CN103455791A - Method for recognizing vehicle types based on linear array CCD automobile chassis imaging - Google Patents
Method for recognizing vehicle types based on linear array CCD automobile chassis imaging Download PDFInfo
- Publication number
- CN103455791A CN103455791A CN2013103568021A CN201310356802A CN103455791A CN 103455791 A CN103455791 A CN 103455791A CN 2013103568021 A CN2013103568021 A CN 2013103568021A CN 201310356802 A CN201310356802 A CN 201310356802A CN 103455791 A CN103455791 A CN 103455791A
- Authority
- CN
- China
- Prior art keywords
- automobile chassis
- image
- overbar
- curve
- identified
- 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)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method for recognizing vehicle types based on linear array CCD automobile chassis imaging. The method comprises the first step of detecting the edges of an image in an automobile chassis area, the second step of projecting the edge image horizontally and vertically, the third step of smoothing projection curves, the fourth step of positioning and cutting an automobile chassis according to the characters of the smoothed projection curves, the fifth step of establishing a standard library of the automobile chassis image, the sixth step of matching the automobile chassis image of the standard library and an automobile chassis image to be recognized, the seventh step of adjusting one projection curve of the unbalanced automobile running speed, the eighth step of judging the similarity between the automobile chassis to be recognized and the automobile chassis of the standard library, and the ninth step of obtaining the vehicle type recognition result of the automobile chassis. By means of the method, the types of the automobile chassis can be accurately recognized.
Description
Technical field
The invention belongs to the image recognition technology field, relate to a kind of model recognizing method based on the imaging of line array CCD automobile chassis.
Background technology
The automatic detection mode of automobile chassis, can, in safety-protection system, carry out detection and vehicle identification that automobile chassis is concealed the foreign matters such as dangerous material easily.
Existing vehicle safety check place, detection mode for automobile chassis, be mainly to rely on manual detection and manual instrument and equipment to be detected for a long time, the part place also needs the staff to utilize mirror image observation automobile chassis whether to carry contraband goods secretly, also has department to be banned by police dog.This just level method is wasted time and energy, inefficiency.
Summary of the invention
The purpose of this invention is to provide a kind of model recognizing method based on the imaging of line array CCD automobile chassis, solved the manual type observation automobile chassis existed in the prior art, waste time and energy, ineffective problem.
The technical solution adopted in the present invention is, a kind of model recognizing method based on the imaging of line array CCD automobile chassis is implemented according to following steps:
Step 1, the image in automobile chassis zone is carried out to rim detection;
Step 2, edge image carry out level and vertical projection;
Step 3, drop shadow curve is carried out to smoothing processing;
Step 4, according to the drop shadow curve's feature after level and smooth, automobile chassis is positioned and cuts apart;
The registration of step 6, java standard library automobile chassis image and automobile chassis image to be identified;
Step 7, the unbalanced drop shadow curve of automobile driving speed is adjusted;
Step 8, judge the similarity of automobile chassis to be identified and java standard library automobile chassis;
Step 9, obtain the vehicle recognition result of automobile chassis.
The invention has the beneficial effects as follows, the view data collected from line array CCD accurately, zone, running automobile chassis is positioned, extract the correlated characteristic of this automobile chassis, fast the automobile chassis feature in itself and java standard library is compared, find vehicle classification or the unusual circumstance of automobile chassis.
The accompanying drawing explanation
Fig. 1 is the horizontal projection curve of the automobile chassis image of the inventive method after to the horizontal direction sharpening;
Fig. 2 is the vertical projection curve of the automobile chassis image of the inventive method after to the horizontal direction sharpening;
Fig. 3 is the horizontal projection curve smoothing result curve of the inventive method to Fig. 1;
Fig. 4 is the vertical projection curve smoothing result curve of the inventive method to Fig. 2;
Fig. 5 is the vertical projection curve of the inventive method after to the same model registration;
Fig. 6 is the drop shadow curve of the inventive method after to the different automobile types registration;
Fig. 7 is the correlation curve of the inventive method before to the unbalanced adjustment of automobile driving speed;
Fig. 8 is the correlation curve of the inventive method after to the unbalanced adjustment of automobile driving speed.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is described in detail.
Because gather the pattern of automobile chassis view data for buried line array CCD mode, in the view data obtained, comprise the part sheltered from by automobile chassis, be called the automobile chassis zone in following text, and the part of not blocked by automobile chassis, be called background area in following text.
Model recognizing method based on the imaging of line array CCD automobile chassis of the present invention, due to automobile through buried line array CCD zone the time, the actual route the crossed predetermined paths that can not fit like a glove, so need to position the automobile chassis zone, eliminate the impact of background on vehicle identification correctness.Afterwards, feature extraction is carried out in the automobile chassis zone of orienting, and, according to the feature of extracted different automobile types, carry out vehicle identification.Specifically according to following steps, implement:
In order to resist the inconsistency of vehicle running path, the automobile chassis image that carries out vehicle identification is all to have removed the image after the background, be that automobile chassis image in java standard library only has body part, and actual photographed to test pattern comprise the background information of redundancy, therefore carry out at first will from the image collected, extracting needed body part before automobile chassis identification, to realize the extraction of cutting apart of automobile chassis image
Step 1, the image in automobile chassis zone is carried out to rim detection
Speed while considering running car is difficult to control to constant, therefore, when the line array CCD imaging, is bound to occur stretching or the compression on vertical direction, so, the edge of the horizontal direction of the automobile chassis area image that only extraction collects here,
If the automobile chassis area image collected is [I (x, y)]
m * n, size is m * n, its edge image is
computing formula is as follows:
what mean is the matrix of a m * n size, and in formula (1), what provide is all elements in this matrix
x=1,2 ..., m; Y=1,2 ..., the computing formula of n, in other words,
with
what mean is not same concept,
Step 2, edge image carry out level and vertical projection
To the outline map obtained by step 1
ask its horizontal projection P
x(x), x=1,2,3 ..., m, and vertical projection P
y(y), y=1,2,3 ..., n, computing formula is as follows:
Image according to above-mentioned formula after to horizontal sharpening carries out projection, the results are shown in Figure 1, Fig. 2;
Step 3, drop shadow curve is carried out to smoothing processing
The drop shadow curve that step 2 is obtained carries out smoothing processing according to following formula, obtains drop shadow curve after smoothing processing
with
x=1,2,3 ..., m, y=1,2,3 ..., n, computing formula is as follows:
Wherein, the size that q is smooth window, be set to odd number, the experience value, and span is q ∈ [11,21]; Fig. 1 and Fig. 2 are carried out to effect curve after level and smooth respectively with reference to Fig. 3 and Fig. 4;
Step 4, according to the drop shadow curve's feature after level and smooth, automobile chassis is positioned and cuts apart
4.1) data for projection after level and smooth is carried out to difference processing, obtain
with
x=1,2,3 ..., m-1, y=1,2,3 ..., n-1, computing formula is as follows:
4.2) to height displacement's component curve
x=1,2,3 ..., m-1, respectively from left to right, from right to left, to height displacement's component curve
y=1,2,3 ..., n-1, respectively from top to bottom, judged from top to bottom, judge two sections mild places of curve and be background parts, and concrete judgment rule is as follows:
Coordinate x=x for drop shadow curve
0if,
and
k=1,2,3 ..., x
0-1, x
0it is the coboundary of automobile chassis;
Coordinate x=x for drop shadow curve
1if,
and
k=1,2,3 ..., m-x
1, x
1it is the lower boundary of automobile chassis;
Coordinate y=y for drop shadow curve
0if,
and
k=1,2,3 ..., y
0-1, y
0it is the left margin of automobile chassis;
Coordinate y=y for drop shadow curve
1if,
and
k=1,2,3 ..., m-y
1, y
1it is the right margin of automobile chassis;
Wherein, ε is a judgment threshold, value rule of thumb, and span is
Coordinate (x by four points of resulting rectangle
0, y
0), (x
0, y
1), (x
1, y
0) and (x
1, y
1) substitution image [I (x, y)]
m * n, just the automobile chassis zone location can be cut apart to the automobile chassis area image that obtains being partitioned into;
Automobile chassis image by different styles, select to pollute few automobile chassis, according to step 1 after step 4 is carried out the automobile chassis Region Segmentation, the automobile chassis image after it is cut apart, with and drop shadow curve's data stored, in order to identification that automobile chassis to be measured is compared;
The registration of step 6, java standard library automobile chassis image and automobile chassis image to be identified
If the vertical projection curve of the java standard library automobile chassis image calculated according to formula (3) and formula (4) is
x=1,2,3 ..., m
s, the vertical projection curve of automobile chassis image to be identified is
x=1,2,3 ..., m
d, suppose m
s>=m
dif do not meet, when calculating, by the drop shadow curve of java standard library automobile chassis image and automobile chassis image projection curve to be identified displacement, replace once mutually, because one of drop shadow curve is long, one short, and the purpose of displacement is the convenience of calculation for formula (9)
6.1) ask for the correlation coefficient ρ of the drop shadow curve of the drop shadow curve of java standard library automobile chassis image and automobile chassis image to be identified
x ds(k), computing formula is as follows:
6.2) ask for the maximum of points of related coefficient, that is:
x
*=arg max{ρ
x ds(k)|k=0,1,...,m
d-m
s}, (10)
The registration position that obtains java standard library automobile chassis image and automobile chassis image to be identified is, the starting point coordinate (1,1) of java standard library automobile chassis image is corresponding to the coordinate (x*, 1) of automobile chassis image to be identified,
Through after registration, be the vertical projection curve registration result afterwards of java standard library automobile chassis image and automobile chassis image to be identified, can find out, when the vehicle of two cars is consistent, the consistance of its curve waveform is fine, as shown in Figure 5; When the vehicle of two cars is inconsistent, after carrying out the registration calculation process, the waveform of vertical projection curve still has larger difference, as shown in Figure 6;
Step 7, the unbalanced drop shadow curve of automobile driving speed is adjusted
Automobile, when travelling through check point, if speed is slower, will cause line array CCD imaging image stretch out, stretch image later when carrying out vertical projection, can cause peak value too high, affect matching result, for this reason, formula (7) is calculated to automobile chassis image to be identified
x=1,2,3 ..., m-1, adjust its vertical projection value according to following rules modification:
If
x=1,2,3 ..., m-1, by edge image
the capable deletion of x, eliminate stretch effects; Afterwards, then carry out vertical projection;
Adjust the above two waveform difference obviously, as shown in Figure 7;
Through after adjusting, more approaching with the waveform of the vertical projection curve of same money vehicle in java standard library, as shown in Figure 8.
Step 8, judge the similarity of automobile chassis to be identified and java standard library automobile chassis
The similarity characteristic parameter comprises two indexs, i.e. the vertical projection correlation coefficient ρ of automobile chassis and automobile chassis to be identified in java standard library
x ds(x
*) number and Euclidean distance d
x ds(x
*):
8.1) the vertical projection correlation coefficient ρ
x ds(x
*)
Obtained registration parameter x according to formula (10)
*afterwards, then process through the Delete Row of step 7, then return to the vertical projection curve that step 2 and step 3 are obtained automobile chassis to be identified
x=1,2,3., m
d, afterwards, by x
*bring formula (9) into, obtain correlation coefficient ρ
x ds(x
*);
8.2) Euclidean distance d
x ds(x
*)
Euclidean distance d
x ds(x
*) computing formula as follows:
Step 9, obtain the vehicle recognition result of automobile chassis
Successively automobile chassis to be identified and the automobile chassis in java standard library are carried out to the calculating of Similarity Parameter, obtain its correlation coefficient ρ
x ds(x
*) and Euclidean distance d
x ds(x
*),
Keep foreign matter in order to resist automobile chassis to be measured, or have when stained, the form of automobile chassis will change and cause the impact on the similarity parameter, carries out the vehicle identification of automobile chassis according to following rule:
9.1) related coefficient the maximum of automobile chassis to be identified and java standard library automobile chassis is considered as to possible vehicle;
9.2) by step 9.1) and in the drop shadow curve of the java standard library automobile chassis that obtains, the drop shadow curve with automobile chassis to be identified after registration, carry out point-to-point comparison, the judging point dot spacing is from the position of larger point,
Judging distance larger according to being the experience value, be preferably this distance and be greater than more than 30% of an amplitude, little if these account for apart from larger some the ratio that curve always counts, this ratio is the experience value, is preferably 10% to 20%, deletes these points; Otherwise, judge the vehicle of this vehicle to be identified for not listing in standard database;
9.3) will delete the curve of large range points, again ask its Euclidean distance according to formula (11), if enough little, judge that it is and step 9.1) vehicle same model in the java standard library that provides; Otherwise, judge the vehicle of this vehicle to be identified for not listing in standard database.
By above step, after the automobile chassis of vehicle to be identified and the automobile chassis in standard database are contrasted, identify the vehicle of vehicle to be identified.
Method of the present invention based on the line array CCD automobile chassis imaging carry out vehicle identification, by extracting the feature of automobile chassis data, with the feature of the automobile chassis data of known vehicle, carry out the similarity judge, realize thus vehicle is identified accurately.This method can be resisted the interference of the stained and foreign matter existence of automobile chassis.
Claims (10)
1. the model recognizing method based on the imaging of line array CCD automobile chassis, its characteristics are: according to following steps, implement:
Step 1, the image in automobile chassis zone is carried out to rim detection;
Step 2, edge image carry out level and vertical projection;
Step 3, drop shadow curve is carried out to smoothing processing;
Step 4, according to the drop shadow curve's feature after level and smooth, automobile chassis is positioned and cuts apart;
Step 5, set up the java standard library of automobile chassis image;
The registration of step 6, java standard library automobile chassis image and automobile chassis image to be identified;
Step 7, the unbalanced drop shadow curve of automobile driving speed is adjusted;
Step 8, judge the similarity of automobile chassis to be identified and java standard library automobile chassis;
Step 9, obtain the vehicle recognition result of automobile chassis.
2. the model recognizing method based on the imaging of line array CCD automobile chassis according to claim 1, its characteristics are: the detailed process of described step 1 is,
The edge of the horizontal direction of the automobile chassis area image that extraction collects, establishing the automobile chassis area image collected is [I (x, y)]
m * n, size is m * n, its edge image is
, computing formula is as follows:
3. the model recognizing method based on the imaging of line array CCD automobile chassis according to claim 1, its characteristics are: the detailed process of described step 2 is,
To the outline map obtained by step 1
, ask its horizontal projection P
x(x), x=1,2,3 ..., m, and vertical projection P
y(y), y=1,2,3 ..., n, computing formula is as follows:
4. the model recognizing method based on the imaging of line array CCD automobile chassis according to claim 1, its characteristics are: the detailed process of described step 3 is,
The drop shadow curve that step 2 is obtained carries out smoothing processing according to following formula, obtains drop shadow curve after smoothing processing
with
x=1,2,3 ..., m, y=1,2,3 ... n, computing formula is as follows:
Wherein, the size that q is smooth window, be set to odd number, and span is q ∈ [11,21].
5. the model recognizing method based on the imaging of line array CCD automobile chassis according to claim 1, its characteristics are: the detailed process of described step 4 is,
4.1) data for projection after level and smooth is carried out to difference processing, obtain
with
x=1,2,3 ..., m-1, y=1,2,3 ..., n-1, computing formula is as follows:
4.2) to height displacement's component curve
x=1,2,3 ..., m-1, respectively from left to right, from right to left, to height displacement's component curve
y=1,2,3 ..., n-1 respectively from top to bottom, judged from top to bottom, judges two sections mild places of curve and is background parts, and concrete judgment rule is as follows:
Coordinate x=x for drop shadow curve
0if,
and
k=1,2,3 ..., x
0-1, x
0it is the coboundary of automobile chassis;
Coordinate x=x for drop shadow curve
1if,
and
k=1,2,3 ..., m-x
1, x
1it is the lower boundary of automobile chassis;
Coordinate y=y for drop shadow curve
0if,
and
k=1,2,3 ..., y
0-1, y
0it is the left margin of automobile chassis;
Coordinate y=y for drop shadow curve
1if,
and
k=1,2,3., m-y
1, y
1it is the right margin of automobile chassis;
Coordinate (x by four points of resulting rectangle
0, y
0) (x
0, y
1) (x
1, y
0) and (x
1, y
1) substitution image [I (x, y)]
m * n, the automobile chassis area image that obtains being partitioned into.
6. the model recognizing method based on the imaging of line array CCD automobile chassis according to claim 1, its characteristics are: the detailed process of described step 5 is,
By the automobile chassis image of different styles, select to pollute few automobile chassis, according to step 1 after step 4 is carried out the automobile chassis Region Segmentation, the automobile chassis image after it is cut apart, with and drop shadow curve's data stored.
7. the model recognizing method based on the imaging of line array CCD automobile chassis according to claim 1, its characteristics are: the detailed process of described step 6 is,
If the vertical projection curve of the java standard library automobile chassis image calculated according to formula (3) and formula (4) is
x=1,2,3..., m
s, the vertical projection curve of automobile chassis image to be identified is
x=1,2,3 ..., m
d, suppose m
s>=m
dif, do not meet,, when calculating, the drop shadow curve of java standard library automobile chassis image and automobile chassis image projection curve to be identified are replaced,
6.1) ask for the correlation coefficient ρ of the drop shadow curve of the drop shadow curve of java standard library automobile chassis image and automobile chassis image to be identified
x ds(k), computing formula is as follows:
6.2) ask for the maximum of points of related coefficient, computing formula is as follows:
x
*=arg max{ρ
x ds(k)|k=0,1,...,m
d-m
s}, (10)
The registration position that obtains java standard library automobile chassis image and automobile chassis image to be identified is that the starting point coordinate (1,1) of java standard library automobile chassis image is corresponding to the coordinate (x of automobile chassis image to be identified
*, 1).
8. the model recognizing method based on the imaging of line array CCD automobile chassis according to claim 1, its characteristics are: the detailed process of described step 7 is,
Automobile, when travelling through check point, if speed is slower, will cause line array CCD imaging image stretch out, for this reason, formula (7) calculated to automobile chassis image to be identified
x=1,2,3 ..., m-1, according to its vertical projection value of following rule adjustment:
9. the model recognizing method based on the imaging of line array CCD automobile chassis according to claim 1, its characteristics are: the detailed process of described step 8 is,
The similarity characteristic parameter comprises two indexs, i.e. the vertical projection correlation coefficient ρ of automobile chassis and automobile chassis to be identified in java standard library
x ds(x
*) and Euclidean distance d
x ds(x
*):
8.1) the vertical projection correlation coefficient ρ
x ds(x
*)
Obtained registration parameter x according to formula (10)
*afterwards, then process through the Delete Row of step 7, then return to the vertical projection curve that step 2 and step 3 are obtained automobile chassis to be identified
x=1,2,3 ..., m
d, afterwards, by x
*bring formula (9) into, obtain correlation coefficient ρ
x ds(x
*);
8.2) Euclidean distance d
x ds(x
*)
Euclidean distance d
x ds(x
*) computing formula as follows:
10. the model recognizing method based on the imaging of line array CCD automobile chassis according to claim 1, its characteristics are: the detailed process of described step 9 is,
Successively automobile chassis to be identified and the automobile chassis in java standard library are carried out to the calculating of Similarity Parameter, obtain its correlation coefficient ρ
x ds(x
*) and Euclidean distance d
x ds(x
*),
When automobile chassis to be measured keeps foreign matter, or have when stained, carry out the vehicle identification of automobile chassis according to following rule:
9.1) related coefficient the maximum of automobile chassis to be identified and java standard library automobile chassis is considered as to possible vehicle;
9.2) by step 9.1) and in the drop shadow curve of the java standard library automobile chassis that obtains, drop shadow curve with automobile chassis to be identified after registration, carry out point-to-point comparison, the judging point dot spacing is from the position of larger point, judging distance larger according to being, be preferably this distance and be greater than more than 30% of an amplitude, little if these account for apart from larger some the ratio that curve always counts, this ratio is preferably 10% to 20%, deletes these points; Otherwise, judge the vehicle of this vehicle to be identified for not listing in standard database;
9.3) will delete the curve of large range points, again ask its Euclidean distance according to formula (11), if enough little, judge that it is and step 9.1) vehicle same model in the java standard library that provides;
Otherwise, judge the vehicle of this vehicle to be identified for not listing in standard database.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310356802.1A CN103455791B (en) | 2013-08-15 | 2013-08-15 | Model recognizing method based on line array CCD automobile chassis imaging |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310356802.1A CN103455791B (en) | 2013-08-15 | 2013-08-15 | Model recognizing method based on line array CCD automobile chassis imaging |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103455791A true CN103455791A (en) | 2013-12-18 |
CN103455791B CN103455791B (en) | 2016-10-05 |
Family
ID=49738135
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310356802.1A Expired - Fee Related CN103455791B (en) | 2013-08-15 | 2013-08-15 | Model recognizing method based on line array CCD automobile chassis imaging |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103455791B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105320705A (en) * | 2014-08-05 | 2016-02-10 | 北京大学 | Retrieval method and device for similar vehicle |
CN108919372A (en) * | 2018-04-23 | 2018-11-30 | 爱驰汽车有限公司 | A kind of vehicle bottom foreign matter method for early warning and system, car-mounted terminal |
CN109344677A (en) * | 2017-11-07 | 2019-02-15 | 长城汽车股份有限公司 | Identify method, apparatus, vehicle and the storage medium of stereoscopic article |
CN112418100A (en) * | 2020-11-24 | 2021-02-26 | 哈尔滨市科佳通用机电股份有限公司 | Road vehicle abnormity detection method based on image classification and image comparison |
-
2013
- 2013-08-15 CN CN201310356802.1A patent/CN103455791B/en not_active Expired - Fee Related
Non-Patent Citations (4)
Title |
---|
卢鹏辉: "基于面阵相机阵列的车底安全检查方法研究", 《中国优秀硕士学位论文信息科技辑》 * |
汪杰: "基于序列图像的车型识别技术研究", 《中国优秀硕士学位论文信息科技辑》 * |
罗棻: "基于投影的FMRI时间序列图像配准方法研究", 《电子科技大学学报》 * |
马政德: "遥感图像配准中相似度测度的比较与分析", 《计算机工程与科学》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105320705A (en) * | 2014-08-05 | 2016-02-10 | 北京大学 | Retrieval method and device for similar vehicle |
CN105320705B (en) * | 2014-08-05 | 2018-09-14 | 北京大学 | The search method and device of similar vehicle |
CN109344677A (en) * | 2017-11-07 | 2019-02-15 | 长城汽车股份有限公司 | Identify method, apparatus, vehicle and the storage medium of stereoscopic article |
CN109344677B (en) * | 2017-11-07 | 2021-01-15 | 长城汽车股份有限公司 | Method, device, vehicle and storage medium for recognizing three-dimensional object |
US11195305B2 (en) | 2017-11-07 | 2021-12-07 | Great Wall Motor Company Limited | Method and device for identifying stereoscopic object, and vehicle and storage medium |
CN108919372A (en) * | 2018-04-23 | 2018-11-30 | 爱驰汽车有限公司 | A kind of vehicle bottom foreign matter method for early warning and system, car-mounted terminal |
CN112418100A (en) * | 2020-11-24 | 2021-02-26 | 哈尔滨市科佳通用机电股份有限公司 | Road vehicle abnormity detection method based on image classification and image comparison |
Also Published As
Publication number | Publication date |
---|---|
CN103455791B (en) | 2016-10-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104951784B (en) | A kind of vehicle is unlicensed and license plate shading real-time detection method | |
CN105069415B (en) | Method for detecting lane lines and device | |
CN104197897B (en) | A kind of downtown roads graticule automatic classification method based on Vehicle-borne Laser Scanning point cloud | |
CN101859382B (en) | License plate detection and identification method based on maximum stable extremal region | |
CN102999749B (en) | Based on the securing band violation event intelligent detecting method of Face datection | |
CN106951879A (en) | Multiple features fusion vehicle checking method based on camera and millimetre-wave radar | |
CN108090429A (en) | Face bayonet model recognizing method before a kind of classification | |
CN103593981B (en) | A kind of model recognizing method based on video | |
CN108389430A (en) | A kind of intersection pedestrian based on video detection and collision of motor-driven vehicles prediction technique | |
CN103034861B (en) | The recognition methods of a kind of truck brake shoe breakdown and device | |
CN104573646A (en) | Detection method and system, based on laser radar and binocular camera, for pedestrian in front of vehicle | |
CN103854320A (en) | Automatic vehicle type identification device and method based on laser radar | |
CN103903018A (en) | Method and system for positioning license plate in complex scene | |
CN104123531B (en) | A kind of method and apparatus of palmmprint interest zone location | |
CN104881661B (en) | Vehicle checking method based on structural similarity | |
CN104778444A (en) | Method for analyzing apparent characteristic of vehicle image in road scene | |
CN106682641A (en) | Pedestrian identification method based on image with FHOG- LBPH feature | |
CN103136528A (en) | Double-edge detection based vehicle license plate identification method | |
CN104217217A (en) | Vehicle logo detection method and system based on two-layer classification | |
CN103455791A (en) | Method for recognizing vehicle types based on linear array CCD automobile chassis imaging | |
CN104268589B (en) | A kind of front vehicles detection method | |
CN104299009A (en) | Plate number character recognition method based on multi-feature fusion | |
CN106446792A (en) | Pedestrian detection feature extraction method in road traffic auxiliary driving environment | |
CN103942546A (en) | Guide traffic marking identification system and method in municipal environment | |
CN104240515A (en) | Exclusive bus lane occupation snapshotting method based on image processing |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20161005 Termination date: 20200815 |
|
CF01 | Termination of patent right due to non-payment of annual fee |