CN105005766A - Vehicle body color identification method - Google Patents
Vehicle body color identification method Download PDFInfo
- Publication number
- CN105005766A CN105005766A CN201510374911.5A CN201510374911A CN105005766A CN 105005766 A CN105005766 A CN 105005766A CN 201510374911 A CN201510374911 A CN 201510374911A CN 105005766 A CN105005766 A CN 105005766A
- Authority
- CN
- China
- Prior art keywords
- color
- vehicle
- image
- pixels
- space
- 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
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
- G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention discloses a vehicle body color identification method. The method comprises the following steps of: a detection unit performing motion detection on input video data; removing interference regions of window glasses and vehicle shadows by using an RGB space maximum value and minimum value method; differentiating colored vehicles and black, white and grey vehicles by calculating ratios of numbers of pixels; and performing identification on the total 7 colors comprising red, orange, yellow, green, cyan, blue, purple by using H space histograms; performing identification on the total 3 colors comprising black, white and grey by using V space histograms and a fan-shaped area based color voting method. According to the present invention, vehicle body colors are identified by using an RGB color channel difference based vehicle color identification strategy, so that interference regions of the vehicle body color identification can be effectively removed, thereby improving the correct rate of vehicle body color identification.
Description
Technical field
The present invention relates to image processing techniques, be specifically related to a kind of body color recognition methods.
Background technology
In current intelligent transportation system, along with vehicle fleet size increases, traffic environment becomes day by day complicated, only vehicle is identified to the needs that can not meet people by car plate.Vehicle color information more easily causes the interest of people, thus makes up because vehicle fake-license, car many boards phenomenon cause the deficiency of Car license recognition, and to identification and the search of vehicle, to improve and to strengthen intelligent transportation system function significant.
At present, mainly contain two kinds of approach to the identification of vehicle color: the first identifies entire vehicle color, the color characteristic extracting interested area of vehicle identifies.First obtain vehicle foreground image by Target Segmentation, then carry out UNICOM's interference region such as regional analysis deletion wheel, reflective mirror etc. and obtain the obvious region of vehicle color.At sorting phase, adopt and color is divided into the types such as black, white, grey, red, yellow, green, blue based on the two layers of classified device of support vector machine, but the method is easily subject to the interference that vehicle shadow color and glass for vehicle window color cause, lower to the classification accuracy rate of body color.In addition, the microscopic characteristics value of entire vehicle colour recognition by utilizing three passages of HSI color space (colour model be namely made up of form and aspect (Hue), saturation degree (Saturation) and intensity (Intensity)) to extract each pixel of body color, the threshold range of define color and mutual relationship, finally by methods such as K nearest neighbor algorithm, artificial neural network and support vector machine by color classification.The second body color recognizer is first positioning licence plate position, then to extract above car plate corresponding region and to go forward side by side driving body colour recognition as body color identified region.But when processing the video of car plate None-identified, the body color that these class methods cannot process, is difficult to meet consumers' demand.In a word, the impact that current most of recognition methods still cannot overcome vehicle shadow color preferably, glass for vehicle window Color pair vehicle color recognition result causes.
Therefore, how a kind of body color recognition methods that can overcome vehicle shadow color, glass for vehicle window color of development and Design, has become one of technical barrier being badly in need of at present solving.
Summary of the invention
The object of this invention is to provide a kind of body color recognition methods, to solve currently available technology to exist larger interference problem for colour recognition, to identify the color of vehicle body more accurately.
The body color recognition methods that the present invention proposes, comprises the following steps:
Obtain the video image of vehicle and set line of stumbling, moving object detection is carried out to the vedio data of input, obtain the foreground moving object images of binaryzation, intercept this foreground moving object circumscribed rectangular region, and the color pixel values that the foreground moving object region extracted in this rectangular image is corresponding in former frame of video, then calculate the difference of maxima and minima in the RGB Color Channel of all pixels in this boundary rectangle image, remove glass for vehicle window and vehicle shadow interference region by threshold segmentation method;
Calculate number of pixels ratio, distinguish colored vehicle and black and white grey vehicle, the ratio of the number of pixels comprised in the number of pixels in the vehicle body region that described number of pixels ratio obtains after the difference process of rgb space maxima and minima and binary conversion treatment for foreground moving object and the boundary rectangle image of foreground moving object;
Use H spatial histogram to amount to seven kinds of colors to red, orange, yellow, green, blue, blue, purple to identify; Adopt V spatial histogram and based on the color of sector region method of voting, three kinds of colors are amounted to black, white, ash and identify;
Export the body color identified.
The difference of described rgb space maxima and minima is the maximal value Max (R for the R passage of pixel each in body color recognition image in rgb color space, G passage and channel B, G, and minimum M in (R B), G, B) difference (i.e. Max (R, G, B)-Min (R, G, B)).
The number of pixels that the boundary rectangle of described foreground target comprises is adopt to detect based on VIBE (visual background extracting) algorithm the moving target obtained in foreground area, when detected moving target cross over stumble line time, namely the line pixel point set of stumbling on display foreground object pixel point set and image has first time to occur simultaneously, then calculate the number of pixels comprised in the boundary rectangle image of this foreground moving object pixel point set.
Described H spatial histogram be vehicle image after maxima and minima in rgb space subtracts each other and carries out binary conversion treatment, the H spatial histogram in corresponding color region in former frame of video, the vehicle body region obtained.
The described color method of voting based on sector region is will be hsv color space by the RGB color space conversion of all pixels in the foreground moving object boundary rectangle image of line of stumbling, find the barycenter of this foreground moving object, then be the center of circle with barycenter, with this barycenter to vehicle boundary rectangle edge edge bee-line for radius draw circle, by disc area by 72 degree be divided into 5 fan-shaped, and calculate pixel in each sector region (not containing the background area pixels) histogram in V space, detect the color that the position at place, top in histogram is corresponding in V space, this color is the color of sector region, finally having the maximum color of sector region quantity is the body color of this vehicle.
In one particular embodiment of the present invention, described body color recognition methods comprises the following steps:
Step S110. obtains the automobile video frequency on road to be detected;
Step S120. adopts VIBE (visual background extracting) algorithm to carry out moving object detection to video data, obtain the foreground moving object images of binaryzation, extract the foreground moving object by line of stumbling, the edge of traversal foreground moving object, maximal value and point coordinate corresponding to minimum value in the vertical and horizontal direction of record respectively, according to the boundary rectangle of the point coordinate determination foreground moving object obtained, and intercept boundary rectangle image P1, then extract foreground moving object region in this rectangular image P1 color pixel values corresponding in former frame of video and obtain image P2,
Step S130. calculates the maximal value Max of each pixel in rgb space (R, G, B) and minimum M in (R, G, B) in boundary rectangle image P2, and the difference then calculating maxima and minima obtains RGB error image P3;
The given threshold value M1 of step S140., uses this threshold value to carry out to RGB error image P3 the vehicle body area image P4 that Threshold segmentation obtains binaryzation;
The number of pixels N that in step S150. statistical picture P4, vehicle body region comprises
p4, the number N of all pixels in the boundary rectangle image P2 of statistics foreground moving object
p2, calculate number of pixels ratio R=N
p4/ N
p2;
Vehicle, by carrying out threshold decision to number of pixels ratio, is divided into colored car and black and white grey car by step S160.;
Step S170., when vehicle is colored car, uses H spatial histogram to amount to seven kinds of colors to red, orange, yellow, green, blue, blue, purple and identifies;
Step S180., when vehicle is achromaticity car, adopts V spatial histogram and identifies black, white, grey three kinds of colors altogether based on the color of sector region method of voting;
Step S190. stores or exports recognition result.
The present invention extracts foreground moving object from video, and the boundary rectangle image of the moving target by line of stumbling can be preserved, by calculating the maximal value of R passage, G passage and channel B in rgb space and minimum difference and Threshold segmentation, by vehicle window and shadow removal, colored vehicle and black-white-gray car two type is then divided into by vehicle to judge.If result of determination is colored car, identify according to form and aspect spatial histogram; If be judged to be black-white-gray car, identify according to V spatial histogram with based on the color method of voting of sector region.There is the problem of larger interference by the candidate regions judged for color that the method for RGB channel difference values solves that prior art chooses in the present invention, improves the accuracy of body color identification.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of a preferred embodiment of the present invention;
Fig. 2 is the present invention one automobile video frequency image (white dashed line is line of stumbling) to be identified;
Fig. 3 is the present invention's color space histogram.
Embodiment
The invention discloses a kind of body color recognition methods, comprise the following steps: that the start frame first reading video also sets line of stumbling, then moving object detection is carried out to the video data of input, calculate foreground moving object circumscribed rectangular region and intercept obtain boundary rectangle image, then the difference of all pixels maxima and minima in RGB Color Channel in this boundary rectangle image is calculated, and this vehicle color is divided into colored vehicle or black-white-gray vehicle, finally by the histogram distribution identification body color calculating shades of colour.The present invention adopts based on the maximal value of R passage, G passage and channel B in RGB color space and the body color recognition strategy identification body color of minimum difference, effectively can remove the interference region of body color identification, improve the accuracy of body color identification.
Below in conjunction with drawings and Examples, invention is described in detail.
As shown in Figure 1, body color recognition methods disclosed by the invention comprises the following steps:
Step S110, the automobile video frequency obtained on road to be detected.
By the traffic surveillance videos under camera shooting static scene, obtain the frame of video position in the video sequence captured, from first frame, read video by millisecond; Or directly import existing traffic surveillance videos, from importing the first frame of video, import video frame by frame according to the time sequencing of presentation of video frames.
Fig. 2 is automobile video frequency image to be identified.White dashed line is the line of stumbling that user marks, and the blue vehicle of linear contact lay of stumbling is vehicle to be identified.
Step S120, moving object detection is carried out to video data, intercept by the foreground moving object boundary rectangle image P1 of the binaryzation of line of stumbling, then extract foreground moving object region in this rectangular image color pixel values corresponding in former frame of video and obtain image P2.
User is according to self needs, and at automobile video frequency start frame, artificial dotted line of describing is as line of stumbling (see Fig. 2).Adopt based on VIBE (visual background extracting, Visual Background Extractor) algorithm detects the moving target obtained in foreground area, obtain the movement destination image P1 (namely in image, moving target pixel value is 1, and background pixel value is 0) of binaryzation.Concrete operations be when detected moving target cross over stumble line time, namely the line pixel point set of stumbling on display foreground object pixel point set and image has first time to occur simultaneously, the boundary rectangle then calculating this foreground moving object pixel point set (namely travels through the edge of foreground moving object, maximal value and pixel coordinate corresponding to minimum value in the vertical and horizontal direction of record respectively, boundary rectangle according to the coordinate determination foreground moving object calculated), then boundary rectangle image is intercepted, obtain the vehicle image P1 of binaryzation, then the color pixel values that foreground moving object region in this rectangle (i.e. pixel value be greater than 0 pixel region) is corresponding in former frame of video is extracted, obtain vehicle image P2.
Maximal value Max (the R of the RGB of each pixel in step S130, computed image P2, G, and minimum M in (R B), G, B), the difference then calculating maxima and minima obtains rgb space error image P3 (i.e. P3=Max (R, G, B)-Min (R, G, B)).
Step S140, a given threshold value M1, use this threshold value to carry out Threshold segmentation to image P3 and obtain binary image P4 (namely vehicle body area pixel value is 1, and background area pixels value is 0).The pixel that pixel value is greater than M1 is 1, otherwise is 0.
The number of pixels N that vehicle body region in step S150, statistical picture P4 (i.e. pixel value be greater than 0 region) comprises
p4, the number N of all pixels in the boundary rectangle image P2 of statistics foreground moving object
p2(i.e. the product of the number of pixels of rectangle frame length and rectangle frame width), calculates number of pixels ratio R=N
p4/ N
p2.
Step S160, by judging the size of number of pixels ratio R, vehicle is divided into colored car and black and white grey car.A given threshold value M2, if R is greater than threshold value M2, then proceeds to step S170 and carries out colour identification; If R is less than threshold value M2, then proceeds to step S180 and carry out black, white, grey recognition.
Step S170, in H (form and aspect) space define color span (see table 1), be defined as red, orange, yellow, green, blue, blue, purple respectively and amount to 7 kinds of colors.Extract the color value of region corresponding in image P2 that pixel value in binary image P4 is greater than 0, then RGB is transferred to hsv color space (colour model be namely made up of form and aspect (Hue), saturation degree (Saturation) and brightness (Value)).Calculate the histogram in H space, detect the color gamut in the H space that top falls in histogram, color corresponding to this scope is the color of this vehicle body.
Define color span in table 1:H (form and aspect) space
Step S180, in V (brightness) space define color span (see table 2), be defined as respectively black, white, ash amount to 3 kinds of colors.Be hsv color space by the RGB color space conversion of foreground moving object P2, finding the barycenter of P2, is then the center of circle with barycenter, to draw justify with this barycenter to the bee-line of vehicle boundary rectangle edge edge for radius.By disc area by 72 degree be divided into 5 fan-shaped, and calculate pixel in each sector region (not containing the background area pixels) histogram in V space, detect the color that the position at place, top in histogram is corresponding in V space, this color is the color of sector region, finally by the mode of voting, body color is voted, determine that having the maximum color of sector region quantity is the body color of this vehicle.
Define color span in table 2:V (brightness) space
| Brightness space (0 ~ 255) | Black | Ash | In vain |
| Maximal value | 0 | 47 | 221 |
| Minimum value | 46 | 220 | 255 |
Step S190, storage vehicle body colour recognition result.
Step S200, judge whether it is last frame video, if not, proceed to step S120, continue segmentation foreground moving object and also judge color; If not, proceed to step S210.
Step S210, return all body colors.Terminate body color identification process.
The color histogram distribution in H space of Fig. 3 corresponding to the vehicle body region of binaryzation in image P3, this histogram presents obvious single-peak response, and therefore the present invention can identify body color preferably.
More than show and describe ultimate principle of the present invention, principal character and feature of the present invention.Owing to employing the maxima and minima difference algorithm of color space, reduce the impact of vehicle shadow color, the identification of glass for vehicle window Color pair vehicle body colour, improve the accuracy of body color identification.Secondly, the present invention can go out body color from any angle recognition, carries out body color identification without the need to extracting further feature (as used License Plate color region).
Claims (8)
1. a body color recognition methods, is characterized in that, comprises the following steps:
Obtain the video image of vehicle and set line of stumbling, moving object detection is carried out to the vedio data of input, intercepting the boundary rectangle of foreground moving object; Then calculate the difference of maxima and minima in the RGB Color Channel of all pixels in this boundary rectangle image, remove glass for vehicle window and vehicle shadow interference region;
Calculate number of pixels ratio, distinguish colored vehicle and black and white grey vehicle, the ratio of the number of pixels that described number of pixels ratio comprises for the number of pixels in vehicle body region of foreground moving object after the difference process of rgb space maxima and minima and binary conversion treatment and the boundary rectangle of foreground moving object;
Use H spatial histogram to amount to seven kinds of colors to red, orange, yellow, green, blue, blue, purple to identify; Adopt V spatial histogram and based on the color of sector region method of voting, three kinds of colors are amounted to black, white, ash and identify;
Export the body color identified.
2. the method for claim 1, it is characterized in that: described rgb space maxima and minima difference is for the maximal value Max (R of pixel each in body color recognition image in rgb color space, G, and the difference of minimum M in (R, G, B) (i.e. Max (R B), G, B)-Min (R, G, B)).
3. the method for claim 1, it is characterized in that: the number of pixels that the boundary rectangle of foreground target comprises is adopt to detect based on VIBE (visual background extracting) algorithm the moving target obtained in foreground area, when detected moving target cross over stumble line time, namely the line pixel point set of stumbling on display foreground object pixel point set and image has first time to occur simultaneously, then the number of pixels that the boundary rectangle calculating this foreground moving object pixel point set comprises.
4. the method for claim 1, it is characterized in that: described H spatial histogram be vehicle image after the three-channel maximal value of R, G and B in rgb space and minimum value are subtracted each other and are carried out binary conversion treatment, the vehicle body region that obtains (i.e. pixel value be greater than 0 region) the H spatial histogram in corresponding color region in former frame of video.
5. the method for claim 1, it is characterized in that: the described color method of voting based on sector region is be hsv color space by the RGB color space conversion of extraneous for moving target rectangular image, find the barycenter of foreground moving object, then be the center of circle with barycenter, with this barycenter to vehicle boundary rectangle edge edge bee-line for radius draw circle, by disc area by 72 degree be divided into 5 fan-shaped, and calculate pixel in each sector region (not containing the background area pixels) histogram in V space, detect the color that the position at place, top in histogram is corresponding in V space, this color is the color of sector region, finally having the maximum color of sector region quantity is the body color of this vehicle.
6. a body color recognition methods, is characterized in that, comprises the following steps:
Step S110. obtains the automobile video frequency on road to be detected;
Step S120. carries out moving object detection to video data, extracts the foreground moving object by line of stumbling, and intercepts the boundary rectangle of foreground moving object;
Step S130. calculates maximal value Max (R, G, B) and the minimum M in (R, G, B) of the RGB of each pixel in boundary rectangle image, and the difference then calculating maxima and minima obtains RGB color difference image;
The given threshold value M1 of step S140., uses this threshold value to carry out Threshold segmentation to rgb space error image and obtains binary image;
Step S150. adds up the number of pixels N in vehicle body region in binary image P4
p4, the number N of all pixels in the image P2 corresponding to boundary rectangle of statistics foreground moving object
p2, calculate number of pixels ratio R=N
p2/ N
p4;
Vehicle, by number of pixels fractional threshold determination methods, is divided into colored car and black and white grey car by step S160.;
Step S170., when vehicle is colored car, uses H spatial histogram to amount to seven kinds of colors to red, orange, yellow, green, blue, blue, purple and identifies;
Step S180., when vehicle is achromaticity car, adopts V spatial histogram and identifies black, white, grey three kinds of colors altogether based on the color of sector region method of voting;
Step S190. stores or exports recognition result.
7. body color recognition methods as claimed in claim 6, it is characterized in that, in H (form and aspect) space, define color span is as following table:
8. body color recognition methods as claimed in claim 6, it is characterized in that, in described V (brightness) space, define color span is as following table:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510374911.5A CN105005766B (en) | 2015-07-01 | 2015-07-01 | A kind of body color recognition methods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510374911.5A CN105005766B (en) | 2015-07-01 | 2015-07-01 | A kind of body color recognition methods |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105005766A true CN105005766A (en) | 2015-10-28 |
| CN105005766B CN105005766B (en) | 2018-06-01 |
Family
ID=54378431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510374911.5A Active CN105005766B (en) | 2015-07-01 | 2015-07-01 | A kind of body color recognition methods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105005766B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106503638A (en) * | 2016-10-13 | 2017-03-15 | 金鹏电子信息机器有限公司 | For the image procossing of colour recognition, vehicle color identification method and system |
| CN107563374A (en) * | 2017-08-14 | 2018-01-09 | 华中科技大学 | A kind of digital display instrument charactron localization method |
| CN109190639A (en) * | 2018-08-16 | 2019-01-11 | 新智数字科技有限公司 | A kind of vehicle color identification method, apparatus and system |
| CN109508720A (en) * | 2018-10-23 | 2019-03-22 | 高新兴科技集团股份有限公司 | A kind of vehicle color identification method and device |
| CN109584317A (en) * | 2018-12-24 | 2019-04-05 | 天津天地伟业机器人技术有限公司 | Body color recognition methods based on HSV color space histogram |
| CN109635825A (en) * | 2018-12-19 | 2019-04-16 | 苏州市科远软件技术开发有限公司 | Vehicle attribute detection method, device and storage medium |
| CN110502986A (en) * | 2019-07-12 | 2019-11-26 | 平安科技（深圳）有限公司 | Identify character positions method, apparatus, computer equipment and storage medium in image |
| CN110751053A (en) * | 2019-09-26 | 2020-02-04 | 高新兴科技集团股份有限公司 | Vehicle color identification method, device, equipment and storage medium |
| CN112507801A (en) * | 2020-11-14 | 2021-03-16 | 武汉中海庭数据技术有限公司 | Lane road surface digital color recognition method, speed limit information recognition method and system |
| CN112907516A (en) * | 2021-01-27 | 2021-06-04 | 山东省计算中心（国家超级计算济南中心） | Sweet corn seed identification method and device for plug seedling |
| CN114170522A (en) * | 2022-02-14 | 2022-03-11 | 北京中科慧眼科技有限公司 | Color classification identification method and system based on chromatographic similarity measurement |
| CN115482474A (en) * | 2022-08-24 | 2022-12-16 | 湖南科技大学 | Bridge deck vehicle load identification method and system based on high-altitude aerial image |
| CN116563770A (en) * | 2023-07-10 | 2023-08-08 | 四川弘和数智集团有限公司 | Method, device, equipment and medium for detecting vehicle color |
| CN117152167A (en) * | 2023-10-31 | 2023-12-01 | 海信集团控股股份有限公司 | Target removing method and device based on segmentation large model |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101504717A (en) * | 2008-07-28 | 2009-08-12 | 上海高德威智能交通系统有限公司 | Characteristic area positioning method, car body color depth and color recognition method |
| US8340420B2 (en) * | 2009-10-05 | 2012-12-25 | National Taiwan University | Method for recognizing objects in images |
| CN103425989A (en) * | 2013-08-07 | 2013-12-04 | 中山大学 | Vehicle color identification method based on significance analysis |
| CN103544480A (en) * | 2013-10-17 | 2014-01-29 | 公安部第三研究所 | Vehicle color recognition method |
-
2015
- 2015-07-01 CN CN201510374911.5A patent/CN105005766B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101504717A (en) * | 2008-07-28 | 2009-08-12 | 上海高德威智能交通系统有限公司 | Characteristic area positioning method, car body color depth and color recognition method |
| US8340420B2 (en) * | 2009-10-05 | 2012-12-25 | National Taiwan University | Method for recognizing objects in images |
| CN103425989A (en) * | 2013-08-07 | 2013-12-04 | 中山大学 | Vehicle color identification method based on significance analysis |
| CN103544480A (en) * | 2013-10-17 | 2014-01-29 | 公安部第三研究所 | Vehicle color recognition method |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106503638B (en) * | 2016-10-13 | 2019-09-13 | 金鹏电子信息机器有限公司 | Image procossing, vehicle color identification method and system for color identification |
| CN106503638A (en) * | 2016-10-13 | 2017-03-15 | 金鹏电子信息机器有限公司 | For the image procossing of colour recognition, vehicle color identification method and system |
| CN107563374A (en) * | 2017-08-14 | 2018-01-09 | 华中科技大学 | A kind of digital display instrument charactron localization method |
| CN109190639A (en) * | 2018-08-16 | 2019-01-11 | 新智数字科技有限公司 | A kind of vehicle color identification method, apparatus and system |
| CN109508720A (en) * | 2018-10-23 | 2019-03-22 | 高新兴科技集团股份有限公司 | A kind of vehicle color identification method and device |
| CN109508720B (en) * | 2018-10-23 | 2021-04-23 | 高新兴科技集团股份有限公司 | Vehicle color identification method and device |
| CN109635825A (en) * | 2018-12-19 | 2019-04-16 | 苏州市科远软件技术开发有限公司 | Vehicle attribute detection method, device and storage medium |
| CN109584317A (en) * | 2018-12-24 | 2019-04-05 | 天津天地伟业机器人技术有限公司 | Body color recognition methods based on HSV color space histogram |
| CN110502986B (en) * | 2019-07-12 | 2024-09-17 | 平安科技（深圳）有限公司 | Method, device, computer equipment and storage medium for identifying positions of persons in image |
| CN110502986A (en) * | 2019-07-12 | 2019-11-26 | 平安科技（深圳）有限公司 | Identify character positions method, apparatus, computer equipment and storage medium in image |
| CN110751053A (en) * | 2019-09-26 | 2020-02-04 | 高新兴科技集团股份有限公司 | Vehicle color identification method, device, equipment and storage medium |
| CN110751053B (en) * | 2019-09-26 | 2022-02-22 | 高新兴科技集团股份有限公司 | Vehicle color identification method, device, equipment and storage medium |
| CN112507801A (en) * | 2020-11-14 | 2021-03-16 | 武汉中海庭数据技术有限公司 | Lane road surface digital color recognition method, speed limit information recognition method and system |
| CN112907516A (en) * | 2021-01-27 | 2021-06-04 | 山东省计算中心（国家超级计算济南中心） | Sweet corn seed identification method and device for plug seedling |
| CN114170522A (en) * | 2022-02-14 | 2022-03-11 | 北京中科慧眼科技有限公司 | Color classification identification method and system based on chromatographic similarity measurement |
| CN115482474A (en) * | 2022-08-24 | 2022-12-16 | 湖南科技大学 | Bridge deck vehicle load identification method and system based on high-altitude aerial image |
| CN116563770A (en) * | 2023-07-10 | 2023-08-08 | 四川弘和数智集团有限公司 | Method, device, equipment and medium for detecting vehicle color |
| CN116563770B (en) * | 2023-07-10 | 2023-09-29 | 四川弘和数智集团有限公司 | Method, device, equipment and medium for detecting vehicle color |
| CN117152167A (en) * | 2023-10-31 | 2023-12-01 | 海信集团控股股份有限公司 | Target removing method and device based on segmentation large model |
| CN117152167B (en) * | 2023-10-31 | 2024-03-01 | 海信集团控股股份有限公司 | Target removing method and device based on segmentation large model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105005766B (en) | 2018-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105005766A (en) | Vehicle body color identification method | |
| CN109886896B (en) | Blue license plate segmentation and correction method | |
| CN106709436B (en) | Track traffic panoramic monitoring-oriented cross-camera suspicious pedestrian target tracking system | |
| CN102609686B (en) | Pedestrian detection method | |
| CN103824066B (en) | A kind of licence plate recognition method based on video flowing | |
| CN105046196B (en) | Front truck information of vehicles structuring output method based on concatenated convolutional neutral net | |
| CN103279736B (en) | A kind of detection method of license plate based on multi-information neighborhood ballot | |
| CN103035013B (en) | A kind of precise motion shadow detection method based on multi-feature fusion | |
| CN102567727B (en) | Method and device for replacing background target | |
| CN105160691A (en) | Color histogram based vehicle body color identification method | |
| CN104978567B (en) | Vehicle checking method based on scene classification | |
| CN111369596B (en) | Escalator passenger flow volume statistical method based on video monitoring | |
| CN102880863B (en) | Method for positioning license number and face of driver on basis of deformable part model | |
| CN102915433B (en) | Character combination-based license plate positioning and identifying method | |
| Zhang et al. | A multi-feature fusion based traffic light recognition algorithm for intelligent vehicles | |
| CN105741324A (en) | Moving object detection identification and tracking method on moving platform | |
| CN111553214B (en) | Method and system for detecting smoking behavior of driver | |
| CN102842037A (en) | Method for removing vehicle shadow based on multi-feature fusion | |
| CN102915544A (en) | Video image motion target extracting method based on pattern detection and color segmentation | |
| CN103577832B (en) | A kind of based on the contextual people flow rate statistical method of space-time | |
| Zhu et al. | Fast detection of moving object based on improved frame-difference method | |
| CN112115800A (en) | Vehicle combination recognition system and method based on deep learning target detection | |
| CN109800693B (en) | Night vehicle detection method based on color channel mixing characteristics | |
| CN101866553A (en) | Location method of plate number based on difference and color pair characteristic point | |
| CN104268509A (en) | Method and system for detecting license plate of dump truck |
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 |