CN102184413A - Automatic vehicle body color recognition method of intelligent vehicle monitoring system - Google Patents
Automatic vehicle body color recognition method of intelligent vehicle monitoring system Download PDFInfo
- Publication number
- CN102184413A CN102184413A CN 201110124540 CN201110124540A CN102184413A CN 102184413 A CN102184413 A CN 102184413A CN 201110124540 CN201110124540 CN 201110124540 CN 201110124540 A CN201110124540 A CN 201110124540A CN 102184413 A CN102184413 A CN 102184413A
- Authority
- CN
- China
- Prior art keywords
- color
- body color
- space
- value
- pixel
- 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
Images
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an automatic vehicle body color recognition method of an intelligent vehicle monitoring system. The method comprises the following steps: firstly detecting a feature region on the behalf of a vehicle body color according to the position of a plate number and the textural features of the vehicle body; then, carrying out color space conversion and vector space quantization synthesis on pixels of the vehicle body feature region, and extracting normalization features of an obscure histogram Bin from the quantized vector space; carrying feature dimension reduction on the acquired high-dimension features by adopting an LDA (Linear Discriminant Analysis) method; carrying out various subspace analysis on the vehicle body color, then carrying out vehicle body color recognition of the subspaces by utilizing the recognition parameters of an offline training classifier, and adopting a multi-feature template matching or SVM (Space Vector Modulation) method; and finally, correcting color with easy intersection and low reliability according to the initial recognition reliability and color priori knowledge, so as to obtain the final vehicle body recognition result. The automatic vehicle body color recognition method is applicable to conditions of daylight, night and sunshine and is fast in recognition speed and high in recognition accuracy.
Description
Technical field
The present invention relates to pattern-recognition and image processing techniques, relate in particular to a kind of based on the recognition methods of representing body color in stationary vehicle image or the video flowing.
Background technology
Along with socioeconomic fast development, road traffic develops rapidly, vehicle population and traffic trip amount increase severely, develop the problems that caused rapidly for solving traffic above-ground, intelligent transportation system (ITS) plays an important role, and in the intelligent monitoring technology of intelligent transportation field, automatic Recognition of License Plate is its core, through the years of researches development, its technology is ripe gradually, and is well used in the various application scenarios of traffic administration.For example charge automatically, non-parking charge, stolen vehicle are pursued etc.
Yet, along with vehicle fleet size constantly increases, the diversity and the complicacy of traffic hazard, escape violating the regulations and offender's crime, it is not enough that the Vehicular intelligent monitoring only relies on the information of car plate identification, when car plate was subjected to factor affecting such as partial occlusion, stain, ambient lighting in addition, its accurate discrimination sharply descended; In addition, Vehicle License Plate Recognition System seems powerless under situations such as the many boards of a car, the many cars of a board, cover board.Therefore, when license plate number is discerned, also need other characteristic informations of vehicle are discerned, as body color, car mark, vehicle etc.
The literature research of at present relevant body color identification is also fewer, obtain with the equal correct recognition rata difficulty of car plate very big, mainly be to be subjected to multifactor interference such as illumination, different weather and noise to produce cross-color owing to the bodywork surface color, and body color change with illumination condition and incident light change color.For this reason, the present invention has proposed body color automatic identifying method in the Vehicular intelligent supervisory system on the basis of existing achievement in research and independent development automatic license plate identification system.
The body color recognition methods of Shi Yonging now has:
(1) based on the method for aberration.This method mainly is transformed into rgb space the HSI space, then in HIS SPATIAL CALCULATION value of chromatism, searches corresponding color according to the value of chromatism minimum in color table and is recognition result.This method is simply quick, can obtain recognition result preferably under the situation preferably at light, but in the practical application, illumination, weather all has very big variation, and its body color of the vehicle image that obtains is all unstable.See document for details: Li Guijun, Liu Zhengxi etc. a kind of based on aberration and colored normalized body color recognition methods. computer utility, 2004,9.
(2) based on chrominance information and SVM recognition methods.It at first is transformed into rgb space Lab, HSV, utilizes chrominance information ab and HS then respectively, and the color with similar chrominance information is merged, and carries out SVM and arest neighbors Classification and Identification in different characteristic color intervals respectively, has improved its accuracy rate.But this body color recognition methods is based on the color identification of whole vehicle image (except that background), and operand is big, and has introduced some interference regions, and as vehicle window, the car light part causes system reliability and the not enough robust of identification.See document for details: Wang Yunqiong, You Zhisheng. utilize support vector machine identification vehicle color. computer-aided design (CAD) and graphics journal, 2004,5.
Above-mentioned body color recognition methods can obtain better effects to the more single car of body color under certain photoenvironment, but some large trucks and lorry are subjected to usually the influence of headstock or tailstock heat radiation case, vent fan, body color surveyed area not accurate enough (containing other non-body colors in the zone of detection) and producing under the disturbed condition, its reliability is relatively poor, and accuracy of identification is lower; In addition, at night, body portion light is darker and inhomogeneous in the vehicle image, even have under the flashlamp condition, the body color identification error is bigger, and the above method does not also solve well to the reliable recognition of vehicle image color at night.
Summary of the invention
The objective of the invention is at the deficiencies in the prior art, body color automatic identifying method in a kind of Vehicular intelligent supervisory system is provided, this method adopts intelligent the processing to vehicular traffic video and the image of gathering, and realizes the fast automatic identification to the automobile body color.
The objective of the invention is to be achieved through the following technical solutions: body color automatic identifying method in a kind of Vehicular intelligent supervisory system may further comprise the steps:
(1), above car plate, detects the thick zone of representing body color with reference to car plate height, wide and location coordinate information by car plate location technology gained.
(2), in the thick zone of detecting of body color, further search for the accurate characteristic area of body color according to the textural characteristics in vehicle body zone.
(3) body color characteristic area rgb pixel is transformed into the HSV space, and based on the synthetic new color space of HSV character
(4) adopt the fuzzy C-means clustering method in the vector color SPATIAL CALCULATION fuzzy histogram Bin normalization feature that quantizes, and utilize the LDA method, obtain the characteristic that helps color discrimination the feature dimensionality reduction.
(5) according to the sorter identification parameter of off-line sample training, adopt feature templates to mate or the identification of SVM method, obtain preliminary body color result based on the multiclass subspace to vehicle image body color characteristic area.
(6) according to first recognition credibility and color priori, commute intersects and vehicle color with a low credibility is carried out the correction and the affirmation of day and night respectively, to obtain final body color recognition result.
Optionally, on behalf of the thick zone of body color, described the detection above car plate comprise that the center with license plate area is a reference point, intercepts one up highly to be
Widely be
The rectangular area, wherein
Be respectively the height and width of car plate,
The empirical value that obtains for statistics.
Optionally, the described vehicle body regional texture feature that utilizes, the accurate characteristic area of the body color of further search comprises:
(a) calculate the difference of thick area pixel gray scale in level and vertical direction;
(b) utilize improved quick Otsu technology, thick area pixel grey scale difference result is carried out binaryzation;
(c) according to the texture threshold value that sets, detect the accurate characteristic area of representing body color: threshold value is set is
,
,
(wherein,
Wide for the thick zone of body color;
Be body color zone texture threshold value in the horizontal direction;
With
Be the two-stage texture threshold value on the vertical direction, and
), bottom-up search on binary picture, add up each the row non-edge pixel quantity (being background pixel), will greater than
Line item get off, and the statistics continuously greater than
Row sum
By comparing
With threshold value
With
Value locate the smart zone of body color.
Optionally, described based on the synthetic new color space of HSV character
, further comprise: pixel RGB is converted to the HSV space; The correction of the unusual pixel of HSV color space; Pixel HSV after the correction synthesizes new color space
, its key step is as follows:
(a) the RGB color conversion is the HSV space: establish max=max(R, and G, B), and min=min(R, G, B); When max ≠ min, definition
H=60*h then, S=(max-min)/and max, V=max/255; As max=min, promptly during R=G=B, H=S=0 then, V=R/255.
(b) bearing calibration of the unusual pixel in HSV space: when
Then
,
,
When
And
Then
,
,
Parameter wherein
,
,
Be unusual pixel corrected threshold, obtain by the training sample statistics.
(c) the synthetic new color space in HSV space comprises behind the pixel correction:
,
,
Optionally, described employing fuzzy C-means clustering method further comprises in the vector color SPATIAL CALCULATION fuzzy histogram Bin normalization feature that quantizes:
(a) quantification of color space and Bin value are chosen: with color component
,
, V is quantified as respectively
,
,
Individual Bin, wherein the non-uniform quantizing mode is adopted in the V space, and
With
Uniform quantization is adopted in the space; Parameter
,
With
Then the best identified rate feedback according to the training sample test obtains an optimal value.
(b) expression of the three-dimensional color space synthesizing one-dimensional vector space after quantizing: the one dimension vector space be designed to Y=
+
+
, and
, wherein
With
Be respectively the quantification progression of color component C2 and V; Consider that in addition vehicle is in different photoenvironments, has incorporated monochrome information V in the expression of one dimension vector space.
(c) calculate fuzzy field histogram Bin normalization feature: establish
Be the normalization histogram feature in Y space, wherein n=
, be the proper vector dimension;
, N is a sum of all pixels,
For belonging to the total pixel number of i color Bin.According to probability theory,
Can be expressed as
, wherein
It is the conditional probability that j pixel belongs to i color Bin; Herein
Adopting fuzzy C-means clustering (FCM) method to calculate the fuzzy membership value that each pixel in the body color zone belongs to i color Bin obtains.
Optionally, the described sorter identification parameter that obtains the off-line sample training further comprises:
(a) some training samples of selection inhomogeneity body color;
(b) multidimensional characteristic vectors of calculating sample;
(c) utilize the LDA method that the high dimensional feature vector is carried out dimensionality reduction, and obtain the dimensionality reduction transition matrix of feature;
(d) adopt K-Means cluster or mixed Gauss model cluster (GMM) analytic approach, sample of all categories is carried out cluster, according to all kinds of feature templates numbers of test discrimination feedback adjusting of training sample and the optimal parameter of feature templates vector.
Optionally, the regional employing of described body color to vehicle image further comprises based on the feature templates coupling or the identification of SVM method of multiclass subspace:
(a) according to tone H value seven kinds of colors (white, grey, black, red, yellow, blue, green) are divided into 5 subclasses in the HSV space, each subclass is comprising different colour types;
(b) utilizing the LDA method to carry out the feature dimensionality reduction is to calculate the dimensionality reduction transition matrix at all categories color card, remove the color class sample that does not have in this subclass when just adopting feature templates coupling or SVM identification in the subclass space, the intrinsic dimensionality that i.e. all subclass spatial color identifications are adopted is identical, and the classification of each subclass space identification is counted difference; In all body color classes, discern relatively, a lot of less at the color class number of subspace identification, thus improved recognition speed and discrimination.
Optionally, described according to the color priori, preliminary recognition credibility is carried out secondary affirmation and correction than hanging down with the easy color of intersecting, further comprise:
(a) temporal mode that current vehicle image is in day and night carries out algorithm and judges: at first the low-light level pixel accounts for the ratio value of all pixels in computed image average brightness value and the image; The proportion threshold value of designed image mean flow rate threshold value and low-light level pixel then, and the brightness of image value of itself and current calculating and ratio value compared judge that vehicle is in daytime or night.
(b) color class that intersects of commute (as, mazarine and black, light red and yellow etc.) and just the lower color of recognition result confidence level carry out secondary and confirm and correction;
Need to prove that the judgment threshold of setting is that a large amount of vehicle image statistical computations according to different time sections obtain; For the vehicle image of video flowing, only need that the background image that obtains is carried out same procedure calculating and get final product; After the differentiation of day and night pattern mainly was based on the preliminary identification of body color, the correction parameter of recognition result that the day and night vehicle is set was different.
Optionally, described priori comprises utilizing easily intersect the proportionate relationship that restriction relation and vehicle body master color account between R, G, the B value of color in surveyed area.
Optionally, the confidence level of the first recognition result of described body color
, wherein d1 is a sorter minimum criteria value, d2 is a sorter time little criterion value; By more current recognition result confidence value that calculates and confidence level threshold value, whether the decision color correction.
The invention has the beneficial effects as follows, compared with prior art, the present invention can overcome different light environment, noise and body color zone and detect the influence of not accurate enough situation to body color identification, simultaneously because of adopting the LDA method to effective dimensionality reduction of high dimensional feature data with based on the identifying schemes in many subclasses of color space, thereby the raising recognition accuracy reduces system's operand.
Description of drawings
Fig. 1 is the schematic flow sheet of body color recognition methods in the embodiment of the present invention;
Fig. 2 is the schematic flow sheet that is used for body color characteristic area location in the embodiment of the present invention;
Fig. 3 is the feature extraction schematic flow sheet that is used for the body color surveyed area in the embodiment of the present invention;
Fig. 4 is the schematic flow sheet of body color multiclass subspace and identification in the embodiment of the present invention;
Fig. 5 is the sample training schematic flow sheet of step S160 in the body color identification shown in Figure 1;
Fig. 6 is that step S170 body color secondary is confirmed and the correcting process synoptic diagram in the body color identification shown in Figure 1.
Embodiment
The invention provides a kind of by thick body color area positioning method to essence, and by calculating effective multidimensional feature in body color zone, employing is based on the feature templates coupling identification body color of color multiclass subspace, meanwhile, to part easily intersect color (as, mazarine and black etc.) adopt color priori and the first method that combines of recognition result confidence level, carry out the secondary of day and night vehicle image body color identification respectively and confirm and correction, body color recognition result finally reliable to obtain, high-accuracy.The technical scheme that the present invention proposes has realized body color identification fast and accurately, for the intelligent traffic vehicle management automation provides new way.
Describe the embodiment of technical solution of the present invention in detail below in conjunction with drawings and Examples, mainly comprise following concrete steps.
1, detects vehicle body color characteristic area
Automobile body color characteristic zone is detected and has been adopted by thick to smart positioning strategy, as depicted in figs. 1 and 2.
1) Shu Ru original vehicle image
By step S100, the vehicle image of input can be static vehicle image, also can be for passing through the detected prospect vehicle image of moving object detection algorithm in the video sequence.
2) the thick zone location of body color feature
In step S200, obtain car plate height, wide and position coordinates according to the car plate location technology, establish the car plate height and width and be respectively H, W, be reference point with the license plate area center, intercept one up and highly be
, widely be
The rectangular area, wherein
The empirical value that obtains for statistics.
3) body color feature precise region location
In step S210, pixel grey scale adopts improved quick Otsu technology to carry out binaryzation to difference result in the difference of level and vertical direction then in the thick zone of calculating body color feature.In step S220, utilize the vehicle body textural characteristics to seek the accurate characteristic area of representing body color, smooth, smooth because of the body color zone, its textural characteristics shows and is large stretch of background pixel on the binary picture.Threshold value is set is
,
,
(wherein,
Wide for the thick zone of body color;
Be body color zone texture threshold value in the horizontal direction;
With
Be the two-stage texture threshold value on the vertical direction, and
), bottom-up search on binary picture, add up each the row non-edge pixel quantity (being background pixel), will greater than
Line item get off, and the statistics continuously greater than
Row sum
By comparing
With threshold value
With
Value locate the smart zone of body color.
2, characteristic area pixel color space conversion and vector space quantization means
1) pixel RGB conversion HSV color space and unusual pixel are proofreaied and correct
In the present embodiment vehicle body characteristic area pixel RGB among the step S300 is transformed into the HSV space in step S310, meets the characteristic of division of human eye vision to color, its concrete conversion formula is as follows, establishes max=max(R, G, B), and min=min(R, G, B); When max ≠ min, definition
H=60*h then, S=(max-min)/and max, V=max/255; As max=min, promptly during R=G=B, H=S=0 then; V=R/255.
The HSV space shows as hemicone, and it is unusual pixel to occur at the two ends of V component, based on this, after pixel RGB is transformed into the HSV space, need proofread and correct the unusual pixel in HSV space, and concrete bearing calibration is as follows: when
, then
,
,
When
And
, then
,
,
Parameter wherein
,
,
Be unusual pixel corrected threshold, obtain by the training sample statistics.
2) synthetic new color space
HSV is non-homogeneous color space, is unfavorable for the identification of color characteristic, for body color is carried out effective feature identification, in step S320, a kind of new even color space is set up in the hsv color space after the unusual pixel correction
, the pass of itself and HSV is
,
,
3) color space quantizes to represent with the one dimension vector space
In step S330, to new color space
The component of V
,
, V is quantified as respectively
,
,
Individual Bin, in the present embodiment, the V component adopts the non-uniform quantizing mode, and
With
Component adopts uniform quantization; Parameter
,
With
There is experiment test to obtain optimum value and is respectively 16,4,4.Three-dimensional color space synthesizing one-dimensional vector space after will quantizing then be expressed as Y=
+
+
, then
, consider that herein vehicle is in different photoenvironments, has incorporated monochrome information V in the expression of one dimension vector space.
3, calculate the normalization feature of the fuzzy histogram Bin in body color zone
Shown in step S340, in one dimension vector space Y, calculate the fuzzy histogram Bin normalization feature in body color zone.If
Be the normalization histogram proper vector in Y space, wherein n=
Be the proper vector dimension,
, N is a sum of all pixels,
For belonging to
The total pixel number of individual color Bin.According to probability theory,
Can be expressed as follows:
,
Be
Individual pixel belongs to
The conditional probability of individual color Bin; Herein
Adopt fuzzy C-means clustering (FCM) method to calculate in the body color zone each pixel and belong to the
The fuzzy membership value of individual color Bin obtains.In the present embodiment, test by experiment, the body color discrimination is the highest when calculating 256 dimensional features.
4, characteristic dimensionality reduction
Relative principal component analysis (PCA) (PCA), linear discriminant analysis method (LDA) is not only to illumination-insensitive, also when removing the characteristic correlativity, considered the characteristic of division between different classes of data, extraction helps differentiating the feature of classification information, for this this example in step S130, adopt LDA that the high dimensional feature that calculates is carried out dimensionality reduction, obtain low dimensional feature vector, make sorter recognition speed and discrimination improve.
5, obtain the sorter identification parameter of off-line sample training
In step S160, need use the sorter identification parameter of off-line sample training during body color identification, these parameters mainly comprise the template number of feature dimensionality reduction transition matrix, multi-template proper vector, each color class etc.These parameter amounts are that the off-line training by the body color sample obtains, specifically as shown in Figure 5.
It should be noted that, step S510 to S520 in Fig. 5, identical with step S120 to S130 shown in Figure 1, promptly the body color pixel is carried out color space conversion and feature extraction and LDA dimensionality reduction.
In step S530, utilize K-Means cluster or mixed Gauss model cluster (GMM) analytic approach training sample of all categories to be carried out the training of cluster identification.
In step S550, according to the discrimination feedback of body color training sample, adjust the parameter that influences the training sample discrimination, mainly comprise the dimensionality reduction matrix of feature, the template number and the template corresponding proper vector of different colours.When the discrimination of test sample book meets the demands, obtain sorter training parameter the best, just preserved.
6, the many feature templates couplings or the SVM identification of body color multiclass subspace
In step S140, color is divided into 5 sub-space-likes according to tone H different range value: (white, ash, black, red, Huang); (white, ash, black, Huang, green); (white, ash, black, green, indigo plant); (white, ash, black, indigo plant); (white, ash, black, red, indigo plant), color to be identified is divided into corresponding subspace earlier before sending into sorter identification; Utilize LDA method feature dimensionality reduction still all categories color card to be carried out, remove the color class sample that does not have in this subclass when just discerning in the subclass space, promptly the final intrinsic dimensionality that adopts of all each subclass spatial color identifications is identical, and the classification number of each subclass space identification is then different; In all body color classes, discern relatively, less at the color class number of subspace identification, thus improve recognition speed and discrimination.Particular flow sheet can be consulted Fig. 4.
In step S150, adopt the identification of many feature templates couplings in the identification of all kinds of subspaces, the sorter criterion function is an Euclidean distance, and the feature templates proper vector of each color class has been set up when sample training, and the feature templates quantity of each color class is obtained automatically according to the distribution of each color card; The recognition methods of adopting in the present embodiment is not limited to many feature templates couplings, also can adopt the SVM method directly to discern in the subspace.
7, the body color correction and the affirmation of recognition result just
Be subjected to the influence of factors such as video camera imaging quality and illumination variation, make more approaching color own intersect easily, as mazarine and black, light red and yellow etc. cause sorter that mistake identification takes place when these colors of identification.In addition, same color vehicle by day with embody very big difference night, also causing the same color of day and night is mistake identification.Consider based on above factor, the present invention is on the basis of first recognition result, color combining priori and recognition credibility, the commute intersection vehicle color and the vehicle color at daytime and night are proofreaied and correct respectively and are confirmed, set up the body color recognition system of high robust.Idiographic flow as shown in Figure 6.
1) temporal mode of vehicle image is differentiated
In step S610, the temporal mode of current vehicle image is judged the two kinds of situations that mainly contain: the first, morning 5 vehicle images to 7 these time periods of night; The second, the vehicle image of 7 o'clock to second day 5:00 AMs at night in the time period.For second kind of situation, the temporal mode of vehicle image all is set to night, so the discriminatory analysis of the temporal mode of vehicle image proposes at first kind of situation in the present embodiment.Account for the ratio value of all pixels by low-light level pixel in computed image average brightness value and the histogram, usually daytime the mean picture brightness value with night differ bigger, and in the daytime image high luminance pixel accounting example greatly, night is then opposite; According to this attribute, the proportion threshold value of setting mean picture brightness threshold value and low-light level pixel can judge daytime and evening pattern; For improving accuracy rate, preset threshold is obtained by a large amount of vehicle image statistics of different time sections; For video flowing, only need that the background image that obtains is carried out same procedure calculating and get final product.
2) correction of body color recognition result and affirmation
In step S650, the rgb value restriction relation of color mainly refers to R between different colours, G, B ratio varies in size, and for example the light red obscured of commute has following relation with yellow rgb value: yellow R and G value are quite, and it is all big than B, and red G and B value are suitable, and all little than the R value, and the difference of red R and G is greater than the difference between G and B, and light red is opposite, other similar successively calculating; Then according to the confidence level of the recognition result that calculates
Wherein d1 is a sorter minimum criteria value, d2 is a sorter time little criterion value, the low difference of recognition credibility is intersected color ratio value that parameter threshold and auxiliary main color account in the vehicle body characteristic area is set first recognition result is proofreaied and correct and is confirmed, to obtain final body color recognition result.
Body color automatic identifying method in the Vehicular intelligent supervisory system provided by the invention, simple, many feature templates couplings based on LDA feature dimensionality reduction and multiclass color sub-spaces, reduce calculated amount, thereby obtain the body color recognition result fast, meanwhile, utilize color priori etc., commute intersection and the first low color of recognition credibility are proofreaied and correct and are confirmed, improve discrimination, for the reliability of the automatic identification of body color in the Vehicular intelligent supervisory system is given security.
Embodiment disclosed in this invention as above; but be not in order to limit the present invention; any those skilled in the art are not in breaking away from spirit and scope disclosed by the invention; all can implement possible variation and modification; but scope of patent protection of the present invention still should be as the criterion with the scope that appending claims of the present invention was defined.
Claims (9)
1. body color automatic identifying method in the Vehicular intelligent supervisory system is characterized in that this method may further comprise the steps:
(1), above car plate, detects the thick zone of representing body color with reference to car plate height, wide and location coordinate information by car plate location technology gained;
(2), in the thick zone of detecting of body color, further search for the accurate characteristic area of body color according to the textural characteristics in vehicle body zone;
(3) body color characteristic area rgb pixel is transformed into the HSV space, and based on the synthetic new color space of HSV character
(4) adopt the normalization feature of fuzzy C-means clustering method, utilize the LDA method, obtain the characteristic that helps color discrimination the feature dimensionality reduction at the vector color SPATIAL CALCULATION fuzzy histogram Bin that quantizes;
(5) according to the sorter identification parameter of off-line sample training, adopt feature templates to mate or the identification of SVM method, obtain preliminary body color result based on the multiclass subspace to vehicle image body color characteristic area;
(6) according to first recognition credibility and color priori, commute intersects and vehicle color with a low credibility is carried out the correction and the affirmation of day and night respectively, to obtain final body color recognition result.
2. according to body color automatic identifying method in the described Vehicular intelligent supervisory system of claim 1, it is characterized in that in the described step (1), on behalf of the thick zone of body color, described the detection above car plate be specially: the center with license plate area is a reference point, intercepts one up highly to be
, widely be
The rectangular area, wherein
Be respectively the height and width of car plate,
,
The empirical value that obtains for statistics.
3. according to body color automatic identifying method in the described Vehicular intelligent supervisory system of claim 1, it is characterized in that described step (2) comprises following substep:
(1) calculates the difference of thick area pixel gray scale in level and vertical direction;
(2) utilize improved quick Otsu technology, thick area pixel grey scale difference result is carried out binaryzation;
(3) according to the texture threshold value that sets, detect the accurate characteristic area of representing body color: threshold value is set is
,
,
(wherein,
Wide for the thick zone of body color;
Be body color zone texture threshold value in the horizontal direction;
With
Be the two-stage texture threshold value on the vertical direction, and
), bottom-up search on binary picture, add up each the row non-edge pixel quantity (being background pixel), will greater than
Line item get off, and the statistics continuously greater than
Row sum
By comparing
With threshold value
With
Value locate the smart zone of body color.
4. according to body color automatic identifying method in the described Vehicular intelligent supervisory system of claim 1, it is characterized in that described step (3) comprises following substep:
(1) area pixel RGB is converted to the HSV space, and its concrete grammar is as follows: establish max=max(R, and G, B), and min=min(R, G, B); When max ≠ min, definition:
H=60*h then, S=(max-min)/and max, V=max/255;
As max=min, promptly during R=G=B, H=S=0 then, V=R/255;
(2) correction of the unusual pixel of HSV color space, its bearing calibration is specially: when
Then
,
,
When
,
The time; Then
,
,
(3) the synthetic new color space of the HSV after the correction
:
,
,
5. according to body color automatic identifying method in the described Vehicular intelligent supervisory system of claim 1, it is characterized in that described step (4) comprises following substep:
(1) quantification of color space and Bin value are chosen: with color component
,
, V is quantified as respectively
,
,
Individual Bin, wherein the non-uniform quantizing mode is adopted in the V space, and
With
Uniform quantization is adopted in the space; Parameter
,
With
Then the best identified rate feedback according to the training sample test obtains an optimal value;
(2) expression of the three-dimensional color space synthesizing one-dimensional vector space after quantizing: the one dimension vector space be designed to Y=
+
+
, and
, wherein
With
Be respectively the quantification progression of color component C2 and V; Consider that in addition vehicle is in different photoenvironments, has incorporated monochrome information V in the expression of one dimension vector space;
(3) calculate fuzzy field histogram Bin normalization feature: establish
Be the normalization histogram feature in Y space, wherein n=
, be the proper vector dimension;
, N is a sum of all pixels,
For belonging to the total pixel number of i color Bin; According to probability theory,
Can be expressed as
, wherein
It is the conditional probability that j pixel belongs to i color Bin; Herein
Adopting fuzzy C-means clustering (FCM) method to calculate the fuzzy membership value that each pixel in the body color zone belongs to i color Bin obtains.
6. according to body color automatic identifying method in the described Vehicular intelligent supervisory system of claim 1, it is characterized in that in the described step (5), the sorter identification parameter of described off-line sample training obtains by following steps:
(1) some training samples of selection inhomogeneity body color;
(2) multidimensional characteristic vectors of calculating sample;
(3) utilize the LDA method that the high dimensional feature vector is carried out dimensionality reduction, and obtain the dimensionality reduction transition matrix of feature;
(4) adopt K-Means cluster or mixed Gauss model cluster (GMM) analytic approach, sample of all categories is carried out cluster, according to all kinds of feature templates numbers of test discrimination feedback adjusting of training sample and the optimal parameter of feature templates vector.
7. according to body color automatic identifying method in the described Vehicular intelligent supervisory system of claim 1, it is characterized in that described step (6) comprises following substep:
(1) temporal mode that current vehicle image is in day and night carries out algorithm and judges that its key step is as follows:
(a) the low-light level pixel accounts for the ratio value of all pixels in computed image average brightness value and the image;
(b) proportion threshold value of designed image mean flow rate threshold value and low-light level pixel, and the value of itself and current calculating compared judge that vehicle is in daytime or night;
(2) color class that intersects of commute (as, mazarine and black, light red and yellow etc.) and just the lower color of recognition result confidence level proofread and correct and confirm.
8. according to body color automatic identifying method in the described Vehicular intelligent supervisory system of claim 1, it is characterized in that, in the described step (6), described priori comprises: utilize the proportionate relationship that restriction relation and vehicle body master color account in surveyed area between R, the G easily obscure color, B value.
9. according to body color automatic identifying method in the described Vehicular intelligent supervisory system of claim 1, it is characterized in that in the described step (6), described body color is the confidence level of recognition result just
, wherein d1 is a sorter minimum criteria value, d2 is a sorter time little criterion value; By more current recognition result confidence value that calculates and confidence level threshold value, whether the decision color correction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110124540 CN102184413B (en) | 2011-05-16 | 2011-05-16 | Automatic vehicle body color recognition method of intelligent vehicle monitoring system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110124540 CN102184413B (en) | 2011-05-16 | 2011-05-16 | Automatic vehicle body color recognition method of intelligent vehicle monitoring system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102184413A true CN102184413A (en) | 2011-09-14 |
| CN102184413B CN102184413B (en) | 2013-06-12 |
Family
ID=44570586
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110124540 Active CN102184413B (en) | 2011-05-16 | 2011-05-16 | Automatic vehicle body color recognition method of intelligent vehicle monitoring system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102184413B (en) |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102542275A (en) * | 2011-12-15 | 2012-07-04 | 广州商景网络科技有限公司 | Automatic identification method for identification photo background and system thereof |
| CN102799872A (en) * | 2012-07-17 | 2012-11-28 | 西安交通大学 | Image processing method based on face image characteristics |
| CN102800199A (en) * | 2012-05-24 | 2012-11-28 | 武汉中科通达高新技术有限公司 | Automobile body color identifying system based on smart camera |
| CN103310201A (en) * | 2013-06-26 | 2013-09-18 | 武汉烽火众智数字技术有限责任公司 | Identification method for mixed colors of targets |
| CN103440503A (en) * | 2013-09-12 | 2013-12-11 | 青岛海信网络科技股份有限公司 | Vehicle body color detection and identification method |
| CN103473567A (en) * | 2013-08-27 | 2013-12-25 | 东莞中国科学院云计算产业技术创新与育成中心 | Vehicle detection method based on partial models |
| CN103996041A (en) * | 2014-05-15 | 2014-08-20 | 武汉睿智视讯科技有限公司 | Vehicle color identification method and system based on matching |
| CN104700620A (en) * | 2014-03-28 | 2015-06-10 | 杭州海康威视数字技术股份有限公司 | Traffic checkpoint-based method and device for recognizing fake-licensed vehicles |
| CN105184299A (en) * | 2015-08-29 | 2015-12-23 | 电子科技大学 | Vehicle body color identification method based on local restriction linearity coding |
| CN105354530A (en) * | 2015-09-22 | 2016-02-24 | 浙江宇视科技有限公司 | Vehicle body color identification method and apparatus |
| CN105635583A (en) * | 2016-01-27 | 2016-06-01 | 宇龙计算机通信科技(深圳)有限公司 | Shooting method and device |
| CN106326848A (en) * | 2016-08-17 | 2017-01-11 | 刘华英 | Attribute recognition method and device for traffic equipment |
| CN106408033A (en) * | 2016-10-09 | 2017-02-15 | 深圳市捷顺科技实业股份有限公司 | Vehicle body's smoothly sliding area positioning method and device |
| CN106529553A (en) * | 2016-10-27 | 2017-03-22 | 深圳市捷顺科技实业股份有限公司 | Vehicle body color recognition region positioning method and device |
| CN106971391A (en) * | 2017-02-24 | 2017-07-21 | 北京理工大学珠海学院 | A kind of multi-template positioning wicking detection method based on colour recognition |
| CN107133574A (en) * | 2017-04-12 | 2017-09-05 | 浙江大华技术股份有限公司 | A kind of vehicle feature recognition method and device |
| CN107292933A (en) * | 2017-04-27 | 2017-10-24 | 浙江工业大学 | A kind of vehicle color identification method based on BP neural network |
| CN107452039A (en) * | 2016-06-01 | 2017-12-08 | 上海东方传媒技术有限公司 | Compress the method and device of RGB color |
| CN107729812A (en) * | 2017-09-18 | 2018-02-23 | 南京邮电大学 | A kind of vehicle color suitable for monitoring scene knows method for distinguishing |
| WO2018107983A1 (en) * | 2016-12-16 | 2018-06-21 | 深圳市中达瑞和科技有限公司 | Colour recognition method and system |
| CN108256521A (en) * | 2017-12-29 | 2018-07-06 | 济南中维世纪科技有限公司 | For the effective coverage localization method of body color identification |
| CN108280415A (en) * | 2018-01-17 | 2018-07-13 | 武汉理工大学 | Driving behavior recognition methods based on intelligent mobile terminal |
| CN108335507A (en) * | 2017-01-17 | 2018-07-27 | 星克跃尔株式会社 | Utilize the driving guiding providing method and device of the filmed image of camera |
| CN108600746A (en) * | 2018-05-22 | 2018-09-28 | 重庆瑞景信息科技有限公司 | The eclipsed detection method of Color image of visual, system and device |
| CN109087363A (en) * | 2018-06-28 | 2018-12-25 | 江南大学 | A kind of sewage discharge detection method based on hsv color space |
| CN109583349A (en) * | 2018-11-22 | 2019-04-05 | 北京市首都公路发展集团有限公司 | A kind of method and system for being identified in color of the true environment to target vehicle |
| CN109993213A (en) * | 2019-03-06 | 2019-07-09 | 上海工程技术大学 | A kind of automatic identifying method for garment elements figure |
| CN110163291A (en) * | 2019-05-28 | 2019-08-23 | 北京史河科技有限公司 | A kind of indicator light recognition methods and device |
| US10423855B2 (en) | 2017-03-09 | 2019-09-24 | Entit Software Llc | Color recognition through learned color clusters |
| CN110349069A (en) * | 2019-08-07 | 2019-10-18 | 北京桔电出行科技有限公司 | Automobile welcomes the emperor lamp image display method, device, server, system, automobile and storage medium |
| CN110705338A (en) * | 2018-07-10 | 2020-01-17 | 浙江宇视科技有限公司 | Vehicle detection method and device and monitoring equipment |
| CN111860505A (en) * | 2020-07-23 | 2020-10-30 | 北京以萨技术股份有限公司 | Testing method, system, terminal and medium for license plate color recognition accuracy |
| CN111915779A (en) * | 2020-07-31 | 2020-11-10 | 浙江大华技术股份有限公司 | Gate control method, device, equipment and medium |
| CN112883857A (en) * | 2021-02-05 | 2021-06-01 | 南京恩瑞特实业有限公司 | Method for recognizing low-cloud large-fog-drop area by NRIET satellite |
| CN113780306A (en) * | 2021-08-11 | 2021-12-10 | 北京工业大学 | Waste mobile phone color identification method based on deep convolutional neural network |
| CN117853937A (en) * | 2024-03-08 | 2024-04-09 | 吉林农业大学 | Rice disease identification method and system based on secondary color cluster analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060204082A1 (en) * | 2005-03-11 | 2006-09-14 | Honeywell International Inc. | Fusion of color space data to extract dominant color |
| US20090097746A1 (en) * | 2006-06-29 | 2009-04-16 | Fujitsu Limited | Color classification method, color recognition method, color classification apparatus, color recognition apparatus, color recognition system, computer program, and recording medium |
| CN101436252A (en) * | 2008-12-22 | 2009-05-20 | 北京中星微电子有限公司 | Method and system for recognizing vehicle body color in vehicle video image |
| CN101630361A (en) * | 2008-12-30 | 2010-01-20 | 北京邮电大学 | Plate number, body color and mark identification-based equipment and plate number, body color and mark identification-based method for identifying fake plate vehicles |
| CN102034080A (en) * | 2009-09-24 | 2011-04-27 | 北京汉王智通科技有限公司 | Vehicle color identification method and device |
-
2011
- 2011-05-16 CN CN 201110124540 patent/CN102184413B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060204082A1 (en) * | 2005-03-11 | 2006-09-14 | Honeywell International Inc. | Fusion of color space data to extract dominant color |
| US20090097746A1 (en) * | 2006-06-29 | 2009-04-16 | Fujitsu Limited | Color classification method, color recognition method, color classification apparatus, color recognition apparatus, color recognition system, computer program, and recording medium |
| CN101436252A (en) * | 2008-12-22 | 2009-05-20 | 北京中星微电子有限公司 | Method and system for recognizing vehicle body color in vehicle video image |
| CN101630361A (en) * | 2008-12-30 | 2010-01-20 | 北京邮电大学 | Plate number, body color and mark identification-based equipment and plate number, body color and mark identification-based method for identifying fake plate vehicles |
| CN102034080A (en) * | 2009-09-24 | 2011-04-27 | 北京汉王智通科技有限公司 | Vehicle color identification method and device |
Cited By (54)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102542275B (en) * | 2011-12-15 | 2014-04-23 | 广州商景网络科技有限公司 | Automatic identification method for identification photo background and system thereof |
| CN102542275A (en) * | 2011-12-15 | 2012-07-04 | 广州商景网络科技有限公司 | Automatic identification method for identification photo background and system thereof |
| CN102800199A (en) * | 2012-05-24 | 2012-11-28 | 武汉中科通达高新技术有限公司 | Automobile body color identifying system based on smart camera |
| CN102799872A (en) * | 2012-07-17 | 2012-11-28 | 西安交通大学 | Image processing method based on face image characteristics |
| CN102799872B (en) * | 2012-07-17 | 2015-04-29 | 西安交通大学 | Image processing method based on face image characteristics |
| CN103310201A (en) * | 2013-06-26 | 2013-09-18 | 武汉烽火众智数字技术有限责任公司 | Identification method for mixed colors of targets |
| CN103310201B (en) * | 2013-06-26 | 2016-03-23 | 武汉烽火众智数字技术有限责任公司 | The recognition methods of target blend color |
| CN103473567A (en) * | 2013-08-27 | 2013-12-25 | 东莞中国科学院云计算产业技术创新与育成中心 | Vehicle detection method based on partial models |
| CN103473567B (en) * | 2013-08-27 | 2016-09-14 | 东莞中国科学院云计算产业技术创新与育成中心 | A kind of vehicle checking method based on department pattern |
| CN103440503A (en) * | 2013-09-12 | 2013-12-11 | 青岛海信网络科技股份有限公司 | Vehicle body color detection and identification method |
| CN103440503B (en) * | 2013-09-12 | 2016-05-11 | 青岛海信网络科技股份有限公司 | The recognition methods of a kind of automobile body color detection |
| CN104700620B (en) * | 2014-03-28 | 2018-06-22 | 杭州海康威视数字技术股份有限公司 | A kind of fake license plate vehicle recognition method and device based on traffic block port |
| CN104700620A (en) * | 2014-03-28 | 2015-06-10 | 杭州海康威视数字技术股份有限公司 | Traffic checkpoint-based method and device for recognizing fake-licensed vehicles |
| CN103996041B (en) * | 2014-05-15 | 2015-07-22 | 武汉睿智视讯科技有限公司 | Vehicle color identification method and system based on matching |
| CN103996041A (en) * | 2014-05-15 | 2014-08-20 | 武汉睿智视讯科技有限公司 | Vehicle color identification method and system based on matching |
| CN105184299A (en) * | 2015-08-29 | 2015-12-23 | 电子科技大学 | Vehicle body color identification method based on local restriction linearity coding |
| CN105354530A (en) * | 2015-09-22 | 2016-02-24 | 浙江宇视科技有限公司 | Vehicle body color identification method and apparatus |
| CN105354530B (en) * | 2015-09-22 | 2019-07-16 | 浙江宇视科技有限公司 | A kind of body color recognition methods and device |
| CN105635583A (en) * | 2016-01-27 | 2016-06-01 | 宇龙计算机通信科技(深圳)有限公司 | Shooting method and device |
| CN107452039A (en) * | 2016-06-01 | 2017-12-08 | 上海东方传媒技术有限公司 | Compress the method and device of RGB color |
| CN106326848A (en) * | 2016-08-17 | 2017-01-11 | 刘华英 | Attribute recognition method and device for traffic equipment |
| CN106408033A (en) * | 2016-10-09 | 2017-02-15 | 深圳市捷顺科技实业股份有限公司 | Vehicle body's smoothly sliding area positioning method and device |
| CN106408033B (en) * | 2016-10-09 | 2020-01-03 | 深圳市捷顺科技实业股份有限公司 | Method and equipment for positioning smooth area of vehicle body |
| CN106529553B (en) * | 2016-10-27 | 2020-01-03 | 深圳市捷顺科技实业股份有限公司 | Method and device for positioning vehicle body color identification area |
| CN106529553A (en) * | 2016-10-27 | 2017-03-22 | 深圳市捷顺科技实业股份有限公司 | Vehicle body color recognition region positioning method and device |
| WO2018107983A1 (en) * | 2016-12-16 | 2018-06-21 | 深圳市中达瑞和科技有限公司 | Colour recognition method and system |
| CN108335507A (en) * | 2017-01-17 | 2018-07-27 | 星克跃尔株式会社 | Utilize the driving guiding providing method and device of the filmed image of camera |
| CN106971391A (en) * | 2017-02-24 | 2017-07-21 | 北京理工大学珠海学院 | A kind of multi-template positioning wicking detection method based on colour recognition |
| US10423855B2 (en) | 2017-03-09 | 2019-09-24 | Entit Software Llc | Color recognition through learned color clusters |
| CN107133574B (en) * | 2017-04-12 | 2022-06-17 | 浙江大华技术股份有限公司 | Vehicle feature identification method and device |
| CN107133574A (en) * | 2017-04-12 | 2017-09-05 | 浙江大华技术股份有限公司 | A kind of vehicle feature recognition method and device |
| CN107292933B (en) * | 2017-04-27 | 2020-09-15 | 浙江工业大学 | Vehicle color identification method based on BP neural network |
| CN107292933A (en) * | 2017-04-27 | 2017-10-24 | 浙江工业大学 | A kind of vehicle color identification method based on BP neural network |
| CN107729812A (en) * | 2017-09-18 | 2018-02-23 | 南京邮电大学 | A kind of vehicle color suitable for monitoring scene knows method for distinguishing |
| CN107729812B (en) * | 2017-09-18 | 2021-06-25 | 南京邮电大学 | Method suitable for recognizing vehicle color in monitoring scene |
| CN108256521A (en) * | 2017-12-29 | 2018-07-06 | 济南中维世纪科技有限公司 | For the effective coverage localization method of body color identification |
| CN108256521B (en) * | 2017-12-29 | 2021-06-22 | 济南中维世纪科技有限公司 | Effective area positioning method for vehicle body color identification |
| CN108280415A (en) * | 2018-01-17 | 2018-07-13 | 武汉理工大学 | Driving behavior recognition methods based on intelligent mobile terminal |
| CN108600746B (en) * | 2018-05-22 | 2019-03-29 | 重庆瑞景信息科技有限公司 | The eclipsed detection method of Color image of visual, system and device |
| CN108600746A (en) * | 2018-05-22 | 2018-09-28 | 重庆瑞景信息科技有限公司 | The eclipsed detection method of Color image of visual, system and device |
| CN109087363A (en) * | 2018-06-28 | 2018-12-25 | 江南大学 | A kind of sewage discharge detection method based on hsv color space |
| CN109087363B (en) * | 2018-06-28 | 2021-12-21 | 江南大学 | HSV color space-based sewage discharge detection method |
| CN110705338A (en) * | 2018-07-10 | 2020-01-17 | 浙江宇视科技有限公司 | Vehicle detection method and device and monitoring equipment |
| CN109583349A (en) * | 2018-11-22 | 2019-04-05 | 北京市首都公路发展集团有限公司 | A kind of method and system for being identified in color of the true environment to target vehicle |
| CN109993213A (en) * | 2019-03-06 | 2019-07-09 | 上海工程技术大学 | A kind of automatic identifying method for garment elements figure |
| CN110163291A (en) * | 2019-05-28 | 2019-08-23 | 北京史河科技有限公司 | A kind of indicator light recognition methods and device |
| CN110349069A (en) * | 2019-08-07 | 2019-10-18 | 北京桔电出行科技有限公司 | Automobile welcomes the emperor lamp image display method, device, server, system, automobile and storage medium |
| CN111860505A (en) * | 2020-07-23 | 2020-10-30 | 北京以萨技术股份有限公司 | Testing method, system, terminal and medium for license plate color recognition accuracy |
| CN111915779A (en) * | 2020-07-31 | 2020-11-10 | 浙江大华技术股份有限公司 | Gate control method, device, equipment and medium |
| CN112883857A (en) * | 2021-02-05 | 2021-06-01 | 南京恩瑞特实业有限公司 | Method for recognizing low-cloud large-fog-drop area by NRIET satellite |
| CN113780306A (en) * | 2021-08-11 | 2021-12-10 | 北京工业大学 | Waste mobile phone color identification method based on deep convolutional neural network |
| CN113780306B (en) * | 2021-08-11 | 2024-04-09 | 北京工业大学 | Deep convolutional neural network-based waste mobile phone color recognition method |
| CN117853937A (en) * | 2024-03-08 | 2024-04-09 | 吉林农业大学 | Rice disease identification method and system based on secondary color cluster analysis |
| CN117853937B (en) * | 2024-03-08 | 2024-06-04 | 吉林农业大学 | Rice disease identification method and system based on secondary color cluster analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102184413B (en) | 2013-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102184413B (en) | Automatic vehicle body color recognition method of intelligent vehicle monitoring system | |
| Tian et al. | Rear-view vehicle detection and tracking by combining multiple parts for complex urban surveillance | |
| CN101789080B (en) | Detection method for vehicle license plate real-time positioning character segmentation | |
| CN101436252A (en) | Method and system for recognizing vehicle body color in vehicle video image | |
| CN103345766A (en) | Method and device for identifying signal light | |
| CN108335509A (en) | A kind of bus or train route cooperative system and method based on machine vision | |
| CN105160691A (en) | Color histogram based vehicle body color identification method | |
| CN103440503B (en) | The recognition methods of a kind of automobile body color detection | |
| CN101950350A (en) | Clear path detection using a hierachical approach | |
| Le et al. | Real time traffic sign detection using color and shape-based features | |
| CN111553214B (en) | Method and system for detecting smoking behavior of driver | |
| CN107016362B (en) | Vehicle weight recognition method and system based on vehicle front windshield pasted mark | |
| CN106529461A (en) | Vehicle model identifying algorithm based on integral characteristic channel and SVM training device | |
| CN104036246A (en) | Lane line positioning method based on multi-feature fusion and polymorphism mean value | |
| CN102737221B (en) | Method and apparatus for vehicle color identification | |
| CN102915433A (en) | Character combination-based license plate positioning and identifying method | |
| CN105224945B (en) | A kind of automobile logo identification method based on joint-detection and identification algorithm | |
| JP2020064583A (en) | Vehicle detection method, nighttime vehicle detection method based on dynamic light intensity, and system for the same | |
| CN107729812A (en) | A kind of vehicle color suitable for monitoring scene knows method for distinguishing | |
| CN102880863A (en) | Method for positioning license number and face of driver on basis of deformable part model | |
| CN103646241A (en) | Real-time taxi identification method based on embedded system | |
| CN106203338B (en) | Human eye state method for quickly identifying based on net region segmentation and threshold adaptive | |
| CN108520528B (en) | Mobile vehicle tracking method based on improved difference threshold and displacement matching model | |
| CN109191495A (en) | Black smoke vehicle detection method based on self-organizing background subtraction model and multiple features fusion | |
| CN109190455A (en) | Black smoke vehicle recognition methods based on Gaussian Mixture and autoregressive moving-average model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C53 | Correction of patent for invention or patent application | ||
| CB03 | Change of inventor or designer information |
Inventor after: Chen Yichao Inventor after: Pan Shizhu Inventor after: Zhang Xingming Inventor after: Fu Liquan Inventor after: Zhu Jiangming Inventor after: Wu Jun Inventor after: Wu Jian Inventor before: Chen Yichao Inventor before: Zhang Xingming Inventor before: Fu Liquan Inventor before: Zhu Jiangming Inventor before: Wu Jun Inventor before: Wu Jian |
|
| COR | Change of bibliographic data |
Free format text: CORRECT: INVENTOR; FROM: CHEN YICHAO ZHANG XINGMING FU LIQUAN ZHU JIANGMING WU JUN WU JIAN TO: CHENYICHAO PAN SHIZHU ZHANG XINGMING FU LIQUAN ZHU JIANGMING WU JUN WU JIAN |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |