope of value is 0≤R
o≤1, R
oMore greatly then figure approaches circle more, is example with continuous circle, square, equilateral triangle, calculates their circularity R
oFor: circle, R
o=1; Square, R
o=0.79; Equilateral triangle, R
o=0.6.
Inscribed circle radius r representes with following formula:
The implication of S and L is the same in the formula.Equally, be that example result of calculation is: circle, r=R (R is a radius of a circle) with three typical row graphs; Square,
(a is the foursquare length of side); Equilateral triangle,
(a is the equilateral triangle length of side).
Complex-shaped sex index dispersion index e commonly used representes that its calculating formula is:
This formula has been described the girth size of unit area figure, and the e value is big, shows the Zhou Changda of unit area, and promptly figure is discrete, then is complex figure; Otherwise, then be simple graph.The minimum figure of e value is circular.The result of calculation of typical row graph is: circle, e=12.6; Equilateral triangle, e=20.8; Square, e=16.0.
Judge when circular that e is generally 11.6～13.6 and R
oBe 0.9～1.1; Needing e when judging triangle is 18.8～22.8 and R
oBe 0.5～0.7.
(3) discern through the method for template matches
The first step uses dynamic method for normalizing sign image to be normalized to the target image of m * n.
With blueness right-hand rotation Warning Mark is example, and preceding step has been confirmed the boundary rectangle in traffic sign zone, with its called after zone 1; Shown in Fig. 2 dotted line; This rectangle is scaled, make rectangle become the sign circle in connect rectangle, be called the zone 2; To remove the outer background interference of sign, shown in Fig. 2 solid line.
Image in the zone 2 is carried out binary conversion treatment through bimodal method (also can select other existing binarization methods for use; For example: clustering procedure, process of iteration, P quantile method, big Tianjin method etc.); Because the contrast of image is stronger in this time domain 2; Bimodal method binaryzation can be partitioned into arrow and the background in the traffic sign well, and this method changes insensitive to light.Further confirm the boundary rectangle of symbol through bianry image, like the zone in Fig. 2 dot-and-dash line 3.
With the image normalizations of zone in 3 is the image with the template same size.Can use existing method for normalizing, but, the present invention proposes the method for a kind of dynamic setting normalization unit in order to express image information more accurately.If the wide and height of source images is respectively x, y, the characteristic image resolution that extract is m * n, then will source images be divided into m * n zone, and it is following that row is cut apart computing formula:
A. as if x%m=0, then each regional every row of source images should contain x/m pixel
B. as if x%m ≠ 0, the every row in preceding x%m zone contains x/m+1 point; The every row in back m-x%m zone contains the point of x/m.
Column split is calculated and row is cut apart compute classes seemingly; Original image is similar to average m * n the zone that be divided into the most at last, detects the number of the black pixel point in each zone, if the stain number is greater than certain empirical value of fixing; Think that then this zone is a black, the target image corresponding point are changed to stain.Can be through top processing with the target image of source images normalization m * n.
In second step, carry out template matches and calculate.
With all templates in the respective color classification respectively with normalization after image carry out matching operation, establishing picture size to be matched is J * K, w (s; T) and f (s; T) be respectively certain pixel value of template image and image to be matched, (x, expression formula y) does the response results c of then final related operation
(x, threshold value y) think that then this image is the traffic sign of corresponding kind if be higher than this threshold value to set a c according to actual conditions.
(4) extract template according to the actual traffic sign image.
Key step is: camera is aimed at the traffic sign that will extract template; Operation (1)～(3) step is caught this traffic sign in image; Observe and confirm to catch the corresponding two-value array of normalized image that obtains in the correct back output (3), this two-value array file is template.The template array with the image format display effect like Fig. 3, shown in 4.
Claims (7)
1. the method for traffic sign detection based on the classification model coupling comprises the steps:
(1) zone of containing traffic sign in the photographic images is cut apart by regions of different colours, be about to traffic sign and press color classification;
(2) through shape or area features the connected region in the image is screened behind the color classification, locate this mark region;
(3) discern through the method for template matches.
2. a kind of method for traffic sign detection according to claim 1 based on the classification model coupling, it is characterized in that: said step (3) specifically is divided into:
(3.1) use dynamic method for normalizing sign image to be normalized to the target image of m * n: to connect rectangle in the middle mark region of construction step (2), and carry out binary conversion treatment through bimodal method; Utilize bianry image further to confirm the boundary rectangle of mark region internal symbol.
(3.2) template matches is calculated.
3. a kind of method for traffic sign detection based on classification model coupling according to claim 2 is characterized in that: with the image normalization in the symbol circumscribed rectangular region described in the step (3.1) for the image of template same size; During step (3.2) coupling; With all templates in the respective color classification respectively with normalization after image carry out matching operation, establishing picture size to be matched is J * K, w (s; T) and f (s; T) be respectively certain pixel value of template image and image to be matched, (x, expression formula y) does the response results c of then final related operation
According to actual conditions set a c (x, threshold value y), if be higher than this threshold value then this image be the traffic sign of corresponding kind.
4. a kind of method for traffic sign detection according to claim 3 based on the classification model coupling; It is characterized in that: the method for normalizing of said step (3.1) is: the wide and height of establishing source images is respectively x, y; The characteristic image resolution of extracting is m * n; Then will source images be divided into m * n zone, it is following that row is cut apart computing formula:
A. as if x%m=0, then each regional every row of source images should contain x/m pixel
B. as if x%m ≠ 0, the every row in preceding x%m zone contains x/m+1 point; The every row in back m-x%m zone contains the point of x/m;
Column split is calculated and row is cut apart calculating in like manner, detects the number of the black pixel point in each zone, if the stain number thinks then that greater than certain fixing empirical value this zone is a black, the target image corresponding point are changed to stain.
5. a kind of method for traffic sign detection based on the classification model coupling according to claim 1 and 2 is characterized in that: when said step (1) is carried out color classification, adopt following steps:
The rgb image is converted into the hsv image, establishes that (r, g b) are red, the green and blue coordinate of a color respectively; If max is equivalent to r, the maximum among g and the b; If min equals r, the reckling among g and the b; Find in the HSV space (h, s, v) value, h ∈ [0,360) be the hue angle of angle, and s, v ∈ [0,1] is saturation degree and brightness, computing formula is:
v＝max
6. a kind of method for traffic sign detection based on the classification model coupling according to claim 1 and 2, it is characterized in that: shape facility comprises girth, circularity, inscribed circle radius or complex shape property described in the step (2).
7. according to the described a kind of method for traffic sign detection of claim 2～6 based on the classification model coupling; It is characterized in that: also comprise the step of extracting template according to the actual traffic sign image: aim at the traffic sign that will extract template with camera; Carry out step (1)～(3) and in image, catch said traffic sign; Observe and confirm to catch the corresponding two-value array of normalized image that obtains in the correct back output step (3), this two-value array file is template.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012102457741A CN102819728A (en) | 2012-07-17 | 2012-07-17 | Traffic sign detection method based on classification template matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012102457741A CN102819728A (en) | 2012-07-17 | 2012-07-17 | Traffic sign detection method based on classification template matching |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102819728A true CN102819728A (en) | 2012-12-12 |
Family
ID=47303837
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012102457741A Pending CN102819728A (en) | 2012-07-17 | 2012-07-17 | Traffic sign detection method based on classification template matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102819728A (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390167A (en) * | 2013-07-18 | 2013-11-13 | 奇瑞汽车股份有限公司 | Multi-characteristic layered traffic sign identification method |
| CN103413124A (en) * | 2013-08-19 | 2013-11-27 | 中国科学院自动化研究所 | Method for detecting round traffic sign |
| CN104463105A (en) * | 2014-11-19 | 2015-03-25 | 深圳市腾讯计算机系统有限公司 | Guide board recognizing method and device |
| CN104937637A (en) * | 2013-07-31 | 2015-09-23 | Plk科技株式会社 | Vehicular image recognition system for display-board recognition |
| CN105303189A (en) * | 2014-07-29 | 2016-02-03 | 阿里巴巴集团控股有限公司 | Method and device for detecting specific identification image in predetermined area |
| CN105488160A (en) * | 2015-11-30 | 2016-04-13 | 北大方正集团有限公司 | Picture hitching method and device, manufacturing method of mapping knowledge domain |
| CN105550632A (en) * | 2014-10-23 | 2016-05-04 | 现代摩比斯株式会社 | Traffic sign recognizing apparatus and operating method thereof |
| CN106096587A (en) * | 2016-06-29 | 2016-11-09 | 韦醒妃 | There is the vehicle monitoring system of target recognition function |
| CN106295644A (en) * | 2016-08-11 | 2017-01-04 | 海信集团有限公司 | Symbol Recognition and device |
| CN106570494A (en) * | 2016-11-21 | 2017-04-19 | 北京智芯原动科技有限公司 | Traffic signal lamp recognition method and device based on convolution neural network |
| CN107203767A (en) * | 2017-05-24 | 2017-09-26 | 深圳市神视检验有限公司 | A kind of method and device for aiding in realizing automation Non-Destructive Testing |
| CN107218889A (en) * | 2017-05-24 | 2017-09-29 | 深圳市神视检验有限公司 | A kind of method and device of measurement position |
| CN109214434A (en) * | 2018-08-20 | 2019-01-15 | 上海萃舟智能科技有限公司 | A kind of method for traffic sign detection and device |
| CN110414511A (en) * | 2019-07-30 | 2019-11-05 | 深圳市普渡科技有限公司 | Cooperate sign and system |
| CN110674863A (en) * | 2019-09-19 | 2020-01-10 | 北京迈格威科技有限公司 | Hamming code identification method and device and electronic equipment |
| CN111178360A (en) * | 2019-12-09 | 2020-05-19 | 武汉光庭信息技术股份有限公司 | Traffic sign detection positioning method and system |
| CN111723625A (en) * | 2019-03-22 | 2020-09-29 | 上海海拉电子有限公司 | Traffic light image recognition processing method and device, auxiliary traffic system and storage medium |
| CN116524725A (en) * | 2023-07-03 | 2023-08-01 | 钧捷科技（北京）有限公司 | Intelligent driving traffic sign image data identification system |
| CN117422714A (en) * | 2023-12-18 | 2024-01-19 | 大陆汽车电子(济南)有限公司 | Assembly inspection method, apparatus, and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101404117A (en) * | 2008-10-21 | 2009-04-08 | 东软集团股份有限公司 | Traffic sign recognition method and device |
| CN101702197A (en) * | 2009-10-26 | 2010-05-05 | 宁波大学 | Method for detecting road traffic signs |
| WO2011154978A2 (en) * | 2010-06-10 | 2011-12-15 | Tata Consultancy Services Limited | An illumination invariant and robust apparatus and method for detecting and recognizing various traffic signs |
| CN202084185U (en) * | 2011-06-16 | 2011-12-21 | 长安大学 | Automatic traffic sign identification device |
| CN102542260A (en) * | 2011-12-30 | 2012-07-04 | 中南大学 | Method for recognizing road traffic sign for unmanned vehicle |
-
2012
- 2012-07-17 CN CN2012102457741A patent/CN102819728A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101404117A (en) * | 2008-10-21 | 2009-04-08 | 东软集团股份有限公司 | Traffic sign recognition method and device |
| CN101702197A (en) * | 2009-10-26 | 2010-05-05 | 宁波大学 | Method for detecting road traffic signs |
| WO2011154978A2 (en) * | 2010-06-10 | 2011-12-15 | Tata Consultancy Services Limited | An illumination invariant and robust apparatus and method for detecting and recognizing various traffic signs |
| CN202084185U (en) * | 2011-06-16 | 2011-12-21 | 长安大学 | Automatic traffic sign identification device |
| CN102542260A (en) * | 2011-12-30 | 2012-07-04 | 中南大学 | Method for recognizing road traffic sign for unmanned vehicle |
Non-Patent Citations (2)
| Title |
|---|
| 李伦波等: "自然场景下交通标志检测与分类算法研究", 《哈尔滨工业大学学报》, vol. 41, no. 11, 30 November 2009 (2009-11-30) * |
| 许少秋: "户外交通标志检测和形状识别", 《中国图象图形学报》, vol. 14, no. 4, 30 April 2009 (2009-04-30), pages 707 - 711 * |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390167A (en) * | 2013-07-18 | 2013-11-13 | 奇瑞汽车股份有限公司 | Multi-characteristic layered traffic sign identification method |
| CN104937637A (en) * | 2013-07-31 | 2015-09-23 | Plk科技株式会社 | Vehicular image recognition system for display-board recognition |
| CN103413124A (en) * | 2013-08-19 | 2013-11-27 | 中国科学院自动化研究所 | Method for detecting round traffic sign |
| CN103413124B (en) * | 2013-08-19 | 2016-09-21 | 中国科学院自动化研究所 | A kind of circular traffic sign detection method |
| CN105303189A (en) * | 2014-07-29 | 2016-02-03 | 阿里巴巴集团控股有限公司 | Method and device for detecting specific identification image in predetermined area |
| US10885644B2 (en) | 2014-07-29 | 2021-01-05 | Banma Zhixing Network (Hongkong) Co., Limited | Detecting specified image identifiers on objects |
| US10360689B2 (en) | 2014-07-29 | 2019-07-23 | Alibaba Group Holding Limited | Detecting specified image identifiers on objects |
| CN105550632A (en) * | 2014-10-23 | 2016-05-04 | 现代摩比斯株式会社 | Traffic sign recognizing apparatus and operating method thereof |
| CN105550632B (en) * | 2014-10-23 | 2019-02-15 | 现代摩比斯株式会社 | Traffic mark board identification device and its working method |
| CN104463105B (en) * | 2014-11-19 | 2017-11-03 | 清华大学 | Guideboard recognition methods and device |
| CN104463105A (en) * | 2014-11-19 | 2015-03-25 | 深圳市腾讯计算机系统有限公司 | Guide board recognizing method and device |
| CN105488160A (en) * | 2015-11-30 | 2016-04-13 | 北大方正集团有限公司 | Picture hitching method and device, manufacturing method of mapping knowledge domain |
| CN106096587A (en) * | 2016-06-29 | 2016-11-09 | 韦醒妃 | There is the vehicle monitoring system of target recognition function |
| CN106295644A (en) * | 2016-08-11 | 2017-01-04 | 海信集团有限公司 | Symbol Recognition and device |
| CN106295644B (en) * | 2016-08-11 | 2020-06-09 | 海信集团有限公司 | Symbol identification method and device |
| CN106570494A (en) * | 2016-11-21 | 2017-04-19 | 北京智芯原动科技有限公司 | Traffic signal lamp recognition method and device based on convolution neural network |
| CN107218889A (en) * | 2017-05-24 | 2017-09-29 | 深圳市神视检验有限公司 | A kind of method and device of measurement position |
| CN107203767A (en) * | 2017-05-24 | 2017-09-26 | 深圳市神视检验有限公司 | A kind of method and device for aiding in realizing automation Non-Destructive Testing |
| CN109214434A (en) * | 2018-08-20 | 2019-01-15 | 上海萃舟智能科技有限公司 | A kind of method for traffic sign detection and device |
| CN111723625A (en) * | 2019-03-22 | 2020-09-29 | 上海海拉电子有限公司 | Traffic light image recognition processing method and device, auxiliary traffic system and storage medium |
| CN111723625B (en) * | 2019-03-22 | 2023-08-08 | 上海海拉电子有限公司 | Traffic light image recognition processing method and device, auxiliary traffic system and storage medium |
| CN110414511A (en) * | 2019-07-30 | 2019-11-05 | 深圳市普渡科技有限公司 | Cooperate sign and system |
| CN110414511B (en) * | 2019-07-30 | 2022-05-03 | 深圳市普渡科技有限公司 | Cooperative sign recognition method and system for robot |
| CN110674863A (en) * | 2019-09-19 | 2020-01-10 | 北京迈格威科技有限公司 | Hamming code identification method and device and electronic equipment |
| CN110674863B (en) * | 2019-09-19 | 2022-06-21 | 北京迈格威科技有限公司 | Hamming code identification method and device and electronic equipment |
| CN111178360A (en) * | 2019-12-09 | 2020-05-19 | 武汉光庭信息技术股份有限公司 | Traffic sign detection positioning method and system |
| CN116524725A (en) * | 2023-07-03 | 2023-08-01 | 钧捷科技（北京）有限公司 | Intelligent driving traffic sign image data identification system |
| CN116524725B (en) * | 2023-07-03 | 2023-09-01 | 钧捷科技（北京）有限公司 | Intelligent driving traffic sign image data identification system |
| CN117422714A (en) * | 2023-12-18 | 2024-01-19 | 大陆汽车电子(济南)有限公司 | Assembly inspection method, apparatus, and storage medium |
| CN117422714B (en) * | 2023-12-18 | 2024-03-29 | 大陆汽车电子(济南)有限公司 | Assembly inspection method, apparatus, and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102819728A (en) | Traffic sign detection method based on classification template matching | |
| CN106651872B (en) | Pavement crack identification method and system based on Prewitt operator | |
| CN108108761B (en) | Rapid traffic signal lamp detection method based on deep feature learning | |
| CN108765443B (en) | Sign enhancement processing method for self-adaptive color threshold segmentation | |
| CN105046196B (en) | Front truck information of vehicles structuring output method based on concatenated convolutional neutral net | |
| CN104751142B (en) | A kind of natural scene Method for text detection based on stroke feature | |
| CN103824081B (en) | Method for detecting rapid robustness traffic signs on outdoor bad illumination condition | |
| CN103544484A (en) | Traffic sign identification method and system based on SURF | |
| CN104408449B (en) | Intelligent mobile terminal scene literal processing method | |
| CN109726717B (en) | Vehicle comprehensive information detection system | |
| Zhang et al. | A multi-feature fusion based traffic light recognition algorithm for intelligent vehicles | |
| Le et al. | Real time traffic sign detection using color and shape-based features | |
| CN109215364B (en) | Traffic signal recognition method, system, device and storage medium | |
| CN105354599A (en) | Color recognition method based on improved SLIC super-pixel segmentation algorithm | |
| CN103198315A (en) | License plate character segmentation algorithm based on character outline and template matching | |
| Deb et al. | HSI color based vehicle license plate detection | |
| CN103544480A (en) | Vehicle color recognition method | |
| CN107330365A (en) | Traffic sign recognition method based on maximum stable extremal region and SVM | |
| CN104463138A (en) | Text positioning method and system based on visual structure attribute | |
| CN111462140A (en) | Real-time image instance segmentation method based on block splicing | |
| Chakraborty et al. | Bangladeshi road sign detection based on YCbCr color model and DtBs vector | |
| CN113673541A (en) | Image sample generation method for target detection and application | |
| CN115082776A (en) | Electric energy meter automatic detection system and method based on image recognition | |
| CN110633635A (en) | ROI-based traffic sign board real-time detection method and system | |
| CN107392115B (en) | Traffic sign identification method based on hierarchical feature extraction |
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: 20121212 |