CN107424116B - Parking space detection method based on side surround view camera - Google Patents
Parking space detection method based on side surround view camera Download PDFInfo
- Publication number
- CN107424116B CN107424116B CN201710534936.6A CN201710534936A CN107424116B CN 107424116 B CN107424116 B CN 107424116B CN 201710534936 A CN201710534936 A CN 201710534936A CN 107424116 B CN107424116 B CN 107424116B
- Authority
- CN
- China
- Prior art keywords
- parking
- parking space
- area
- vehicle
- image
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 26
- 230000009466 transformation Effects 0.000 claims abstract description 18
- 238000000034 method Methods 0.000 claims description 17
- 238000013528 artificial neural network Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 11
- 230000000007 visual effect Effects 0.000 claims description 11
- 238000012937 correction Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 238000003384 imaging method Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000012790 confirmation Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/04—Context-preserving transformations, e.g. by using an importance map
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a parking space detection method based on a side looking around camera, which comprises the steps of dividing a passable area in a looking around camera scene, calculating the actual size of the plane area through inverse perspective transformation under a ground coordinate system, confirming the area by utilizing multi-frame time sequence image data and combining a vehicle motion state, optimizing the position information of the actual parking space area, and finally realizing effective parking space detection. The invention has the characteristics of improving the parking space detection reliability and improving the parking accuracy.
Description
Technical Field
The invention relates to the technical field of vehicle-mounted electronics, in particular to a parking space detection method based on a side looking-around camera, which can improve the reliability and accuracy of parking space detection.
Background
With the improvement of the automation degree of the automobile, the automatic parking system becomes part of high-end automobile system standard configuration, and the real-time and accurate detection of the effective parking position is an important precondition for the realization of the system. Most of the existing automatic parking systems are based on ultrasonic radar sensors to detect parking spaces. The ultrasonic radar sensor has the characteristics of short measuring distance, easy signal interference between adjacent sensors and the like, and the size and the category of the obstacle cannot be distinguished. Therefore, effective parking space information cannot be accurately obtained in certain scenes, application scenes have certain limitations, and a driver is required to manually confirm the correctness of a parking area.
Visual systems are increasingly used in the field of active safety of vehicles. The 360-degree look-around system is one of the existing advanced automobile auxiliary safety systems, and the system can provide the conditions around the automobile for a driver under the low-speed working condition and provide visual assistance for the low-speed operation of the driver, and is already a standard configuration for a plurality of mass-produced automobile models. However, the existing mass production system can only provide visual assistance for the driver in the relevant area around the vehicle, and can not detect the feasibility of the area.
In order to solve the above problems, chinese patent publication No. CN104916163U discloses a parking space detection method in 2015, 9/16, comprising: acquiring binocular images of the side edges of a running vehicle, wherein the direction of an optical axis of the image acquisition is vertical to the running direction and is parallel to the bottom surface of the vehicle; detecting characteristic pixel points and distribution thereof contained in the binocular image, wherein the characteristic pixel points are used for representing the characteristics of a parking space line or a vehicle shape; judging whether any image in the binocular images contains all characteristic pixel points corresponding to a parking space line, and if so, prompting that a parking space is detected; and determining the distance between two obstacle vehicles parked at the parking space according to the characteristic pixel points in the binocular image and the distribution of the characteristic pixel points, and prompting that the parking space is between the two obstacle vehicles if the distance is not smaller than a preset threshold value. The invention has the advantages that: the parking space can be detected under the condition that the number of vehicles in the parking space is small, and the detection range is large. But has the following disadvantages: the invention utilizes an image recognition mode to detect the parking space, although the detection range is larger, the effective parking space information can not be accurately obtained, such as a hidden pit in the parking space area or an entrance/exit channel in the parking space area.
Disclosure of Invention
The invention provides a parking space detection method based on a side looking around camera, which can improve the reliability of parking space detection and improve the parking accuracy and can improve the parking space detection, aiming at overcoming the problems that in the prior art, an ultrasonic radar sensor has short measuring distance, signal interference is easy to generate between adjacent sensors, the sizes and the types of obstacles cannot be distinguished, effective parking space information cannot be accurately obtained in certain scenes, and a driver needs to manually confirm the correctness of a parking area.
In order to achieve the purpose, the invention adopts the following technical scheme:
a parking space detection method based on a side looking camera comprises the following steps:
(1-1) fisheye camera distortion correction
(1-1-1) removing radial distortion in the fisheye image by using the camera internal parameters obtained by calibration through a distortion model as follows: θ' ═ θ (1+ θ)2+θ4) Wherein, theta is an imaging perspective angle corresponding to the middle point of the image;
(1-2) side-looking around visible parking area segmentation
(1-2-1) training a deep neural network by adopting a supervised learning method, and dividing a side perspective view, a parking available area and a parking unavailable area in an image area;
(1-3) inverse perspective transformation of image parking available area
(1-3-1) calibrating parameters by using a side looking camera, carrying out inverse perspective transformation on the parking available area, and calculating the area of the parking available area under a road plane coordinate system;
(1-4) plan view parking space search
(1-4-1) confirming a potential parking space according to the area obtained in the step (1-3-1) by adopting set transverse and longitudinal parking space geometric dimension threshold information;
(1-5) timing confirmation of parking available area on road surface
(1-5-1) estimating vehicle motion information by using a steering wheel corner and a vehicle speed signal according to a vehicle kinematics model of a yaw plane, confirming a parking area by using a multi-frame time sequence picture, and screening and calculating the final position of a parking space according to a geometric size threshold of the parking space.
The invention calculates the actual size of the plane area by dividing the passable area in the panoramic camera scene and by inverse perspective transformation under a ground coordinate system, confirms the area by utilizing multi-frame time sequence image data and combining the vehicle motion state, optimizes the position information of the actual parking space area and finally realizes effective parking space detection. The invention has the characteristics of improving the parking space detection reliability and improving the parking accuracy.
Preferably, the side-view parking area dividing step further includes the steps of:
(1-2) further comprising the steps of:
(1-2-2) acquiring a side-looking video sample, performing distortion correction on video data, performing pixel-level calibration, dividing the image into a parking available area and a parking unavailable area, and improving calibration efficiency by adopting a network open source algorithm coarse calibration auxiliary manual calibration method;
(1-2-3) designing a deep neural network architecture, and adopting a full convolution network structure;
(1-2-4) carrying out supervision training on the deep neural network architecture designed in the step (1-2-3) by using the samples and the calibration labels collected in the step (1-2-2), wherein a gradient descent method based on a mini batch mode is adopted in the training process: in each cycle, solving an optimal solution for the softmax loss based on a reverse recursion method to optimize a network weight parameter until the set cycle iteration number is completed, wherein a calculation formula of the softmax loss is as follows:
wherein z isjFor each element of the output vector;
and (1-2-5) predicting image parking available areas by using the depth neural network parameters obtained by training in the step (1-2-4) and corrected video data acquired by the side surround view camera in real time to obtain parking available image areas in the visual angle of the side surround view camera at the moment.
Preferably, the step of performing inverse perspective transformation on the parking available area by using the calibration parameters of the side looking camera and calculating the area of the parking available area in the road plane coordinate system further comprises:
distortion corrected points [ u, v, 1 ] in the image coordinate system by a projective transformation matrix H]TCoordinate transformation to a point [ X, Y, 1 ] in the ground coordinate system according to the following formula]T：
[X，Y，1]T＝H*[u，v，1]T；
The calibration method of the projective transformation matrix H comprises the following steps: a checkerboard with a known size is placed in a specific area of a camera visual angle, corner detection is carried out on the checkerboard to obtain a corner set I1, a ground coordinate point set I2 corresponding to the corner set can be obtained through the geometric size and the relative position of the checkerboard, and H is obtained by matching I1 and I2 sets and calculating a set point distance matching error through least squares.
Preferably, the step of confirming the potential parking space according to the region obtained in the step (1-3-1) by using the set geometric size threshold information of the horizontal and vertical parking spaces further comprises the following steps:
defining the moment of starting parking space search, taking the right rear side angular point of the vehicle as the origin of a ground coordinate system, searching all potential parking spaces which accord with the threshold value of the size in a parking area in a top view by utilizing a transverse and longitudinal template with the minimum size of the parking spaces, defining the average position of the parking spaces as the arithmetic average value of the vertex positions of the potential parking spaces, defining the maximum range of the parking spaces as the maximum rectangular range in the potential parking space area, and recording the position of the vehicle at the current moment, the average position of the parking spaces and the maximum range in a parking space register.
Preferably, the method comprises the steps of estimating vehicle motion information by using a steering wheel corner and a vehicle speed signal according to a vehicle kinematics model of a yaw plane, confirming a parking area by using a multi-frame time sequence picture, and screening and calculating the final position of a parking space according to a parking space geometric size threshold, and further comprises the following steps:
according to the vehicle model of the yaw plane, the motion information of the vehicle under the ground coordinate system is calculated according to the following formula by using the vehicle speed v and the steering wheel angle signal delta:
x＝x0+∫vcos(δ)dt
y＝y0+∫vsin(δ)dt
according to the vehicle motion information, compensating the relative motion of the parking space area in the ground coordinate system, and if the average parking space position in the parking space register is continuously confirmed as an effective parking space at n moments, confirming and prompting the effective parking space information. And if the average parking position in the parking position register is not confirmed to be the effective parking position at the continuous n moments, updating the information in the parking position register again according to the maximum parking position range in the parking position register and the current-moment potential parking position information.
Therefore, the invention has the following beneficial effects: (1) the method can be fused with a parking space detection method based on an ultrasonic radar, so that the robustness of the system is improved; (2) the system can independently utilize the look-around system to detect the parking position and automatically park, thereby saving the system cost; (3) the parking space detection reliability can be improved; (4) the parking accuracy can be improved.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The invention is further described with reference to the following detailed description and accompanying drawings:
as shown in fig. 1, a parking space detection method based on a side view camera includes the following steps:
Step 110, removing radial distortion in the fisheye image by using the calibrated internal parameters of the camera through the following distortion model: θ' ═ θ (1+ θ)2+θ4) Wherein, theta is an imaging perspective angle corresponding to the middle point of the image;
Step 210, training a deep neural network by adopting a supervised learning method, and dividing a side-looking visual angle, a parking available area and a parking unavailable area in an image area;
step 220, collecting a side-looking video sample, performing distortion correction on video data, performing pixel-level calibration, dividing the image into a parking available area and a parking unavailable area, and improving calibration efficiency by adopting a network open source algorithm coarse calibration auxiliary manual calibration method;
step 230, designing a deep neural network architecture, and adopting a full convolution network structure;
in step 240, the deep neural network architecture designed in step 230 is supervised and trained by using the sample and the calibration label acquired in step 220. The training process adopts a gradient descent method based on a mini batch mode: in each cycle, solving an optimal solution for the softmax loss based on a reverse recursion method to optimize a network weight parameter until the set cycle iteration number is completed, wherein a calculation formula of the softmax loss is as follows:
where zj is each element of the output vector;
and 250, predicting image parking available areas by using the corrected video data acquired by the side looking-around camera in real time by using the deep neural network parameters obtained by training in the step 240, so as to obtain parking available image areas in the visual angle of the side looking-around camera at the moment.
Step 310, calibrating parameters by using a side looking camera, performing inverse perspective transformation on the parking available area, and calculating the area of the parking available area under a road plane coordinate system:
distortion corrected points [ u, v, 1 ] in the image coordinate system by a projective transformation matrix H]TCoordinate transformation to a point [ X, Y, 1 ] in the ground coordinate system according to the following formula]T：
[X，Y，1]T＝H*[u，v，1]T；
The calibration method of the projective transformation matrix H comprises the following steps: a checkerboard with a known size is placed in a specific area of a camera visual angle, corner detection is carried out on the checkerboard to obtain a corner set I1, a ground coordinate point set I2 corresponding to the corner set can be obtained through the geometric size and the relative position of the checkerboard, and H is obtained by matching I1 and I2 sets and calculating a set point distance matching error through least squares.
Step 410, defining the moment of starting parking space search, taking the right rear side angular point of the vehicle as the origin of a ground coordinate system, searching all potential parking spaces meeting the threshold value of the size in a parking area in a top view by using a transverse and longitudinal template with the minimum size of the parking spaces, defining the average position of the parking spaces as the arithmetic average value of the top positions of the potential parking spaces, defining the maximum range of the parking spaces as the maximum rectangular range in the potential parking space area, and recording the position of the vehicle at the current moment, the average position of the parking spaces and the maximum range in a parking space register.
Step 510, according to the vehicle model of the yaw plane, calculating the motion information of the vehicle under the ground coordinate system according to the following formula by using the vehicle speed v and the steering wheel angle signal delta:
x＝x0+∫vcos(δ)dt
y＝y0+∫vsin(δ)dt
compensating the relative motion of a parking space area in a ground coordinate system according to the motion information of the vehicle, confirming and prompting the effective parking space information if the average parking space position in the parking space register is confirmed as an effective parking space at each moment within 2 minutes, and updating the information in the parking space register again according to the maximum parking space range in the parking space register and the potential parking space information at the current moment if the average parking space position in the parking space register is not confirmed as an effective parking space at each moment within 2 minutes.
It should be understood that this example is for illustrative purposes only and is not intended to limit the scope of the present invention. Further, it should be understood that various changes or modifications of the present invention may be made by those skilled in the art after reading the teaching of the present invention, and such equivalents may fall within the scope of the present invention as defined in the appended claims.
Claims (3)
1. A parking space detection method based on a side looking camera is characterized by comprising the following steps:
(1-1) fisheye camera distortion correction
(1-1-1) removing radial distortion in the fisheye image by using the camera internal parameters obtained by calibration through a distortion model as follows: θ' ═ θ (1+ θ)2+θ4) Wherein, theta is an imaging perspective angle corresponding to the middle point of the image;
(1-2) side-looking around visible parking area segmentation
(1-2-1) training a deep neural network by adopting a supervised learning method, and dividing a side perspective view, a parking available area and a parking unavailable area in an image area;
(1-3) inverse perspective transformation of image parking available area
(1-3-1) calibrating parameters by using a side looking camera, carrying out inverse perspective transformation on the parking available area, and calculating the area of the parking available area under a road plane coordinate system;
distortion corrected points [ u, v, 1 ] in the image coordinate system by a projective transformation matrix H]TCoordinate transformation to a point [ X, Y, 1 ] in the ground coordinate system according to the following formula]T：
[X，Y，1]T＝H*[u，v，1]T；
The calibration method of the projective transformation matrix H comprises the following steps: placing a checkerboard with a known size in a specific area of a camera visual angle, carrying out corner detection on the checkerboard to obtain a corner set I1, wherein a ground coordinate point set I2 corresponding to the corner set can be obtained through the geometric size and the relative position of the checkerboard, and obtaining H by matching I1 with an I2 set and calculating a set point distance matching error by using least squares;
(1-4) plan view parking space search
(1-4-1) confirming a potential parking space according to the area obtained in the step (1-3-1) by adopting set transverse and longitudinal parking space geometric dimension threshold information;
defining the moment of starting parking space search, taking an angular point on the right rear side of a vehicle as the origin of a ground coordinate system, searching all potential parking spaces which accord with a threshold value of the size in a parking area in a top view by utilizing a transverse and longitudinal template with the minimum size of the parking spaces, defining the average position of the parking spaces as the arithmetic average value of the vertex positions of the potential parking spaces, defining the maximum range of the parking spaces as the maximum rectangular range in the potential parking space area, and recording the position of the vehicle at the current moment and the average position and the maximum range of the parking spaces in a parking space register;
(1-5) timing confirmation of parking available area on road surface
(1-5-1) estimating vehicle motion information by using a steering wheel corner and a vehicle speed signal according to a vehicle kinematics model of a yaw plane, confirming a parking area by using a multi-frame time sequence picture, and screening and calculating the final position of a parking space according to a geometric size threshold of the parking space.
2. A parking space detection method based on a side looking around camera as claimed in claim 1, wherein the step (1-2) further comprises the steps of:
(1-2-2) acquiring a side-looking video sample, performing distortion correction on video data, performing pixel-level calibration, dividing the image into a parking available area and a parking unavailable area, and improving calibration efficiency by adopting a network open source algorithm coarse calibration auxiliary manual calibration method;
(1-2-3) designing a deep neural network architecture, and adopting a full convolution network structure;
(1-2-4) carrying out supervision training on the deep neural network architecture designed in the step (1-2-3) by using the samples and the calibration labels collected in the step (1-2-2), wherein a gradient descent method based on a mini batch mode is adopted in the training process: in each cycle, solving an optimal solution for the softmax loss based on a reverse recursion method to optimize a network weight parameter until the set cycle iteration number is completed, wherein a calculation formula of the softmax loss is as follows:
wherein z isjFor each element of the output vector;
and (1-2-5) carrying out image parking area prediction on corrected video data acquired by the side looking-around camera in real time by utilizing the deep neural network parameters obtained by training in the step (1-2-4) to obtain a parking image area in the visual angle of the real-time side looking-around camera.
3. A parking space detection method based on a side looking around camera as claimed in claim 1, wherein the step (1-5-1) further comprises the steps of:
according to the vehicle model of the yaw plane, the motion information of the vehicle under the ground coordinate system is calculated according to the following formula by using the vehicle speed v and the steering wheel angle signal delta:
x＝x0+∫v cos(δ)dt
y＝y0+∫v sin(δ)dt
compensating the relative motion of a parking space area in a ground coordinate system according to the motion information of the vehicle, confirming and prompting the effective parking space information if the average parking space position in the parking space register is continuously confirmed as an effective parking space at n moments, and updating the information in the parking space register again according to the maximum parking space range in the parking space register and the potential parking space information at the current moment if the average parking space position in the parking space register is not continuously confirmed as an effective parking space at n moments.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710534936.6A CN107424116B (en) | 2017-07-03 | 2017-07-03 | Parking space detection method based on side surround view camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710534936.6A CN107424116B (en) | 2017-07-03 | 2017-07-03 | Parking space detection method based on side surround view camera |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107424116A CN107424116A (en) | 2017-12-01 |
| CN107424116B true CN107424116B (en) | 2020-06-19 |
Family
ID=60426806
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710534936.6A Active CN107424116B (en) | 2017-07-03 | 2017-07-03 | Parking space detection method based on side surround view camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107424116B (en) |
Families Citing this family (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107680113A (en) * | 2017-10-27 | 2018-02-09 | 武汉大学 | The image partition method of multi-layer segmentation network based on Bayesian frame edge prior |
| CN107993488B (en) * | 2017-12-13 | 2021-07-06 | 深圳市航盛电子股份有限公司 | Parking space identification method, system and medium based on fisheye camera |
| CN108090455B (en) * | 2017-12-27 | 2023-08-22 | 北京纵目安驰智能科技有限公司 | Cascade mechanism-based parking space line vertex positioning method, system, terminal and medium |
| CN108281041A (en) * | 2018-03-05 | 2018-07-13 | 东南大学 | A kind of parking space's detection method blended based on ultrasonic wave and visual sensor |
| CN108860139B (en) * | 2018-04-11 | 2019-11-29 | 浙江零跑科技有限公司 | A kind of automatic parking method for planning track based on depth enhancing study |
| CN108357269B (en) * | 2018-04-12 | 2023-05-02 | 电子科技大学中山学院 | Intelligent pen rack |
| CN110654373A (en) * | 2018-06-29 | 2020-01-07 | 比亚迪股份有限公司 | Automatic parking method and device and vehicle |
| CN109146978B (en) * | 2018-07-25 | 2021-12-07 | 南京富锐光电科技有限公司 | High-speed camera imaging distortion calibration device and method |
| CN109086708A (en) * | 2018-07-25 | 2018-12-25 | 深圳大学 | A kind of parking space detection method and system based on deep learning |
| CN109063632B (en) * | 2018-07-27 | 2022-02-01 | 重庆大学 | Parking space characteristic screening method based on binocular vision |
| CN109147321A (en) * | 2018-08-23 | 2019-01-04 | 重庆文理学院 | ITS new model and its construction method and intelligent transportation system under Internet of Things |
| CN109446909B (en) * | 2018-09-27 | 2020-12-29 | 山东省科学院自动化研究所 | Monocular distance measurement auxiliary parking system and method |
| CN111028534B (en) * | 2018-10-09 | 2022-04-26 | 杭州海康威视数字技术股份有限公司 | Parking space detection method and device |
| CN111098850A (en) * | 2018-10-25 | 2020-05-05 | 北京初速度科技有限公司 | Automatic parking auxiliary system and automatic parking method |
| CN109584304A (en) * | 2018-12-07 | 2019-04-05 | 中国科学技术大学 | A kind of steering wheel angle measurement method and device, system |
| CN109800658B (en) * | 2018-12-26 | 2023-05-26 | 中汽研（天津）汽车工程研究院有限公司 | Parking space type online identification and positioning system and method based on neural network |
| CN109801234B (en) * | 2018-12-28 | 2023-09-22 | 南京美乐威电子科技有限公司 | Image geometry correction method and device |
| CN111508260A (en) * | 2019-01-30 | 2020-08-07 | 上海欧菲智能车联科技有限公司 | Vehicle parking space detection method, device and system |
| CN109649384B (en) * | 2019-02-15 | 2020-08-14 | 华域汽车系统股份有限公司 | Parking assisting method |
| CN112016349B (en) * | 2019-05-29 | 2024-06-11 | 北京市商汤科技开发有限公司 | Parking space detection method and device and electronic equipment |
| CN110942434B (en) * | 2019-11-22 | 2023-05-05 | 华兴源创(成都)科技有限公司 | Display compensation system and method of display panel |
| CN111746525B (en) * | 2020-07-07 | 2022-06-21 | 东风柳州汽车有限公司 | Parking path planning method, device, equipment and storage medium |
| CN112668588B (en) * | 2020-12-29 | 2023-09-12 | 禾多科技（北京）有限公司 | Parking space information generation method, device, equipment and computer readable medium |
| CN112983085A (en) * | 2021-04-30 | 2021-06-18 | 的卢技术有限公司 | Parking space line identification method based on vision |
| CN113705474B (en) * | 2021-08-30 | 2022-04-15 | 北京易航远智科技有限公司 | Parking space detection method and device |
| CN114347982B (en) * | 2021-12-24 | 2024-06-14 | 北京经纬恒润科技股份有限公司 | Path planning method and system for automatic parking |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103763517A (en) * | 2014-03-03 | 2014-04-30 | 惠州华阳通用电子有限公司 | Vehicle-mounted around view display method and system |
| CN104512328A (en) * | 2013-09-27 | 2015-04-15 | 比亚迪股份有限公司 | Automobile all-round looking image generation method and automobile all-round looking system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6467838B2 (en) * | 2014-09-26 | 2019-02-13 | アイシン精機株式会社 | Perimeter monitoring device and perimeter monitoring system |
-
2017
- 2017-07-03 CN CN201710534936.6A patent/CN107424116B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104512328A (en) * | 2013-09-27 | 2015-04-15 | 比亚迪股份有限公司 | Automobile all-round looking image generation method and automobile all-round looking system |
| CN103763517A (en) * | 2014-03-03 | 2014-04-30 | 惠州华阳通用电子有限公司 | Vehicle-mounted around view display method and system |
Non-Patent Citations (2)
| Title |
|---|
| a Line Segment Detector;Rafael Grompone von Gioi等;《Image Processing On Line》;20120324;第35-55页 * |
| 一种基于环视相机的自动泊车方法;王旭东 等;《上海交通大学学报》;20130731;第47卷(第7期);第1077-1081、1086页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107424116A (en) | 2017-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107424116B (en) | Parking space detection method based on side surround view camera | |
| JP6821712B2 (en) | Calibration of integrated sensor in natural scene | |
| CN111448478B (en) | System and method for correcting high-definition maps based on obstacle detection | |
| CN109945858B (en) | Multi-sensing fusion positioning method for low-speed parking driving scene | |
| CN111436216B (en) | Method and system for color point cloud generation | |
| CN106952308B (en) | Method and system for determining position of moving object | |
| CN110031829B (en) | Target accurate distance measurement method based on monocular vision | |
| CN109435942A (en) | A kind of parking stall line parking stall recognition methods and device based on information fusion | |
| CN108759823B (en) | Low-speed automatic driving vehicle positioning and deviation rectifying method on designated road based on image matching | |
| CN102682292A (en) | Method based on monocular vision for detecting and roughly positioning edge of road | |
| CN112136021B (en) | System and method for constructing landmark-based high definition map | |
| CN112740225A (en) | Method and device for determining road surface elements | |
| CN111678518A (en) | Visual positioning method for correcting automatic parking path | |
| CN114693787A (en) | Parking garage map building and positioning method and system and vehicle | |
| CN112424568B (en) | System and method for constructing high-definition map | |
| CN111539278A (en) | Detection method and system for target vehicle | |
| US11138448B2 (en) | Identifying a curb based on 3-D sensor data | |
| AU2018102199A4 (en) | Methods and systems for color point cloud generation | |
| US20240142590A1 (en) | Online sensor alignment using feature registration | |
| Albrecht et al. | Concept on landmark detection in road scene images taken from a top-view camera system | |
| CN117808895A (en) | Non-target external parameter calibration method and related device for laser radar and camera | |
| CN117953046A (en) | Data processing method, device, controller, vehicle and storage medium | |
| CN118711398A (en) | Road condition detection method, device, equipment, medium and program product | |
| CN117292355A (en) | Target fusion perception method and device, computer equipment and storage medium | |
| CN113218361A (en) | Camera ranging method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 6 / F, Xintu building, 451 Internet of things street, Binjiang District, Hangzhou City, Zhejiang Province, 310051 Patentee after: Zhejiang Zero run Technology Co.,Ltd. Address before: 6 / F, Xintu building, 451 Internet of things street, Binjiang District, Hangzhou City, Zhejiang Province, 310051 Patentee before: ZHEJIANG LEAPMOTOR TECHNOLOGY Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |