CN105718859B - A kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot - Google Patents
A kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot Download PDFInfo
- Publication number
- CN105718859B CN105718859B CN201610026902.1A CN201610026902A CN105718859B CN 105718859 B CN105718859 B CN 105718859B CN 201610026902 A CN201610026902 A CN 201610026902A CN 105718859 B CN105718859 B CN 105718859B
- Authority
- CN
- China
- Prior art keywords
- image
- speed limitation
- red
- limitation board
- graphics processor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/582—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of traffic signs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot, its image for being 1200*900 to resolution ratio by the image down for being 1600*1200 by resolution ratio, it is obtained and red image similar in speed limitation board red in GPU equipment using CUDA C language, and the red image that will acquire passes back on CPU device, use connected component analysis method, obtain the information such as the position of each connected component, again in the original color image of 1600*1200, according to support vector machine classifier, judge whether each connected component is speed limitation board.The position of all speed limitation boards and Pixel Dimensions in the final original image for obtaining 1600*1200 pixel size.The present invention proposes executing the step of a part is easy to parallel computation in detection method in GPU equipment, by combining the calculating advantage of central processing unit and graphics processor, while being gone here and there and being calculated, to improve the verification and measurement ratio and detection speed of speed limitation board.The method of the invention has operation frame per second higher, and the development cycle is short, and method is it can be readily appreciated that advantage convenient for safeguarding.
Description
Technical field
The present invention relates to the traffic information detection fields of automatic Pilot industry, and in particular to a kind of knot in a kind of automatic Pilot
Close the method that graphics processor carries out speed limitation board detection.
Background technique
With the development of computer science and robot technology, automatic driving vehicle is in military, civilian and scientific research etc.
All various aspects are widely used, it has concentrated structure, electronics, cybernetics and artificial intelligence etc. multi-disciplinary newest
Research achievement has broad application prospects.
In unpiloted technical research, speed limitation board detection method is a kind of more time-consuming method, is needed pixel-by-pixel
Image is handled.Color extraction and contours extract are more time-consuming.And contour extraction method is a kind of side of recursive form
Method is not easy to be revised as the parallel method of operation, and therefore, how to carry out the quick detection of speed limitation board is those skilled in the art
Technical problem anxious to be resolved.
Summary of the invention
In view of this, it is necessary to provide in a kind of verification and measurement ratio that can be improved speed limitation board and the automatic Pilot of detection speed one
The method that kind combines graphics processor to carry out speed limitation board detection.
A kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot, a kind of knot in the automatic Pilot
Close graphics processor carry out speed limitation board detection method the following steps are included:
S1, the video image that vehicle camera obtains is received, and received video image is uploaded to graphics processor；
S2, in graphics processor, binaryzation analysis is carried out to the red area in video image and is extracted, and carries out mean value
Obscure the red image after being simplified；
S3, the red image after simplifying is back on central processing unit, the red image simplified repeatedly is connected to
Body analysis, obtains the rough position of speed limitation board；
S4, the rough position according to speed limitation board use support vector machine classifier in original color image, differentiate the position
Whether set is speed limitation board.
The present invention has that time-consuming is more, frame per second is low when running on CPU device for existing speed limitation board detection method,
It proposes executing the step of a part is easy to parallel computation in detection method in GPU equipment, by combining central processing unit and figure
The calculating advantage of shape processor, while being gone here and there and being calculated, to improve the verification and measurement ratio and detection speed of speed limitation board.The present invention
The method has operation frame per second higher, and the development cycle is short, and method is it can be readily appreciated that advantage convenient for safeguarding.
Detailed description of the invention
Fig. 1 is a kind of stream of the method for combination graphics processor progress speed limitation board detection in automatic Pilot of the present invention
Journey block diagram；
Fig. 2 is the flow diagram of step S1 in Fig. 1；
Fig. 3 is the flow diagram of step S2 in Fig. 1；
Fig. 4 is the flow diagram of step S3 in Fig. 1；
Fig. 5 is the flow diagram of step S4 in Fig. 1.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated, it should be understood that and the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
As shown in Figure 1, the embodiment of the present invention provides a kind of combination graphics processor progress speed limitation board detection in automatic Pilot
Method, in the automatic Pilot a kind of combination graphics processor carry out speed limitation board detection method the following steps are included:
S1, the video image that vehicle camera obtains is received, and received video image is uploaded to graphics processor；
S2, in graphics processor, binaryzation analysis is carried out to the red area in video image and is extracted, and carries out mean value
Obscure the red image after being simplified；
S3, the red image after simplifying is back on central processing unit, the red image simplified repeatedly is connected to
Body analysis, obtains the rough position of speed limitation board；
S4, the rough position according to speed limitation board use support vector machine classifier in original color image, differentiate the position
Whether set is speed limitation board.
Graphics processor is a kind of equipment for being suitable for Large-scale parallel computing, and the performance of equipment processing parallel computation is wanted
Exponentially it is better than the central processing unit of corresponding cost.Since graphics processor is not suitable for serial computing, and currently, still having
The method of many serial computings, these methods are not appropriate for being transplanted in graphics processor and be run.Therefore, the present invention uses
Central processing unit carries out serial computing, and carries out parallel computation using graphics processor, by combining central processing unit and figure
The calculating advantage of processor, while being gone here and there and being calculated, to improve the verification and measurement ratio and detection speed of speed limitation board.
As shown in Fig. 2, the step S1 include it is following step by step:
S11, the color RGB image that 1600 pixels are wide, 1200 pixels are high is obtained from camera, stored to central processing unit
In；
S12, the color RGB image in central processing unit is uploaded in graphics processor；
S13, in graphics processor, it is 1200*900 that the color RGB image that size is 1600*1200, which is compressed to size,
Color RGB image.
In order to improve the verification and measurement ratio of speed limitation board detection, the image of detection should select resolution ratio as high as possible；But by
The limitation of power is calculated in equipment, in order to improve the speed of detection, the image of detection should select alap resolution ratio.Cause
This, needs to be weighed between the two, chooses a preferable resolution ratio, so that the verification and measurement ratio of speed limitation board and detection speed
Two aspects can better meet demand.The present invention is detected by many experiments, and obtaining 1200*900 is optimal resolution, because
This, the color RGB image that size is 1600*1200, which is compressed to the color RGB image that size is 1200*900, can be improved limit
The verification and measurement ratio and detection speed of fast board.
Specifically, graphics processor of the present invention is GPU equipment, the central processing unit is CPU device.First
On CPU device, using the method for uploading of the OpenCV cuda::GpuMat class provided, by the colored RGB of 1600*1200 size
Image uploads in GPU equipment；For GPU equipment, the cuda::resize function provided using OpenCV, acquisition is narrowed down to
The image of 1200*900 size；
As shown in figure 3, the step S2 include it is following step by step:
S21, in graphics processor, color RGB image is converted into HSV image, according to H component and S component, is obtained red
Color binary map R and red degree figure RL；
Wherein, the H component obtains respectively according to following five kinds of situations:
As max=min,
H (x, y)=0；
As max=R (x, y), and when G (x, y) >=B (x, y),
As max=R (x, y), and when G (x, y) < B (x, y),
As max=G (x, y),
As max=B (x, y),
Shown in the following formula of acquisition modes of the S component:
In these formula that the H component obtains and S component obtains, max indicates R (x, y), G (x, y), B (x, y) most
Big value, min indicate the minimum value of R (x, y), G (x, y), B (x, y).
Shown in the following formula of acquisition modes of the red degree figure RL and red binary map R；
S22, red degree figure and red binary map are obscured using the block size progress mean value of 5*5, after being obscured
Red degree figure RLblurWith it is fuzzy after red binary map Rblur；
S23, according to red degree figure RLblurWith red binary map RblurObtain the red image R after simplifyingrefine, and make
With OpenCV provide cuda::GpuMat class method for down loading, by red image RrefineIt is back in central processing unit.
The red image R simplifiedrefineAcquisition modes it is as follows；
It is programmed since graphics processor programming exists relative to common C language, study threshold with higher and programming
Threshold, and the present invention greatly reduces these technical thresholds using CUDA C language, so that being not required to it is to be understood that figure is compiled
The relevant rudimentary knowledge of journey, it will be able to the programming of general-purpose computations is carried out directly in graphics processor.
Specifically, being directed to GPU equipment, it is programmed using CUDA C and realizes that RGB image, which is converted to HSV image, (to be disregarded
Calculate V component), according to H component and S component, obtain red binary map RB and red degree figure RL.
In this step, the value that each of GpuMat image pixel is directly accessed using CUDA C language, is reached
The development cycle short double dominant with the larger promotion of performance.Specifically, make the parameter type of GPU function be claimed as PtrStepSz <
T > para, T indicate image in each element type, para is parameter name, when function call, the parameter with
The form of GpuMat type inputs, and calls directly that para (x, y) is available or setting is located at xth row inside function, y column
The value of element.
As shown in figure 4, the step S3 include it is following step by step:
S31, to the red image R simplifiedrefineConnected component analysis is carried out, the information of each connected component is obtained；
S32, each connected component is filled with to solid convex polygon, obtains filled red image R 'refine, from
And get the image R of the rough position of the speed limitation board not comprising red sidehole；
Specifically, described image RholeAcquisition formula it is as follows；
Rhole(x, y)=R'refine(x,y)-Rrefine(x,y)；
S33, to image RholeCorroded, the image R ' after being corrodedhole.In image R 'holeUpper progress connected component point
Analysis, obtains all connected component information；Wherein, to image RholeThe purpose corroded is to remove some zero in image
Star noise spot reduces interference information；
S34, by image RholeThe connected component information of acquisition is converted to corresponding connected component information in original color image, i.e.,
Corresponding connected component information on 1600*1200 image, and the limitation of the image factor according to speed limitation board are converted to, speed limitation board is obtained
Rough position.
Wherein, in the original color image corresponding connected component information refer to it is corresponding each in original color image
The elemental area of a connected component, the pixel of the boundary rectangle of each connected component are wide and pixel is high.Specifically, being by each
The elemental area of connected component is multiplied by 1.3333332, the pixel of each connected component is wide and pixel Gao Jun is multiplied by 1.333333, most
The connected component information on 1600*1200 image has been obtained afterwards.
The image factor of the speed limitation board includes that the minimum pixel area of speed limitation board, maximum pixel are wide, and minimum pixel is wide, most
Big pixel is high, minimum pixel is high, minimum depth-width ratio, maximum aspect ratio, elemental area and corresponding area of a circle minimum ratio, elemental area
With corresponding area of a circle high specific.
As shown in figure 5, the step S4 include it is following step by step:
S41, the rough position shape of confining be rectangle frame, it is high according to the starting point of each rectangle frame and width, and
Color RGB image calculates the HOG feature of the rectangle；
S42, according to HOG feature Training Support Vector Machines classifier, and load support vector machine classifier after training；
S43, judged using HOG feature of the support vector machine classifier to rectangle frame, when the function of classifier returns
When value is 1, indicate that the rectangle frame is speed limitation board, it is on the contrary then be not speed limitation board, speed limitation board number and position are obtained to finally detect
It sets.
Preferably to embody obtained testing result by entity number, further, the present invention will be examined finally
The speed limitation board number and position measured is exported according to the form of output parameter.
Specifically, the present invention divides the rough position of speed limitation board using rectangle frame, pass through will scheme in step S3
As RholeThe connected component information of acquisition is converted to corresponding connected component information in original color image, further available thick
Slightly the center of mass point coordinate of the rectangle frame of position and position and the upper left of rectangle frame point x coordinate, upper left point y-coordinate, lower-right most point
X coordinate, lower-right most point y-coordinate determine whether the rough position is speed limitation board by step S4, if it is, above-mentioned data are
The center of mass point coordinate of speed limitation board and position and the upper left of speed limitation board point x coordinate, upper left point y-coordinate, lower-right most point x coordinate, the right side
Lower y-coordinate.
The center of mass point coordinate for the speed limitation board that will test out and position are provided according to the form of output parameter.One of them
Output parameter using x coordinate, y-coordinate as constituent parameters structural body array, indicate the matter of all speed limitation boards that detected
Heart point coordinate；Another output parameter is with upper left point x coordinate, upper left point y-coordinate, lower-right most point x coordinate, lower-right most point y-coordinate
The array of structural body as constituent parameters indicates the position of all speed limitation boards that detected, and the function finally obtained returns
The integer that value is one 32, the integer indicate the number of the speed limitation board detected.
A kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot of the present invention, according to from vehicle
Camera obtain video image image is uploaded in GPU equipment on the image in different resolution of 1600*1200,
In GPU equipment, the function provided using the library OpenCV, by image down to 1200*900；Reuse CUDA C language, obtain with
Red image similar in speed limitation board red.The red image that will acquire again passes back on CPU device, is analyzed using connected component
Method obtains the information such as the position of each connected component, then in the original color image of 1600*1200, according to support vector machines
Classifier judges whether each connected component is speed limitation board.Own in the final original image for obtaining 1600*1200 pixel size
The position of speed limitation board and Pixel Dimensions.
The present invention has that time-consuming is more, frame per second is low when running on CPU device for existing speed limitation board detection method,
It proposes executing the step of a part is easy to parallel computation in detection method in GPU equipment, by combining central processing unit and figure
The calculating advantage of shape processor, while being gone here and there and being calculated, to improve the verification and measurement ratio and detection speed of speed limitation board.The present invention
The method has operation frame per second higher, and the development cycle is short, and method is it can be readily appreciated that advantage convenient for safeguarding.
The present invention utilizes a kind of speed limitation board detection method in graphics processor top partite transport row, so that in embedded device
On, it can be with the speed limitation board in the velocity measuring image of 10 frame per second.The purpose of this method is obtained during automatic Pilot
More information relevant to road, auxiliary driver more safely drive a car, and for automated driving system obtain highest when
Speed limit system, which provides, refers to distinguishing rule.
Apparatus above embodiment and embodiment of the method are one-to-one, the simple places of Installation practice, referring to method reality
Apply example.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with other
The difference of embodiment, the same or similar parts in each embodiment may refer to each other.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And method and step, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to functionality in the above description.This
A little functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Specially
Industry technical staff can use different methods to achieve the described function each specific application, but this realization is not
It should be more than the scope of the present invention.
The step of method or method described in conjunction with the examples disclosed in this document, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory, memory, read-only memory,
Institute is public in electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technical field
In the storage medium for any other forms known.
The embodiment of the present invention is described with above attached drawing, but the invention is not limited to above-mentioned specific
Embodiment, the above mentioned embodiment is only schematical, rather than restrictive, those skilled in the art
Under the inspiration of the present invention, without breaking away from the scope protected by the purposes and claims of the present invention, it can also make very much
Form, all of these belong to the protection of the present invention.
Claims (6)
1. a kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot, which is characterized in that described to drive automatically
Sail a kind of middle combination graphics processor carry out speed limitation board detection method the following steps are included:
S1, the video image that vehicle camera obtains is received, and received video image is uploaded to graphics processor；
S11, the color RGB image that 1600 pixels are wide, 1200 pixels are high is obtained from camera, stored into central processing unit；
S12, the color RGB image in central processing unit is uploaded in graphics processor；
S13, in graphics processor, the color RGB image that size is 1600*1200 is compressed to the coloured silk that size is 1200*900
Color RGB image；
S2, in graphics processor, binaryzation analysis is carried out to the red area in video image and is extracted, and it is fuzzy to carry out mean value
Red image after being simplified；
S21, in graphics processor, color RGB image is converted into HSV image, according to H component and S component, obtains red two
Value figure R and red degree figure RL；
S22, the red that mean value is fuzzy, after being obscured is carried out using the block size of 5*5 to red degree figure and red binary map
Degree figure RLblurWith it is fuzzy after red binary map Rblur；
S23, according to red degree figure RLblurWith red binary map RblurObtain the red image R after simplifyingrefine, and will be red
Image RrefineIt is back in central processing unit；
S3, the red image after simplifying is back on central processing unit, multiple connected component point is carried out to the red image simplified
Analysis, obtains the rough position of speed limitation board；
S4, the rough position according to speed limitation board use support vector machine classifier in original color image, differentiate that the position is
No is speed limitation board.
2. a kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot according to claim 1,
Be characterized in that, the step S3 include it is following step by step:
S31, to the red image R simplifiedrefineConnected component analysis is carried out, the information of each connected component is obtained；
S32, each connected component is filled with to solid convex polygon, obtains filled red image R 'refine, to obtain
Get the image R of the rough position of the speed limitation board not comprising red sidehole；
S33, to image RholeCorroded to obtain image R 'hole, in image R 'holeUpper progress connected component analysis, obtains all
Connected component information；
S34, connected component information is converted to corresponding connected component information in original color image, and according to the image of speed limitation board because
The limitation of element, obtains the rough position of speed limitation board.
3. a kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot according to claim 2,
It is characterized in that, described image RholeAcquisition formula it is as follows；
Rhole(x, y)=R'refine(x,y)-Rrefine(x,y)。
4. a kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot according to claim 2,
Be characterized in that, the image factor of the speed limitation board include the minimum pixel area of speed limitation board, maximum pixel is wide, minimum pixel is wide,
Maximum pixel is high, minimum pixel is high, minimum depth-width ratio, maximum aspect ratio, elemental area and corresponding area of a circle minimum ratio, pixel faces
It accumulates and corresponding area of a circle high specific.
5. a kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot according to claim 1,
Be characterized in that, the step S4 include it is following step by step:
S41, the rough position shape of confining be rectangle frame, it is high according to the starting point of each rectangle frame and width and colored
RGB image calculates the HOG feature of the rectangle；
S42, according to HOG feature Training Support Vector Machines classifier, and load support vector machine classifier after training；
S43, judged using HOG feature of the support vector machine classifier to rectangle frame, when the function return value of classifier is 1
When, indicate that the rectangle frame is speed limitation board, it is on the contrary then be not speed limitation board.
6. a kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot according to claim 1,
It is characterized in that, the speed limitation board number and position that will test out are exported according to the form of output parameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610026902.1A CN105718859B (en) | 2016-01-15 | 2016-01-15 | A kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610026902.1A CN105718859B (en) | 2016-01-15 | 2016-01-15 | A kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105718859A CN105718859A (en) | 2016-06-29 |
| CN105718859B true CN105718859B (en) | 2019-05-03 |
Family
ID=56147847
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610026902.1A Active CN105718859B (en) | 2016-01-15 | 2016-01-15 | A kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105718859B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107808360B (en) * | 2017-09-05 | 2020-08-04 | 百度在线网络技术（北京）有限公司 | Data processing method and device of end-to-end automatic driving system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102542260A (en) * | 2011-12-30 | 2012-07-04 | 中南大学 | Method for recognizing road traffic sign for unmanned vehicle |
| CN103034836A (en) * | 2011-09-29 | 2013-04-10 | 株式会社理光 | Road sign detection method and device |
| CN103971126A (en) * | 2014-05-12 | 2014-08-06 | 百度在线网络技术（北京）有限公司 | Method and device for identifying traffic signs |
| CN104616021A (en) * | 2014-12-24 | 2015-05-13 | 深圳市腾讯计算机系统有限公司 | Method and device for processing traffic sign symbols |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011154978A2 (en) * | 2010-06-10 | 2011-12-15 | Tata Consultancy Services Limited | An illumination invariant and robust apparatus and method for detecting and recognizing various traffic signs |
-
2016
- 2016-01-15 CN CN201610026902.1A patent/CN105718859B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103034836A (en) * | 2011-09-29 | 2013-04-10 | 株式会社理光 | Road sign detection method and device |
| CN102542260A (en) * | 2011-12-30 | 2012-07-04 | 中南大学 | Method for recognizing road traffic sign for unmanned vehicle |
| CN103971126A (en) * | 2014-05-12 | 2014-08-06 | 百度在线网络技术（北京）有限公司 | Method and device for identifying traffic signs |
| CN104616021A (en) * | 2014-12-24 | 2015-05-13 | 深圳市腾讯计算机系统有限公司 | Method and device for processing traffic sign symbols |
Non-Patent Citations (1)
| Title |
|---|
| Efficient Recognition of Speed Limit Signs;Jim Torresen 等;《2004 IEEE Intelligent Transportation Systems Conference》;20041006;第652-656页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105718859A (en) | 2016-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11410277B2 (en) | Method and device for blurring image background, storage medium and electronic apparatus | |
| CN109636771B (en) | Flight target detection method and system based on image processing | |
| Ardianto et al. | Real-time traffic sign recognition using color segmentation and SVM | |
| JP6230751B1 (en) | Object detection apparatus and object detection method | |
| CN106062824B (en) | edge detecting device and edge detection method | |
| CN110490839B (en) | Method and device for detecting damaged area in expressway and computer equipment | |
| US20130342694A1 (en) | Method and system for use of intrinsic images in an automotive driver-vehicle-assistance device | |
| CN106250824A (en) | Vehicle window localization method and system | |
| CN105957107A (en) | Pedestrian detecting and tracking method and device | |
| CN107451976A (en) | A kind of image processing method and device | |
| CN112991374B (en) | Canny algorithm-based edge enhancement method, canny algorithm-based edge enhancement device, canny algorithm-based edge enhancement equipment and storage medium | |
| CN109670517A (en) | Object detection method, device, electronic equipment and target detection model | |
| CN111046746A (en) | License plate detection method and device | |
| CN107844761A (en) | The detection method and device of traffic sign | |
| CN109916415A (en) | Road type determines method, apparatus, equipment and storage medium | |
| CN111950523A (en) | Ship detection optimization method and device based on aerial photography, electronic equipment and medium | |
| CN115393322A (en) | Method and device for generating and evaluating change detection data based on digital twins | |
| CN107730487A (en) | A kind of image detecting method and device | |
| CN105718859B (en) | A kind of method that combination graphics processor carries out speed limitation board detection in automatic Pilot | |
| CN109948630A (en) | Recognition methods, device, system and the storage medium of target sheet image | |
| CN107767366B (en) | A kind of transmission line of electricity approximating method and device | |
| CN109523570B (en) | Motion parameter calculation method and device | |
| CN107343154A (en) | A kind of methods, devices and systems for the exposure parameter for determining camera device | |
| JP2014016710A (en) | Object detection device and program | |
| Tsai et al. | Vision-Based Obstacle Detection for Mobile Robot in Outdoor Environment. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |