CN102646199A - Motorcycle type identifying method in complex scene - Google Patents
Motorcycle type identifying method in complex scene Download PDFInfo
- Publication number
- CN102646199A CN102646199A CN2012100497301A CN201210049730A CN102646199A CN 102646199 A CN102646199 A CN 102646199A CN 2012100497301 A CN2012100497301 A CN 2012100497301A CN 201210049730 A CN201210049730 A CN 201210049730A CN 102646199 A CN102646199 A CN 102646199A
- Authority
- CN
- China
- Prior art keywords
- parts
- vehicle
- video image
- parameter
- score
- 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)
Abstract
The invention discloses a motorcycle type identifying method in a complex scene, which comprises the following steps of: initializing a component dictionary of a video image, learning a parameter of each component in the component dictionary, calculating an optimal composition structure according to the learned parameters of the components and an XOR searching tree, training and integrating vehicle templates by adopting the optimal composition structure, and detecting and identifying a motorcycle type in the video image by using the vehicle templates. According to the motorcycle type identifying method, the optimal composition structure of the vehicle templates is learned by adopting a dynamic planning algorithm, the XOR searching tree and a large quantity of actual samples, thus the efficiency of training the templates is increased, better discrimination is achieved, and actual application is facilitated. According to the motorcycle type identifying method, by combining a Latent SVM (Support Vector Machine) algorithm and a robust HOG (Histograms of Oriented Gradients) characteristic, the motorcycle type in the complex scene can be processed, and instantaneity and generality are ensured.
Description
Technical field
The present invention relates to image model identification, intelligent video monitoring and intelligent transportation field, be specifically related to the model recognizing method in a kind of complex scene.
Background technology
Vehicle identification based on video image refers to from image and video, automatically identify dissimilar cars; Like minibus, car, truck, motor bus or the like; It is the gordian technique in the intelligent transportation system; No matter monitor the field in intelligent transportation, still in the full automatic charging field in highway and parking lot, it all has extremely important application.
Vehicle identification based on video image generally is divided into three parts: 1, vehicle image cuts apart; 2, feature extraction; 3, the identification of vehicle and classification.Relevant model recognizing method mainly comprises in the document at present: (a) based on the model recognizing method of prototype with (b) based on the model recognizing method of classifying.
For method based on prototype, often need at first set up the template database of standard, will mate through vehicle image after over-segmentation, the feature extraction and the template in the database then.It generally can be divided into: i) based on the coupling of vehicle edge; Ii) based on the coupling of vehicle ' s contour; Iii) based on the coupling of vehicle geometric parameter (like height, width, length and length breadth ratio etc.).These class methods are simple, intuitive the most, but its shortcoming is also quite obvious: one of which, and edge, profile or other geometric parameters of accurate extraction vehicle are relatively more difficult from real image; Its two, this method often requires video camera must be installed in fixing position and to its demarcation, has limited its application scenario; They are three years old; This method generally can only be separated size, length breadth ratio differs apparent in view vehicle; Like large car and compact car, and differ unconspicuous car (for example be all the lorry and the passenger vehicle of large car, or be all the car and the just very difficult differentiation of jeep of compact car) for size, length breadth ratio; Its four, the not enough robust of this method is easy to receive the influence of picture noise, weather condition.
For method, often need at first to extract various characteristics the sorter reasonable in design then vehicle of classifying to vehicle based on classification.The performance of these class methods often depends on the selection of characteristic and the design of sorter.It generally can be divided into: i) discern based on the vehicle of neural network; Ii) discern based on the vehicle of Gabor wave filter; Iii) discern based on the vehicle of SVMs (SVM).Wherein, I) with the parameter of the 3 d structure model of vehicle as characteristic, utilize neural network that the type of vehicle is classified then, ii) extracted the Gabor characteristic of vehicle; Utilize the method for template matches to realize vehicle classification then; Iii) extracted some characteristic (like absolute altitude, width and the length of vehicle, SIFT characteristic etc.) of vehicle, utilized SVMs (SVM) then vehicle classification.Though these class methods than the method based on prototype, have stronger robustness, they also exist common shortcoming: one of which, these class methods still depend on the quality of image segmentation very much, often can only handle the simple situation of background; Its two, the characteristic that this class methods are selected still is not enough robust, they are three years old; The model that these class methods adopted is all fairly simple; The coarse information that can only represent target generally also can only be divided into large, medium and small three types with vehicle, and can not carry out further exhaustive division; Its four, these class methods still highly depend on the placement location of video camera.
Recently; Objective classification method based on parts has become a kind of trend; Especially the partial model based on deformation (Deformable part template) that proposes of Felzenswalb has been obtained great success and (has been seen " Object Detection with Discriminatively Trained Part Based Models "; IEEE Transactions on Pattern Analysis and Machine Intelligence, 32 (9): 1627-1645,2010).This method is with two-layer star model of Latent-SVM algorithm training; This model has combined the geometry site between whole object and the target component; Have following advantage than recognition methods before: the more HOG characteristic of robust has been used in (1); Make model have more identification, overcome the sensitivity of existing method effectively complicated applications background and noise; (2) employing allows parts on certain direction, position and yardstick, to change based on the partial model of deformation, therefore can catch the detailed information of target.Traditional recognition methods during vehicle, is often done as a whole identification with car in identification, and this model is not only according to car load, and the parts of also waiting for bus according to wheel, vehicle window are discerned car, have increased the reliability of identification; (3) this method does not need in advance target to be cut apart, and has therefore avoided classic method because target is cut apart the inaccurate difficulty of bringing.Yet, because this method uses heuristic that parts are carried out initialization, the initialized location that therefore can not always find for parts, this tends to cause the inefficacy of model; Secondly, this heuristic also highly depends on the number and the shape of the artificial parts of setting.In reality, the components number and the shape of target are not often fixed, and it depends on the distance of video camera visual angle, distance objective and the difference between the target classification, and like this, for every type vehicle, one group of appropriate parts is selected in very difficult artificially.
Summary of the invention
The object of the present invention is to provide the model recognizing method in a kind of complex scene, it can locate and discern the type of vehicle efficiently, and improves the speed of vehicle identification greatly.
The present invention realizes through following technical scheme:
Model recognizing method under a kind of complex scene may further comprise the steps:
(1) the parts dictionary of initialization video image comprises following substep:
(1-1) confirm the length breadth ratio and the area of detection window in the video image according to positive and negative size in the video image;
The shape of (1-2) confirming parts in the parts dictionary according to the length breadth ratio and the area of detection window, area and point of fixity;
(1-3) according to the shape of parts, area and point of fixity make up with or search tree;
(2) parameter of each parts in the study parts dictionary;
(3) according to the parameter of each parts of study and with or the search tree compute optimal form structure, comprise following substep:
(3-1) according to each parts of calculation of parameter of each parts of study score at positive negative sample,
And initialization and or the leaf node of search tree;
(3-2) according to the score of positive negative sample calculate with or the top score of search tree;
(3-3) on and/or tree, confirm selected node, to obtain the optimum structure of forming according to top score;
(4) adopt optimum structured training and the integration vehicle template formed;
(5) vehicle in use vehicle template detection and the identification video image.
In the step (2), be to use the parameter of each parts in the Latent-SVM algorithm study parts dictionary.
In the step (3-2) be through dynamic programming algorithm calculate from bottom to top with or the top score of search tree.
Be to form structure in the step (3-3) through retrogressive method compute optimal from bottom to top.
Step (4) specifically comprises: different angles, dissimilar vehicles are carried out the training and the integration of template, and the threshold value of different templates is unitized.
Step (5) specifically comprises: when the detection and Identification vehicle, adopt the method for moving window, and video image is extracted HOG characteristic pyramid.
With respect to prior art, the present invention has following advantage and beneficial effect:
(1) the vehicle template among the present invention has been utilized the optimum structure of learning from a large amount of training samples of forming of parts, has improved the identification of template and the accuracy rate of identification effectively;
(2) vehicle detection among the present invention has adopted the moving window method, has overcome existing background subtraction method, frame-to-frame differences method and the optical flow method sensitivity to noise effectively, has effectively overcome the influence of picture noise, has expanded the range of application of this method widely;
(3) the present invention has combined the HOG characteristic of Latent SVM algorithm and robust; And the component model that trains corresponding composition structure according to the type and the visual angle of car; Do not need video camera to fix, the vehicle identification under can the dealing with complicated scene has guaranteed real-time and versatility;
(4) method of the present invention is not limited to rough vehicle detection and classification, and like compact car and large car, it can carry out more careful classification, such as car and jeep, taxi and minibus or the like.
Description of drawings
Fig. 1 is the process flow diagram of the model recognizing method in the complex scene of the present invention.
Fig. 2 is training sample and corresponding template thereof.
Fig. 3 illustrates and or search tree.
Fig. 4 (a) illustrates the vehicle recognition result of the inventive method for car.
Fig. 4 (b) illustrates the vehicle recognition result of the inventive method for truck.
Embodiment
Below at first technical term of the present invention is made an explanation and explain.
Parts:, possibly be the zone or the like together on wheel, vehicle window, car door or the vehicle body corresponding to the part of vehicle;
Component dictionary a: set of forming by all parts;
Positive sample: form: the image and the position (with upper left corner coordinate and the lower right corner coordinate mark of rectangle frame) of vehicle in image that comprise vehicle by two parts.
Negative sample: the image that does not comprise vehicle.
Learning sample: promptly positive negative sample.
With or search tree: a notion in artificial intelligence and the computer vision, be by with or figure promote, with or to scheme be a kind ofly systematically PROBLEM DECOMPOSITION to be minor issue independently mutually, divide then and the method that solves.With or figure in two kinds of representational nodes are arranged: " with node " and " or node "." with node " refers to that it was just separated when all subsequent node were all separated; " or node " refers to that each subsequent node is all fully independent, as long as wherein there is one to separate it and just separate.For and/or tree, except start node, all the other each nodes all have only a father node.
Recall: mainly be meant after having tried to achieve optimum solution, begin to the leaf node search from the root node of and/or tree, with confirm in asking the optimum solution process the node or the state of process.
Detection window: refer to a rectangle frame on the image; When detecting target, with rectangle frame scan image on a plurality of yardsticks, each step of scanning only is concerned about the image information in the rectangle frame; See whether comprise target in this rectangle frame, what this rectangle frame was vivid says to be exactly a window.
As shown in Figure 1, the model recognizing method in the complex scene of the present invention is following:
1, the parts dictionary of initialization video image specifically comprises following substep
(1-1) confirm the length breadth ratio and the area of detection window in the video image according to positive and negative size in the video image; To one group of training sample D={x
1, x
2..., x
n... }, use the length breadth ratio of the peak value of sample length breadth ratio and Gaussian function convolution as detection block, use sample area 20 percent to divide for an area of making the position detection block;
The shape of (1-2) confirming parts in the parts dictionary according to the length breadth ratio and the area of detection window, area and point of fixity;
Particularly, according to the size of detection window, enumerate area, length breadth ratio and the point of fixity of all candidate's parts, wherein, what the area of parts can not be greater than the detection block area is half the.The edge of parts can not surpass the edge of detection block.
(1-3) according to the shape of parts, area and point of fixity make up with or search tree;
As shown in Figure 2, wherein the parts in the father node are split as two sub-components with the node representative, or the different fractionation mode of node representative.With or search tree enumerate the component relationship of all parts, a kind of composition structure of each stalk tree corresponding component.In addition, the candidate's parts in the dictionary satisfy the size that size is no less than 3 * 3 HOG piece and is not more than thick yardstick template.The template of training sample and correspondence thereof is as shown in Figure 1.
2, use the parameter of each parts in the Latent-SVM algorithm study parts dictionary;
Choose candidate's parts all in the dictionary, its parameter can be obtained by the study of Latent-SVM algorithm:
Here, w is the long vector that the parameter of all M parts is formed, w
jIt is the parameter of j parts.σ
j(x
i, h
j) be the HOG characteristic that j parts extract.h
jBe hidden variable, specifically represent the position of the characteristic of each parts extraction, and the anglec of rotation.
3, according to the parameter of each parts of study and with or the search tree compute optimal form structure;
Particularly, from dictionary, select one group not overlapping and cover the parts of detection window fully.Whether each parts is selected is to confirm according to its must assign on all positive negative samples, and this score is calculated through following formula:
According to the score of each candidate's parts, through dynamic programming algorithm with or search tree on the composition structure of compute optimal, as shown in Figure 3, specifically comprise following substep:
(3-1) according to each parts of calculation of parameter of each parts of study score at positive negative sample, and initialization and or the leaf node of search tree.The score of each the candidate's parts that calculates according to formula (3), and compose give with or search tree in corresponding each leaf node.Other leaf node is composed and is divided into 0;
(3-2) according to the score of positive negative sample, use dynamic programming algorithm is from bottom to top calculated top score.According to the score of each leaf node, can calculate the top score of each node.The score of each and node be two leaf nodes branch with, each or node score are the maximal values of all child node scores;
(3-3) recall whole and/or tree from root node to leaf node, confirm selected node, thereby obtain the optimum structure of forming according to top score.According to the method for recalling, can obtain optimal path, the parts that comprise in this optimal path are the optimum parts of forming in the structure.Selected parts have been formed optimum composition structure;
4, adopt optimum structured training and the integration vehicle template formed;
Particularly; Utilize the optimum structure of forming of the parts of learning in the step (3); We carry out the training of template to different angles, dissimilar vehicles; For example in order to discern these two kinds of vehicles of car and truck, we possibly further divide three kinds of visual angles to car and truck: headstock, the tailstock and car are leaned to one side.We just need 6 vehicle templates of training like this, and final vehicle template has just comprised this 6 templates.For the detection threshold of unitized each template, we also need adjust the bias term between each template in addition, and the threshold value of final template is all learnt on training sample through the Latent-SVM algorithm with the bias term of each template.
Each position component and size are not demarcated in advance on training sample, belong to hidden variable, so the training need of template employing coordinate descent algorithm, and the coordinate descent algorithm was divided into for two steps: the 1) parameter of fixed form, locate each position component; 2) position of fixed part, the parameter of study template.Algorithm is iteration between these two processes always, up to satisfying end condition.Simultaneously, for the convergence of accelerating algorithm, we have adopted the technology of data mining difficulty negative sample, when iteration training each time, dynamically add the difficult negative sample that classification makes mistakes, and dynamically remove the simple negative sample away from classifying face.
5, the vehicle in use vehicle template detection and the identification video image.
The moving window method is adopted in vehicle detection and identification; Idiographic flow is shown in Fig. 4 (a) and Fig. 4 (b); For the two field picture in the video flowing, at first we extract HOG characteristic pyramid on a plurality of yardsticks, the template of utilizing the training of the 4th step to obtain then; On the characteristic pyramid, detect successively and recognition image in the vehicle that comprised; This process is exactly the response of calculating vehicle template and HOG proper vector, if response is higher than the threshold value of detection, algorithm is just predicted and detected a car here so.Wherein, for each candidate's vehicle, its pairing vehicle is exactly the pairing vehicle of template with peak response.For example, we are by a car of being made up of 6 vehicle templates (Car) and truck (Truck) integrated template.For a detected car in the image, algorithm contrasts the response of various types of vehicle templates, if the response of the tailstock template of truck is maximum, so detected this car is exactly a truck.For each car in the image, the position and the corresponding vehicle classification at its place of algorithm output are shown in Fig. 5 a and Fig. 5 b.
In addition; When the detection and Identification vehicle; We adopt Cascade beta pruning algorithm, on training sample, learn out a series of parts pruning threshold, have been divided into a plurality of stages to original testing process like this; Can carry out parallel detection and identification to the vehicle template of a plurality of vehicles and angle, improve the travelling speed of algorithm greatly.
Claims (6)
1. the model recognizing method under the complex scene is characterized in that, may further comprise the steps:
(1) the parts dictionary of initialization video image comprises following substep:
(1-1) confirm the length breadth ratio and the area of detection window in the said video image according to positive and negative size in the said video image;
The shape of (1-2) confirming parts in the said parts dictionary according to the length breadth ratio and the area of said detection window, area and point of fixity;
(1-3) according to the shape of said parts, area and point of fixity make up with or search tree;
(2) parameter of each parts in the said parts dictionary of study;
(3) according to the parameter of each parts of said study and said and or the search tree compute optimal form structure, comprise following substep:
(3-1) according to each parts of calculation of parameter of each parts of said study score at said positive negative sample, and initialization said with or the leaf node of search tree;
(3-2) according to the score of said positive negative sample calculate said with or the top score of search tree;
(3-3) on said and/or tree, confirm selected node, to obtain the said optimum structure of forming according to said top score;
(4) adopt said optimum structured training and the integration vehicle template formed;
(5) use said vehicle template detection and the vehicle of discerning in the said video image.
2. model recognizing method according to claim 1 is characterized in that, in the said step (2), is to use the Latent-SVM algorithm to learn the parameter of each parts in the said parts dictionary.
3. the model recognizing method under a kind of complex scene according to claim 1 is characterized in that, in the said step (3-2) be through dynamic programming algorithm calculate from bottom to top said with or the top score of search tree.
4. model recognizing method according to claim 1 is characterized in that, is through the said optimum structure of forming of retrogressive method calculating from bottom to top in the said step (3-3).
5. model recognizing method according to claim 1 is characterized in that, said step (4) specifically comprises: different angles, dissimilar vehicles are carried out the training and the integration of template, and the threshold value of different templates is unitized.
6. model recognizing method according to claim 1 is characterized in that, said step (5) specifically comprises: when the detection and Identification vehicle, adopt the method for moving window, and said video image is extracted HOG characteristic pyramid.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201210049730 CN102646199B (en) | 2012-02-29 | 2012-02-29 | Motorcycle type identifying method in complex scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201210049730 CN102646199B (en) | 2012-02-29 | 2012-02-29 | Motorcycle type identifying method in complex scene |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102646199A true CN102646199A (en) | 2012-08-22 |
| CN102646199B CN102646199B (en) | 2013-12-18 |
Family
ID=46659015
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201210049730 Expired - Fee Related CN102646199B (en) | 2012-02-29 | 2012-02-29 | Motorcycle type identifying method in complex scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102646199B (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102880881A (en) * | 2012-09-25 | 2013-01-16 | 常州大学 | Method for identifying car type on basis of binary support vector machines and genetic algorithm |
| CN102902983A (en) * | 2012-09-21 | 2013-01-30 | 博康智能网络科技股份有限公司 | Taxi recognition method based on support vector machine |
| CN102902960A (en) * | 2012-09-25 | 2013-01-30 | 武汉烽火众智数字技术有限责任公司 | Leave-behind object detection method based on Gaussian modelling and target contour |
| CN102930271A (en) * | 2012-09-21 | 2013-02-13 | 博康智能网络科技股份有限公司 | Method for identifying taxicabs in real time by utilizing video images |
| CN103093249A (en) * | 2013-01-28 | 2013-05-08 | 中国科学院自动化研究所 | Taxi identifying method and system based on high-definition video |
| CN103258213A (en) * | 2013-04-22 | 2013-08-21 | 中国石油大学(华东) | Vehicle model dynamic identification method used in intelligent transportation system |
| CN103886583A (en) * | 2014-02-14 | 2014-06-25 | 杭州电子科技大学 | Target detection sliding window scanning method based on scene geometric constraint |
| CN103942558A (en) * | 2013-01-22 | 2014-07-23 | 日电（中国）有限公司 | Method and apparatus for obtaining object detectors |
| CN104091151A (en) * | 2014-06-30 | 2014-10-08 | 南京信息工程大学 | Vehicle identification method based on Gabor feature extraction and sparse representation |
| WO2015027649A1 (en) * | 2013-08-27 | 2015-03-05 | 东莞中国科学院云计算产业技术创新与育成中心 | Vehicle detection method using multi-scale model |
| CN104751190A (en) * | 2015-04-23 | 2015-07-01 | 武汉大学 | Vehicle part positioning method for vehicle fine recognition |
| CN105975949A (en) * | 2016-05-26 | 2016-09-28 | 大连理工大学 | Visual-information-based automobile identification method |
| CN106056102A (en) * | 2016-07-01 | 2016-10-26 | 哈尔滨工业大学 | Video-image-analysis-based road vehicle type classification method |
| CN107122706A (en) * | 2017-03-17 | 2017-09-01 | 华南理工大学 | A kind of vehicle checking method based on growth form extreme learning machine |
| CN107194338A (en) * | 2017-05-14 | 2017-09-22 | 北京工业大学 | Traffic environment pedestrian detection method based on human body tree graph model |
| CN108154127A (en) * | 2017-12-27 | 2018-06-12 | 天津智芯视界科技有限公司 | A kind of vehicle identification method based on video and radar |
| CN108475471A (en) * | 2016-01-26 | 2018-08-31 | 三菱电机株式会社 | Vehicle decision maker, vehicle determination method and vehicle decision procedure |
| CN108537125A (en) * | 2018-03-13 | 2018-09-14 | 周慧 | Automobile search system based on amendment type image procossing |
| CN108647624A (en) * | 2018-05-04 | 2018-10-12 | 广东工业大学 | A kind of road vehicle detection method, device, equipment and storage medium |
| CN109544907A (en) * | 2018-10-23 | 2019-03-29 | 大唐软件技术股份有限公司 | A kind of vehicle count method, device |
| CN110245670A (en) * | 2019-06-11 | 2019-09-17 | 观博云标(北京)文化科技有限公司 | Day pearl identity identification method and device based on image pyramid histogram of gradients feature |
| CN110765929A (en) * | 2019-10-21 | 2020-02-07 | 东软睿驰汽车技术（沈阳）有限公司 | Vehicle obstacle detection method and device |
| CN112255973A (en) * | 2019-07-02 | 2021-01-22 | 库卡机器人(广东)有限公司 | Target detection method, detection terminal and storage medium in industrial production system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1996346A (en) * | 2005-12-29 | 2007-07-11 | 电子科技大学 | Vehicle recognition algorithm based on contour |
| CN101661559A (en) * | 2009-09-16 | 2010-03-03 | 中国科学院计算技术研究所 | Digital image training and detecting methods |
| CN101872477A (en) * | 2009-04-24 | 2010-10-27 | 索尼株式会社 | Method and device for detecting object in image and system containing device |
-
2012
- 2012-02-29 CN CN 201210049730 patent/CN102646199B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1996346A (en) * | 2005-12-29 | 2007-07-11 | 电子科技大学 | Vehicle recognition algorithm based on contour |
| CN101872477A (en) * | 2009-04-24 | 2010-10-27 | 索尼株式会社 | Method and device for detecting object in image and system containing device |
| CN101661559A (en) * | 2009-09-16 | 2010-03-03 | 中国科学院计算技术研究所 | Digital image training and detecting methods |
Non-Patent Citations (1)
| Title |
|---|
| 方菲菲，余稳: "基于PCA-LDA-SVM 的多普勒雷达车型识别算法", 《数据采集与处理》 * |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902983A (en) * | 2012-09-21 | 2013-01-30 | 博康智能网络科技股份有限公司 | Taxi recognition method based on support vector machine |
| CN102930271A (en) * | 2012-09-21 | 2013-02-13 | 博康智能网络科技股份有限公司 | Method for identifying taxicabs in real time by utilizing video images |
| CN102902983B (en) * | 2012-09-21 | 2016-01-20 | 博康智能网络科技股份有限公司 | A kind of taxi identification method based on support vector machine |
| CN102902960B (en) * | 2012-09-25 | 2015-04-22 | 武汉烽火众智数字技术有限责任公司 | Leave-behind object detection method based on Gaussian modelling and target contour |
| CN102902960A (en) * | 2012-09-25 | 2013-01-30 | 武汉烽火众智数字技术有限责任公司 | Leave-behind object detection method based on Gaussian modelling and target contour |
| CN102880881A (en) * | 2012-09-25 | 2013-01-16 | 常州大学 | Method for identifying car type on basis of binary support vector machines and genetic algorithm |
| CN103942558A (en) * | 2013-01-22 | 2014-07-23 | 日电（中国）有限公司 | Method and apparatus for obtaining object detectors |
| CN103093249A (en) * | 2013-01-28 | 2013-05-08 | 中国科学院自动化研究所 | Taxi identifying method and system based on high-definition video |
| CN103093249B (en) * | 2013-01-28 | 2016-03-02 | 中国科学院自动化研究所 | A kind of taxi identification method based on HD video and system |
| CN103258213B (en) * | 2013-04-22 | 2016-04-27 | 中国石油大学(华东) | A kind of for the dynamic vehicle model recognizing method in intelligent transportation system |
| CN103258213A (en) * | 2013-04-22 | 2013-08-21 | 中国石油大学(华东) | Vehicle model dynamic identification method used in intelligent transportation system |
| WO2015027649A1 (en) * | 2013-08-27 | 2015-03-05 | 东莞中国科学院云计算产业技术创新与育成中心 | Vehicle detection method using multi-scale model |
| CN103886583A (en) * | 2014-02-14 | 2014-06-25 | 杭州电子科技大学 | Target detection sliding window scanning method based on scene geometric constraint |
| CN104091151A (en) * | 2014-06-30 | 2014-10-08 | 南京信息工程大学 | Vehicle identification method based on Gabor feature extraction and sparse representation |
| CN104751190A (en) * | 2015-04-23 | 2015-07-01 | 武汉大学 | Vehicle part positioning method for vehicle fine recognition |
| CN104751190B (en) * | 2015-04-23 | 2017-12-29 | 武汉大学 | A kind of vehicle part localization method finely identified towards vehicle |
| CN108475471B (en) * | 2016-01-26 | 2020-12-18 | 三菱电机株式会社 | Vehicle determination device, vehicle determination method, and computer-readable recording medium |
| CN108475471A (en) * | 2016-01-26 | 2018-08-31 | 三菱电机株式会社 | Vehicle decision maker, vehicle determination method and vehicle decision procedure |
| CN105975949A (en) * | 2016-05-26 | 2016-09-28 | 大连理工大学 | Visual-information-based automobile identification method |
| CN106056102A (en) * | 2016-07-01 | 2016-10-26 | 哈尔滨工业大学 | Video-image-analysis-based road vehicle type classification method |
| CN107122706A (en) * | 2017-03-17 | 2017-09-01 | 华南理工大学 | A kind of vehicle checking method based on growth form extreme learning machine |
| CN107194338A (en) * | 2017-05-14 | 2017-09-22 | 北京工业大学 | Traffic environment pedestrian detection method based on human body tree graph model |
| CN108154127A (en) * | 2017-12-27 | 2018-06-12 | 天津智芯视界科技有限公司 | A kind of vehicle identification method based on video and radar |
| CN108537125A (en) * | 2018-03-13 | 2018-09-14 | 周慧 | Automobile search system based on amendment type image procossing |
| CN108647624A (en) * | 2018-05-04 | 2018-10-12 | 广东工业大学 | A kind of road vehicle detection method, device, equipment and storage medium |
| CN109544907A (en) * | 2018-10-23 | 2019-03-29 | 大唐软件技术股份有限公司 | A kind of vehicle count method, device |
| CN110245670A (en) * | 2019-06-11 | 2019-09-17 | 观博云标(北京)文化科技有限公司 | Day pearl identity identification method and device based on image pyramid histogram of gradients feature |
| CN110245670B (en) * | 2019-06-11 | 2021-08-20 | 观博云标(北京)文化科技有限公司 | Image pyramid gradient histogram feature-based skyhead identity identification method and device |
| CN112255973A (en) * | 2019-07-02 | 2021-01-22 | 库卡机器人(广东)有限公司 | Target detection method, detection terminal and storage medium in industrial production system |
| CN110765929A (en) * | 2019-10-21 | 2020-02-07 | 东软睿驰汽车技术（沈阳）有限公司 | Vehicle obstacle detection method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102646199B (en) | 2013-12-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102646199B (en) | Motorcycle type identifying method in complex scene | |
| Beltrán et al. | Birdnet: a 3d object detection framework from lidar information | |
| CN110942000B (en) | Unmanned vehicle target detection method based on deep learning | |
| CN108345822B (en) | Point cloud data processing method and device | |
| CN102663356B (en) | Method for extraction and deviation warning of lane line | |
| Philipsen et al. | Traffic light detection: A learning algorithm and evaluations on challenging dataset | |
| CN111563412B (en) | Rapid lane line detection method based on parameter space voting and Bessel fitting | |
| CN105160309B (en) | Three lanes detection method based on morphological image segmentation and region growing | |
| CN103034836B (en) | Road sign detection method and road sign checkout equipment | |
| Sivaraman et al. | Vehicle detection by independent parts for urban driver assistance | |
| CN105892471A (en) | Automatic automobile driving method and device | |
| CN111580131B (en) | Method for identifying vehicles on expressway by three-dimensional laser radar intelligent vehicle | |
| DE102006012914A1 (en) | System and method for determining the distance to a preceding vehicle | |
| CN104778444A (en) | Method for analyzing apparent characteristic of vehicle image in road scene | |
| García-Garrido et al. | Robust traffic signs detection by means of vision and V2I communications | |
| CN101996401A (en) | Target analysis method and device based on intensity image and range image | |
| Zhang et al. | A framework for turning behavior classification at intersections using 3D LIDAR | |
| CN107274668A (en) | A kind of congestion in road modeling method based on vehicle detection | |
| Sahu et al. | A comparative analysis of deep learning approach for automatic number plate recognition | |
| CN102945602B (en) | Vehicle trajectory classifying method for detecting traffic incidents | |
| Börcs et al. | A marked point process model for vehicle detection in aerial LiDAR point clouds | |
| CN103295003A (en) | Vehicle detection method based on multi-feature fusion | |
| CN115685237A (en) | Multi-mode three-dimensional target detection method and system combining viewing cones and geometric constraints | |
| CN113306558B (en) | Lane changing decision method and system based on lane changing interaction intention | |
| Mancini et al. | A winner takes all mechanism for automatic object extraction from multi-source data |
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: 20131218 Termination date: 20160229 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |