CN105005766B - A kind of body color recognition methods - Google Patents
A kind of body color recognition methods Download PDFInfo
- Publication number
- CN105005766B CN105005766B CN201510374911.5A CN201510374911A CN105005766B CN 105005766 B CN105005766 B CN 105005766B CN 201510374911 A CN201510374911 A CN 201510374911A CN 105005766 B CN105005766 B CN 105005766B
- Authority
- CN
- China
- Prior art keywords
- color
- vehicle
- image
- pixels
- region
- 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.)
- Active
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 invention discloses a kind of body color recognition methods, comprise the following steps：Detection unit carries out motion detection to the video data of input, and the method using rgb space maxima and minima difference removes glass for vehicle window and vehicle shadow interference region；By calculating number of pixels ratio, colored vehicle and black and white grey vehicle are distinguished；Using H spatial histograms red, orange, yellow, green, blue, blue, purple are amounted to 7 kinds of colors to be identified, voting method using V spatial histograms and the color based on sector region is identified black, white, grey 3 kinds of colors altogether.The present invention can effectively be removed the interference region of body color identification, be improved the accuracy that body color identifies using the body color recognition strategy identification body color based on RGB color channel difference values.
Description
Technical field
The present invention relates to image processing techniques, and in particular to a kind of body color recognition methods.
Background technology
In current intelligent transportation system, as vehicle fleet size increases, traffic environment becomes increasingly complicated, only by car plate pair
The needs that cannot meet people are identified in vehicle.Vehicle color information is easier to arouse people's interest, so as to make up
The deficiency of Car license recognition caused by vehicle fake-license, the more board phenomenons of a vehicle, and to the identification of vehicle and search, perfect and enhancing intelligence
Traffic system function is of great significance.
At present, to the identification of vehicle color, there are mainly two types of approach：The first is that vehicle integral color is identified, and is carried
The color characteristic of interested area of vehicle is taken to be identified.Vehicle foreground image is obtained by Target Segmentation first, is then carried out
Unicom regional analysis deletes the interference regions such as wheel, reflective mirror and obtains the apparent region of vehicle color.In sorting phase, using base
Color is divided into the types such as black, white, grey, red, yellow, green, blue in the two layers of classified device of support vector machines, but this method easily by
It is relatively low to the classification accuracy rate of body color to interference caused by vehicle shadow color and glass for vehicle window color.In addition, vehicle is whole
The identification of body color can be by using HSI color spaces (i.e. by form and aspect (Hue), saturation degree (Saturation) and intensity
(Intensity) composition colour model) three passages extraction each pixel of body color microscopic characteristics value, define color
Threshold range and correlation, finally by the methods of K nearest neighbor algorithms, artificial neural network and support vector machines by color point
Class.Second of body color recognizer is first positioning licence plate position, then extracts car plate top corresponding region as vehicle body face
Color identification region simultaneously carries out body color identification.However, when handling the video of car plate None- identified, such method can not be handled
Body color, it is difficult to meet user demand.In short, major part recognition methods at present still can not preferably overcome vehicle shadow face
Color, glass for vehicle window color are influenced caused by vehicle color recognition result.
Therefore, how to develop and design a kind of body color identification that can overcome vehicle shadow color, glass for vehicle window color
Method, it has also become be badly in need of one of technical barrier solved at present.
The content of the invention
The object of the present invention is to provide a kind of body color recognition methods, are identified with solving currently available technology for color
There are problems that larger interference, more accurately to identify the color of vehicle body.
Body color recognition methods proposed by the present invention, comprises the following steps：
The video image and setting for obtaining vehicle are stumbled line, are carried out moving object detection to the vedio data of input, are obtained
To the foreground moving object images of binaryzation, the foreground moving object circumscribed rectangular region is intercepted, and extracts the rectangular image
In foreground moving object region color pixel values corresponding in former video frame, then calculate institute in the boundary rectangle image
There is the difference of maxima and minima in the RGB color passage of pixel, glass for vehicle window and vehicle are removed by threshold segmentation method
Shadow interference region；Number of pixels ratio is calculated, colored vehicle is distinguished and black and white grey vehicle, the number of pixels ratio is
The vehicle body region that foreground moving object obtains after the processing of RGB color passage maxima and minima difference and binary conversion treatment
Number of pixels and foreground moving object boundary rectangle image included in the ratio between number of pixels；
Using H spatial histograms red, orange, yellow, green, blue, blue, purple are amounted to seven kinds of colors to be identified；It is straight using V spaces
Side's figure and the color based on sector region vote method and black, white, grey three kinds of colors altogether are identified；
Export the body color identified.
The difference of the RGB color passage maxima and minima is to identify each pixel in image for body color
The difference of the maximum Max (R, G, B) and minimum M in (R, G, B) of R passages, G passages and channel B in rgb color space
(i.e. Max (R, G, B)-Min (R, G, B)).
The number of pixels that the boundary rectangle of the foreground target is included is using based on VIBE (visualization background extracting)
Algorithm detection obtains the moving target in foreground area, when detected moving target across stumble line when, i.e. display foreground target
Line pixel point set of stumbling on pixel point set and image has first time intersection, then calculates the outer of the foreground moving object pixel point set
Connect number of pixels included in rectangular image.
The H spatial histograms subtract each other and carry out by maxima and minima in RGB color passage for vehicle image
After binary conversion treatment, the H spatial histograms in obtained vehicle body region corresponding color region in former video frame.
The color based on sector region votes method as will be by the external square of foreground moving object for line of stumbling
The RGB color of all pixels is converted to hsv color space in shape image, finds the barycenter of the foreground moving object, then
It using barycenter as the center of circle, is drawn and justified as radius using the beeline of the barycenter to vehicle boundary rectangle frame edge, disc area is pressed 72
Degree is divided into 5 sectors, and calculates histogram of the pixel in each sector region (without background area pixels) in V spaces, inspection
It is the color of sector region to survey the position in histogram where top corresponding color, the color in V spaces, is finally gathered around
The color for having sector region quantity most is the body color of the vehicle.
In one particular embodiment of the present invention, the body color recognition methods comprises the following steps：
Step S110. obtains road vehicle video to be detected；
Step S120. carries out moving object detection using VIBE (visualization background extracting) algorithms to video data, obtains
The foreground moving object images of binaryzation extract the foreground moving object by line of stumbling, and travel through the edge of foreground moving object, point
Vertically and horizontally upper maximum and the corresponding point coordinates of minimum value are not recorded, and foreground moving is determined according to obtained point coordinates
The boundary rectangle of target, and boundary rectangle image P1 is intercepted, then extract the foreground moving object area in rectangular image P1
Domain color pixel values corresponding in former video frame simultaneously obtain image P2；
Step S130. calculate each pixel in boundary rectangle image P2 in RGB color passage maximum Max (R, G,
B) and minimum M in (R, G, B), the difference for then calculating maxima and minima obtain RGB color error image P3；
Step S140. gives a threshold value M1, and RGB color error image P3 is obtained into row threshold division using the threshold value
The vehicle body area image P4 of binaryzation；
The number of pixels N that vehicle body region includes in step S150. statistical pictures P4P4, count the external of foreground moving object
The number N of all pixels in rectangular image P2P2, calculate number of pixels ratio R=NP4/NP2；
Vehicle is divided into colored vehicle and black and white grey vehicle by step S160. by carrying out threshold decision to number of pixels ratio；
Step S170. amounts to seven using H spatial histograms when vehicle is colored vehicle to red, orange, yellow, green, blue, blue, purple
Kind color is identified；
Step S180. when vehicle be achromaticity vehicle when, using V spatial histograms and the color based on sector region ballot table
Certainly black, white, grey three kinds of colors altogether are identified in method；
Step S190. is stored or output recognition result.
The present invention extracts foreground moving object from video, and can preserve the boundary rectangle of the moving target by line of stumbling
Image, will by calculating R passages in RGB color passage, the maximum of G passages and channel B and minimum difference and Threshold segmentation
Then vehicle is divided into colored vehicle and black-white-gray vehicle two types judges by vehicle window and shadow removal.If it is determined that result is
Colored vehicle is identified according to form and aspect spatial histogram；If it is determined that it is black-white-gray vehicle, according to V spatial histograms and based on sector
The color method of voting in region is identified.The present invention solves prior art selection by the method for RGB channel difference
For color judge candidate regions there are problems that larger interference, improve body color identification accuracy.
Description of the drawings
Fig. 1 is the flow diagram of a preferred embodiment of the present invention；
Fig. 2 is the automobile video frequency image to be identified of the present invention one (white dashed line is line of stumbling)；
Fig. 3 is a color space histogram of the invention.
Specific embodiment
The invention discloses a kind of body color recognition methods, comprise the following steps：Read the start frame of video simultaneously first
Line of stumbling is set, moving object detection then is carried out to the video data of input, calculates the circumscribed rectangular region of foreground moving object
And intercept and obtain boundary rectangle image, then calculate all pixels maximum in RGB color passage in the boundary rectangle image
With the difference of minimum value, and the vehicle color is divided into colored vehicle or black-white-gray vehicle, finally by calculating a variety of colors
Histogram distribution identifies body color.The present invention uses the maximum based on R passages, G passages and channel B in RGB color
Body color is identified with the body color recognition strategy of minimum difference, can effectively remove the interference range of body color identification
Domain improves the accuracy of body color identification.
Invention is described in detail with reference to the accompanying drawings and examples.
As shown in Figure 1, body color recognition methods disclosed by the invention comprises the following steps：
Step S110, road vehicle video to be detected is obtained.
Traffic surveillance videos under static scene are shot by camera, obtain captured video frame in the video sequence
Position, since first frame by millisecond read video；Or existing traffic surveillance videos are introduced directly into, from the first frame for importing video
Start, video is imported frame by frame according to the time sequencing of presentation of video frames.
Fig. 2 is automobile video frequency image to be identified.The blue vehicle that white dashed line contacts for stumble line and the line of stumbling of user's mark
Be vehicle to be identified.
Step S120, moving object detection is carried out to video data, intercepts the foreground moving mesh by the binaryzation for line of stumbling
Boundary rectangle image P1 is marked, it is corresponding in former video frame then to extract the foreground moving object region in the rectangular image
Color pixel values simultaneously obtain image P2.
According to their needs, in automobile video frequency start frame, artificial dotted line of describing is as line of stumbling by user (see Fig. 2).Using base
The fortune in foreground area is obtained in VIBE (visualization background extracting, Visual Background Extractor) algorithm detection
Moving-target obtains the movement destination image P1 of binaryzation (moving target pixel value is 1 i.e. in image, background pixel value 0).Tool
Gymnastics work be when detected moving target across stumble line when, i.e., the line pixel of stumbling on display foreground object pixel point set and image
Point set has first time intersection, then the boundary rectangle for calculating the foreground moving object pixel point set (travels through foreground moving object
Edge, vertically and horizontally upper maximum and the corresponding pixel point coordinates of minimum value are recorded respectively, according to the seat calculated
Mark determines the boundary rectangle of foreground moving object), boundary rectangle image is then intercepted, obtains the vehicle image P1 of binaryzation, so
The foreground moving object region (i.e. pixel region of the pixel value more than 0) extracted afterwards in the rectangle is corresponding in former video frame
Color pixel values, obtain vehicle image P2.
Step S130, calculate the RGB of each pixel in image P2 maximum Max (R, G, B) and minimum M in (R, G,
B), the difference for then calculating maxima and minima obtains RGB color channel difference values image P3 (i.e. P3=Max (R, G, B)-Min
(R,G,B))。
Step S140, a threshold value M1 is given, binary image is obtained into row threshold division to image P3 using the threshold value
(i.e. vehicle body area pixel value is 1 to P4,0) background area pixels value is.Pixel of the pixel value more than M1 is 1, otherwise is 0.
Step S150, the number of pixels N that the vehicle body region in statistical picture P4 (i.e. region of the pixel value more than 0) includesP4,
Count the number N of all pixels in the boundary rectangle image P2 of foreground moving objectP2(i.e. rectangle frame length and rectangle width of frame
The product of number of pixels), calculate number of pixels ratio R=NP4/NP2。
Step S160, by judging the size of number of pixels ratio R, vehicle is divided into colored vehicle and black and white grey vehicle.It gives
A fixed threshold value M2 if R is more than threshold value M2, is transferred to step S170 and carries out colored identification；If R is less than threshold value M2, step is transferred to
Rapid S180 carries out black, white, grey recognition.
Step S170, the color value range (being shown in Table 1) defined in H (form and aspect) space, be respectively defined as it is red, orange, yellow,
It is green, blue or green, blue, purple to amount to 7 kinds of colors.It is corresponding in image P2 to extract region of the pixel value more than 0 in binary image P4
Then RGB is switched to hsv color space (i.e. by form and aspect (Hue), saturation degree (Saturation) and brightness (Value) by color value
The colour model of composition).The histogram in H spaces is calculated, detects the color gamut in the H spaces that top is fallen into histogram,
The corresponding color of the scope is the color of the vehicle body.
Table 1:Color value range defined in H (form and aspect) space
Step S180, the color value range (being shown in Table 2) defined in V (brightness) space, is respectively defined as black, white, grey common
Count 3 kinds of colors.The RGB color of foreground moving object P2 is converted into hsv color space, finds the barycenter of P2, then with
Barycenter is the center of circle, is drawn and justified as radius using the beeline of the barycenter to vehicle boundary rectangle frame edge.Disc area is pressed 72 degree
5 sectors are divided into, and calculate histogram of the pixel in each sector region (without background area pixels) in V spaces, detection
Position in histogram where top corresponding color, the color in V spaces are the color of sector region, finally by
The mode voted votes to body color, determines to possess the vehicle body face that the most color of sector region quantity is the vehicle
Color.
Table 2:Color value range defined in V (brightness) space
| Brightness space (0~255) | It is black | Ash | In vain |
| Maximum | 0 | 47 | 221 |
| Minimum value | 46 | 220 | 255 |
Step S190, body color recognition result is stored.
Step S200, judge whether it is last frame video, if it is not, being transferred to step S120, continue segmentation prospect fortune
Moving-target simultaneously judges color；If it is not, it is transferred to step S210.
Step S210, all body colors are returned.Terminate body color identification process.
Histogram distributions of the Fig. 3 for the color corresponding to the vehicle body region of binaryzation in image P3 in H spaces, the Nogata
Apparent single-peak response is presented in figure, therefore the present invention can preferably identify body color.
Basic principle, main feature and the features of the present invention of the present invention has been shown and described above.Due to the use of face
The maxima and minima difference algorithm of the colour space reduces vehicle shadow color, glass for vehicle window color to the colored identification of vehicle body
Influence, improve body color identification accuracy.Secondly, the present invention can go out body color, nothing from any angle recognition
Other feature (such as with License Plate color region) need to be extracted to carry out body color identification.
Claims (8)
1. a kind of body color recognition methods, which is characterized in that comprise the following steps：
The video image and setting for obtaining vehicle are stumbled line, moving object detection are carried out to the vedio data of input, before interception
The boundary rectangle of scape moving target；Then calculate in the boundary rectangle image in the RGB color passage of all pixels maximum with
The difference of minimum value passes through a threshold value M1 so that the threshold value obtains binaryzation to RGB color error image into row threshold division
Vehicle body image, to remove glass for vehicle window and vehicle shadow interference region；
Number of pixels ratio is calculated, distinguishes colored vehicle and black and white grey vehicle, the number of pixels ratio is foreground moving
Target is by maxima and minima difference processing in RGB color passage and the number of pixels in the vehicle body region after binary conversion treatment
The ratio between number of pixels included with the boundary rectangle of foreground moving object；
Using H spatial histograms red, orange, yellow, green, blue, blue, purple are amounted to seven kinds of colors to be identified；Using V spatial histograms
Method is voted with the color based on sector region black, white, grey three kinds of colors altogether are identified；
Export the body color identified.
2. the method as described in claim 1, it is characterised in that：Maxima and minima difference in the RGB color passage
Maximum Max (R, G, B) and minimum value of each pixel in rgb color space in image are identified to be used for body color
The difference of Min (R, G, B), i.e. Max (R, G, B)-Min (R, G, B).
3. the method as described in claim 1, it is characterised in that：The number of pixels that the boundary rectangle of foreground target is included is to adopt
With the moving target obtained in foreground area is detected based on VIBE (visualization background extracting) algorithm, when having detected moving target
Across stumble line when, i.e., the line pixel point set of stumbling on display foreground object pixel point set and image has first time intersection, then calculates
The number of pixels that the boundary rectangle of the foreground moving object pixel point set is included.
4. the method as described in claim 1, it is characterised in that：The H spatial histograms pass through rgb space for vehicle image
After the maximum and minimum value of middle R, G and B triple channel subtract each other and carry out binary conversion treatment, obtained vehicle body region, i.e. pixel value
Region more than 0, the H spatial histograms in corresponding color region in former video frame.
5. the method as described in claim 1, it is characterised in that：The color method of voting based on sector region is
The RGB color of moving target external world rectangular image is converted into hsv color space, finds the barycenter of foreground moving object,
Then using barycenter as the center of circle, drawn and justified as radius using the beeline of the barycenter to vehicle boundary rectangle frame edge, by disc area
5 sectors are divided by 72 degree, and calculate pixel in each sector region, the pixel without background area, the Nogata in V spaces
Figure, it is the color of sector region to detect the position in histogram where top corresponding color, the color in V spaces,
Finally possess the body color that the most color of sector region quantity is the vehicle.
6. a kind of body color recognition methods, which is characterized in that comprise the following steps：
Step S110. obtains road vehicle video to be detected；
Step S120. carries out moving object detection to video data, extracts the foreground moving object by line of stumbling, interception prospect fortune
The boundary rectangle of moving-target；
Step S130. calculates the maximum Max (R, G, B) and minimum of the RGB color passage of each pixel in boundary rectangle image
Value Min (R, G, B), the difference for then calculating maxima and minima obtain RGB color error image；
Step S140. gives a threshold value M1, and binaryzation is obtained into row threshold division to RGB color error image using the threshold value
Image；
The number of pixels N in vehicle body region in step S150. statistics binary images P4P4, count the external square of foreground moving object
The number N of all pixels in image P2 corresponding to shapeP2, calculate number of pixels ratio R=NP2/ NP4；
Vehicle is divided into colored vehicle and black and white grey vehicle by step S160. by number of pixels fractional threshold determination methods；
Step S170. amounts to red, orange, yellow, green, blue, blue, purple using H spatial histograms seven kinds of face when vehicle is colored vehicle
Color is identified；
Step S180. is when vehicle is achromaticity vehicle, using V spatial histograms and the color side of voting based on sector region
Black, white, grey three kinds of colors altogether are identified in method；
Step S190. is stored or output recognition result.
7. body color recognition methods as claimed in claim 6, which is characterized in that H（Form and aspect）Color value defined in space
Scope such as following table：
。
8. body color recognition methods as claimed in claim 6, which is characterized in that the V（Brightness）Color defined in space
Value range such 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 CN105005766A (en) | 2015-10-28 |
| CN105005766B true 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) |
Families Citing this family (14)
| 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 |
| CN107563374B (en) * | 2017-08-14 | 2020-07-10 | 华中科技大学 | Digital display instrument nixie tube positioning 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 |
| 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 |
| CN109584317B (en) * | 2018-12-24 | 2023-07-25 | 天津天地伟业信息系统集成有限公司 | Vehicle body color identification method 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 |
| 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 |
| CN112907516B (en) * | 2021-01-27 | 2022-08-02 | 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 |
| CN115482474B (en) * | 2022-08-24 | 2023-07-04 | 湖南科技大学 | Bridge deck vehicle load identification method and system based on aerial image |
| 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 |
| 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 |
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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105005766A (en) | 2015-10-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105005766B (en) | A kind of body color recognition methods | |
| CN106709436B (en) | Track traffic panoramic monitoring-oriented cross-camera suspicious pedestrian target tracking system | |
| CN103035013B (en) | A kind of precise motion shadow detection method based on multi-feature fusion | |
| US7664315B2 (en) | Integrated image processor | |
| TWI409718B (en) | Method of locating license plate of moving vehicle | |
| CN102426649B (en) | Simple steel seal digital automatic identification method with high accuracy rate | |
| US8345100B2 (en) | Shadow removal in an image captured by a vehicle-based camera using an optimized oriented linear axis | |
| CN107301405A (en) | Method for traffic sign detection under natural scene | |
| US8433099B2 (en) | Vehicle discrimination apparatus, method, and computer readable medium storing program thereof | |
| CN102194108B (en) | Smile face expression recognition method based on clustering linear discriminant analysis of feature selection | |
| CN111369596B (en) | Escalator passenger flow volume statistical method based on video monitoring | |
| CN104766071B (en) | A kind of traffic lights fast algorithm of detecting applied to pilotless automobile | |
| CN107507206B (en) | Depth map extraction method based on significance detection | |
| CN106778633B (en) | Pedestrian identification method based on region segmentation | |
| CN102880863A (en) | Method for positioning license number and face of driver on basis of deformable part model | |
| CN104182769A (en) | Number plate detection method and system | |
| CN108038458B (en) | Method for automatically acquiring outdoor scene text in video based on characteristic abstract diagram | |
| CN104143077A (en) | Pedestrian target retrieving method and system based on images | |
| CN112115800A (en) | Vehicle combination recognition system and method based on deep learning target detection | |
| CN104268509B (en) | The method and system of dump truck car plate detection | |
| CN106446832B (en) | Video-based pedestrian real-time detection method | |
| CN115601358B (en) | Tongue picture image segmentation method under natural light environment | |
| CN103839232B (en) | A kind of pedestrian's cast shadow suppressing method based on agglomerate model | |
| Le et al. | Pedestrian lane detection in unstructured environments for assistive navigation | |
| Parente et al. | Assessing facial image accordance to ISO/ICAO requirements |
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 |