CN110796023B - Recognition method for parking state wheel positions in interaction area of AGV intelligent parking system - Google Patents

Recognition method for parking state wheel positions in interaction area of AGV intelligent parking system Download PDF

Info

Publication number
CN110796023B
CN110796023B CN201910955512.6A CN201910955512A CN110796023B CN 110796023 B CN110796023 B CN 110796023B CN 201910955512 A CN201910955512 A CN 201910955512A CN 110796023 B CN110796023 B CN 110796023B
Authority
CN
China
Prior art keywords
view
vehicle body
vehicle
wheel
coordinates
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.)
Expired - Fee Related
Application number
CN201910955512.6A
Other languages
Chinese (zh)
Other versions
CN110796023A (en
Inventor
杜启亮
朱伟枝
向照夷
田联房
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China University of Technology SCUT
Zhuhai Institute of Modern Industrial Innovation of South China University of Technology
Original Assignee
South China University of Technology SCUT
Zhuhai Institute of Modern Industrial Innovation of South China University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by South China University of Technology SCUT, Zhuhai Institute of Modern Industrial Innovation of South China University of Technology filed Critical South China University of Technology SCUT
Priority to CN201910955512.6A priority Critical patent/CN110796023B/en
Publication of CN110796023A publication Critical patent/CN110796023A/en
Application granted granted Critical
Publication of CN110796023B publication Critical patent/CN110796023B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • E—FIXED CONSTRUCTIONS
    • E04—BUILDING
    • E04H—BUILDINGS OR LIKE STRUCTURES FOR PARTICULAR PURPOSES; SWIMMING OR SPLASH BATHS OR POOLS; MASTS; FENCING; TENTS OR CANOPIES, IN GENERAL
    • E04H6/00—Buildings for parking cars, rolling-stock, aircraft, vessels or like vehicles, e.g. garages
    • E04H6/42—Devices or arrangements peculiar to garages, not covered elsewhere, e.g. securing devices, safety devices, monitoring and operating schemes; centering devices
    • E04H6/426—Parking guides
    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/08—Detecting or categorising vehicles

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Architecture (AREA)
  • Multimedia (AREA)
  • Geometry (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Civil Engineering (AREA)
  • Structural Engineering (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a method for identifying the wheel position of a parking state in an interactive area of an AGV intelligent parking system, which comprises the steps of collecting images by a main camera at an overlooking angle right above the interactive area of the intelligent parking system, accurately identifying a vehicle target, extracting and analyzing vehicle body position information when the vehicle target is in a parking state, and solving the position of a wheel in the overlooking image collected by the main camera by combining with wheel related information extracted from a vehicle side image collected by an auxiliary camera, thereby providing effective information for an automatic parking robot to carry the vehicle, and effectively improving the intelligent level of the management of the AGV parking system.

Description

Recognition method for parking state wheel positions in interaction area of AGV intelligent parking system
Technical Field
The invention relates to the technical field of intelligent parking system design and management, in particular to a method for identifying parking state wheel positions in an interactive area of an AGV intelligent parking system.
Background
At present, with the rapid increase of the number of motor vehicles kept in China, the problem of difficult parking is increasingly highlighted. The AGV intelligent parking system can effectively solve the problem of difficulty in parking, the parking robot automatically moves forwards, backwards and turns according to actual conditions, wheels are clamped through a mechanical device after the parking robot enters the bottom of a vehicle, then the vehicle is jacked up, and the vehicle is transported to an appointed parking space according to positioning navigation. However, the parking states of the interactive area of the system are various, and after the parking robot drills into the bottom of the vehicle, the wheel positions need to be recognized to jack up the vehicle. If the parking robot does not have the position information of the wheels before, the wheels are difficult to be quickly clamped by a mechanical device only by a sensor arranged on the parking robot, and the working efficiency of the system is influenced.
Therefore, the invention hopes to extract the accurate positions of the vehicle wheels in the interactive area by designing the recognition method of the parking state wheel positions in the interactive area of the AGV intelligent parking system, and provide effective information for the automatic parking robot to carry the vehicle.
Disclosure of Invention
The invention aims to overcome the defects and shortcomings of the prior art, and provides a method for identifying the wheel position of a parking state in an interactive area of an AGV intelligent parking system, which has the characteristics of simple structure, convenience in use, low cost and the like.
In order to achieve the purpose, the technical scheme provided by the invention is as follows: the method for identifying the parking state wheel position of the AGV intelligent parking system interaction area comprises the following steps:
1) acquiring video information containing a complete vehicle body side image through an auxiliary camera, transmitting the video information to a computer, and further preprocessing a video frame through the computer;
2) preprocessing a video frame by a computer in the step 1), and extracting the wheel outline and the minimum external rectangle of the vehicle body of the vehicle in a side view by adopting a feature extraction method, so as to obtain the central coordinates and the wheel diameters of the front wheel and the rear wheel in the side view; the total length of the vehicle body in the side view is obtained through a distance formula between two points, the distances from the centers of the front and rear wheels to a front bumper of the vehicle respectively in the side view are obtained through a distance formula from the points to a straight line, the ratio of the distance from the centers of the front and rear wheels to the front bumper of the vehicle respectively in the side view to the total length of the vehicle body in the side view is further obtained, and the ratio of the diameter of the vehicle wheel in the side view to the total length of the vehicle body in the side view is obtained; adopting a pixel coordinate system as a common coordinate system where all coordinate values in the image are located, wherein the origin of the pixel coordinate system is at the upper left corner of the image, the horizontal direction is the x axis, the horizontal coordinate value is increased to the right, the longitudinal direction is the y axis, and the vertical coordinate value is increased to the lower side;
3) acquiring video information containing a complete interaction area of the AGV intelligent parking system through a main camera arranged right above the interaction area of the AGV intelligent parking system, transmitting the video information to a computer, and further preprocessing a video frame through the computer;
4) preprocessing the video frame by the computer in the step 3), and then acquiring a foreground image of the vehicle under the top view by adopting a moving target detection algorithm, so as to obtain a minimum circumscribed rectangle of the vehicle body under the top view; under the parking state, the center coordinates of the front wheel and the rear wheel under the top view are calculated by utilizing the coordinates of four vertexes of the minimum circumscribed rectangle of the vehicle body under the top view and the ratio of the distance from the center of the front wheel and the center of the rear wheel under the side view to a front bumper of the vehicle obtained in the step 2) to the total length of the vehicle body under the side view; adopting a pixel coordinate system as a common coordinate system where all coordinate values in the image are located, wherein the origin of the pixel coordinate system is at the upper left corner of the image, the horizontal direction is the x axis, the horizontal coordinate value is increased to the right, the longitudinal direction is the y axis, and the vertical coordinate value is increased to the lower side;
5) and (3) calculating the total length of the vehicle body in the top view by using the coordinates of four vertexes of the minimum circumscribed rectangle of the vehicle body in the top view, and calculating the diameter of the vehicle wheel in the top view by combining the ratio of the diameter of the vehicle wheel in the side view obtained in the step 2) to the total length of the vehicle body in the side view.
In step 1), the auxiliary camera is arranged in any area of the whole AGV intelligent parking system, wherein the area can collect video information containing complete images of the side faces of the vehicle body, and the vehicle information collected by the auxiliary camera is required to be in one-to-one correlation with the information collected by the main camera.
In the step 2), the wheel outline and the minimum external rectangle of the vehicle body of the vehicle in the side view are extracted by adopting a characteristic extraction method, so that the central coordinates PWF (x) of the front wheel and the rear wheel in the side view are obtainedPWF,yPWF)、PWB(xPWB,yPWB) And wheel diameter D in side viewCAnd four vertex coordinates of a minimum bounding rectangle of the vehicle body in the side view, wherein the vertex with the maximum longitudinal coordinate value is defined as PC0 (x)PC0,yPC0) If the ordinate values of the two vertexes are the maximum, the vertex having the maximum ordinate value and the minimum abscissa value is defined as PC0 (x)PC0,yPC0) And the remaining three vertices are PC0 (x)PC0,yPC0) As a starting point, sequentially defined as PC1 (x) in clockwise orderPC1,yPC1)、PC2(xPC2,yPC2)、PC3(xPC3,yPC3)；
According to the distance formula between two points, the vertex coordinate PC0 (x) of the minimum bounding rectangle of the car body in the side view is utilizedPC0,yPC0)、PC1(xPC1,yPC1) Determining the total length L of the vehicle body in the side viewCComprises the following steps:
Figure BDA0002227150510000031
according to the distance formula from the point to the straight line, the central coordinates PWF (x) of the front wheel and the rear wheel in the side view are utilizedPWF,yPWF)、PWB(xPWB,yPWB) The vertex coordinates PC0 (x) of the minimum bounding rectangle of the vehicle body in the side viewPC0,yPC0)、PC3(xPC3,yPC3) Determining the distance L from the center of the front and rear wheels to the front bumper of the vehicle in the side viewF、LBComprises the following steps:
Figure BDA0002227150510000041
by means of LC、LF、LBThe Ratio of the distance from the center of the front wheel to the front bumper of the vehicle to the center of the rear wheel to the total length of the vehicle body in the side view is calculatedF、RatioBComprises the following steps:
Figure BDA0002227150510000042
by means of LC、DCAnd determining the Ratio of the diameter of the wheel under the side view to the total length of the vehicle body under the side viewWComprises the following steps:
RatioW＝DCLC。
in the step 3), the main camera is installed right above the interaction area of the AGV intelligent parking system, collects video information containing the complete interaction area of the AGV intelligent parking system from a overlooking angle, and transmits the video information to the computer.
In the step 4), under the pixel coordinate system, the x axis rotates anticlockwise, and the first side of the minimum circumscribed rectangle of the vehicle body under the first encountering plan view is defined as LHThe other with LHThe mutually perpendicular sides are defined as LWDefining θ as x-axis and LHThe value range of theta is [0 degrees and 90 degrees ]; in the parking state, the vehicle body is divided into three states of left deviation, right deviation and body straightening according to different angles of the vehicle body; when L isH＞LWWhen the vehicle body deviates to the right; when L isH＜LWIn time, two cases are distinguished: when theta is larger than 0 DEG, the vehicle body deviates to the left, and when theta is equal to 0 DEG, the vehicle body is straight; among the four vertex coordinates of the minimum circumscribed rectangle of the vehicle body in the plan view, the vertex with the largest ordinate value is defined as P0 (x)P0,yP0) If the ordinate values of the two vertexes are maximum, the vertex having the maximum ordinate value and the minimum abscissa value is defined as P0 (x)P0,yP0) And the other three vertexes are represented by P0 (x)P0,yP0) As a starting point, sequentially defined as P1 (x) in clockwise orderP1,yP1)、P2(xP2,yP2)、P3(xP3,yP3)；
When the vehicle is in a left inclined state and a right straight state, the center coordinates of a left front wheel, a right front wheel, a left rear wheel and a right rear wheel of the vehicle are respectively PW3 (x)PW3,yPW3)、PW0(xPW0,yPW0)、PW2(xPW2,yPW2)、PW1(xPW1,yPW1) (ii) a Using the coordinates P0 (x) of four vertexes of the minimum bounding rectangle of the car body in the top viewP0,yP0)、P1(xP1,yP1)、P2(xP2,yP2)、P3(xP3,yP3) And the Ratio of the distance from the center of the front wheel to the front bumper of the vehicle to the center of the rear wheel to the total length of the vehicle body in the side view obtained in the step 2) isF、RatioBAnd calculating the center coordinates of the front wheel and the rear wheel under the plan view as follows:
Figure BDA0002227150510000051
when the vehicle is in a right-leaning state, the center coordinates of the left front wheel, the right front wheel, the left rear wheel and the right rear wheel of the wheels are respectively PW0 (x)PW0,yPW0)、PW1(xPW1,yPW1)、PW3(xPW3,yPW3)、PW2(xPW2,yPW2) (ii) a Using the coordinates P0 (x) of four vertexes of the minimum bounding rectangle of the car body in the top viewP0,yP0)、P1(xP1,yP1)、P2(xP2,yP2)、P3(xP3,yP3) And the Ratio of the distance from the center of the front wheel to the front bumper of the vehicle to the center of the rear wheel to the total length of the vehicle body in the side view obtained in the step 2) isF、RatioBAnd calculating the center coordinates of the front wheel and the rear wheel under the plan view as follows:
Figure BDA0002227150510000052
in step 5), under the pixel coordinate system,the x axis rotates anticlockwise, and the first side of the minimum circumscribed rectangle of the vehicle body under the first encountering plan view is defined as LHThe other with LHThe mutually perpendicular sides are defined as LWDefining θ as x-axis and LHThe value range of theta is [0 degrees and 90 degrees ]; in the parking state, the vehicle body is divided into three states of left deviation, right deviation and body straightening according to different angles of the vehicle body; when L isH＞LWWhen the vehicle body deviates to the right; when L isH＜LWIn time, two cases are distinguished: when theta is larger than 0 DEG, the vehicle body deviates to the left, and when theta is equal to 0 DEG, the vehicle body is straight; among the four vertex coordinates of the minimum circumscribed rectangle of the vehicle body in the plan view, the vertex with the largest ordinate value is defined as P0 (x)P0,yP0) If the ordinate values of the two vertexes are maximum, the vertex having the maximum ordinate value and the minimum abscissa value is defined as P0 (x)P0,yP0) And the other three vertexes are represented by P0 (x)P0,yP0) As a starting point, sequentially defined as P1 (x) in clockwise orderP1,yP1)、P2(xP2,yP2)、P3(xP3,yP3)；
When the vehicle is in a state that the vehicle body is deviated from the left and the vehicle body is straight, according to a distance formula between two points, the vertex coordinate P0 (x) of the minimum circumscribed rectangle of the vehicle body in the plan view is utilizedP0,yP0)、P1(xP1,yP1) Calculating the total length L of the vehicle body in plan viewDComprises the following steps:
Figure BDA0002227150510000061
when the vehicle is in a right-leaning state, according to a distance formula between two points, the vertex coordinate P1 (x) of the minimum circumscribed rectangle of the vehicle body in a plan view is utilizedP1,yP1)、P2(xP2,yP2) Calculating the total length L of the vehicle body in plan viewDComprises the following steps:
Figure BDA0002227150510000062
the Ratio of the wheel diameter under the side view and the total length of the vehicle body under the side view, which is obtained in the step 2), is reusedWDetermining the wheel diameter D in plan viewDComprises the following steps:
DD＝RatioW·LD。
compared with the prior art, the invention has the following advantages and beneficial effects:
1. the specific positions of the vehicle wheels in the parking interaction area of the AGV intelligent parking system are identified and extracted, effective information is provided for the automatic parking robot to carry the vehicle, and the working efficiency of the system is improved.
2. And effective information is provided for the design and management of the intelligent parking system.
3. And information fusion is carried out with the prior art, so that the identification accuracy is improved.
4. The detection algorithm is simple and accurate, has good real-time performance and is suitable for general application.
Drawings
Fig. 1 is a schematic layout of two main and auxiliary cameras according to the present invention.
Fig. 2 is a schematic side view of the wheel and body contour extraction of the present invention.
Fig. 3 is a schematic side view of the present invention.
Fig. 4 is a schematic diagram showing three states of the vehicle parking in a plan view according to the present invention.
Fig. 5 is a schematic diagram of wheel position information corresponding to a left offset of a vehicle body according to the present invention.
Fig. 6 is a schematic diagram of wheel position information corresponding to a right-hand yaw of a vehicle body according to the present invention.
Detailed Description
The following is a further description with reference to specific examples.
The method for identifying the parking state wheel position of the interactive area of the AGV intelligent parking system comprises the following steps:
step 1: referring to fig. 1, video information including a complete vehicle body side image is acquired by an auxiliary camera, and transmitted to a computer, and then a video frame is preprocessed by the computer. The auxiliary camera is arranged in any area of the whole AGV intelligent parking system, wherein the area can collect video information containing complete vehicle body side images. The vehicle information collected by the auxiliary camera must be associated with the information of the main camera one by one. The invention is explained by taking the side arranged in the interactive area of the AGV intelligent parking system as an example.
Step 2: the wheel contour and the minimum bounding rectangle of the vehicle body of the vehicle in the side view are extracted by adopting a feature extraction method, as shown in fig. 2, the wheel contour and the minimum bounding rectangle of the vehicle body of the vehicle in the side view are extracted by adopting the feature extraction method, and thus the central coordinates PWF (x) of the front wheel and the rear wheel in the side view are obtainedPWF,yPWF)、PWB(xPWB,yPWB) And wheel diameter D in side viewCAnd four vertex coordinates of a minimum bounding rectangle of the vehicle body in the side view, wherein the vertex with the maximum longitudinal coordinate value is defined as PC0 (x)PC0,yPC0) If the ordinate values of the two vertexes are the maximum, the vertex having the maximum ordinate value and the minimum abscissa value is defined as PC0 (x)PC0,yPC0) And the remaining three vertices are PC0 (x)PC0,yPC0) As a starting point, sequentially defined as PC1 (x) in clockwise orderPC1,yPC1)、PC2(xPC2,yPC2)、PC3(xPC3,yPC3) Referring to fig. 3, the side view is shown in fig. 3, wherein a pixel coordinate system is used as a common coordinate system where all coordinate values in the image are located, the origin of the pixel coordinate system is located at the upper left corner of the image, the horizontal direction is the x axis, the horizontal coordinate value increases progressively to the right, the vertical direction is the y axis, and the vertical coordinate value increases progressively downward.
According to the distance formula between two points, the vertex coordinate PC0 (x) of the minimum bounding rectangle of the car body in the side view is utilizedPC0,yPC0)、PC1(xPC1,yPC1) Determining the total length L of the vehicle body in the side viewCComprises the following steps:
Figure BDA0002227150510000081
according to the distance formula from point to straight line, the utilization sideCenter coordinates PWF (x) of front and rear wheels under viewPWF,yPWF)、PWB(xPWB,yPWB) The vertex coordinates PC0 (x) of the minimum bounding rectangle of the vehicle body in the side viewPC0,yPC0)、PC3(xPC3,yPC3) Determining the distance L from the center of the front and rear wheels to the front bumper of the vehicle in the side viewF、LBComprises the following steps:
Figure BDA0002227150510000082
by means of LC、LF、LBThe Ratio of the distance from the center of the front wheel to the front bumper of the vehicle to the center of the rear wheel to the total length of the vehicle body in the side view is calculatedF、RatioBComprises the following steps:
Figure BDA0002227150510000083
by means of LC、DCAnd determining the Ratio of the diameter of the wheel under the side view to the total length of the vehicle body under the side viewWComprises the following steps:
RatioW＝DCLC。
and step 3: referring to fig. 1, video information including a complete interactive area of the AGV intelligent parking system is collected by a main camera disposed right above the interactive area of the AGV intelligent parking system, and the video information is transmitted to a computer, so that a video frame is preprocessed by the computer. The image preprocessing method includes image graying, histogram equalization processing, and the like.
And 4, step 4: extracting the background of the image through Gaussian mixture modeling, separating a vehicle foreground image under the top view by adopting a background difference method, and performing morphological processing, shadow detection and elimination on the vehicle foreground image to obtain the minimum external rectangle of the vehicle body under the top view.
Under a pixel coordinate system, the x axis rotates anticlockwise, and a first side of a minimum circumscribed rectangle of the car body under the condition of first encountering a top view is defined as LHThe other with LHThe mutually perpendicular sides are defined as LWDefining θ as x-axis and LHThe value range of theta is [0 degrees and 90 degrees ]; referring to fig. 4, in the parking state, the vehicle body is divided into three states of left-side vehicle body deviation, right-side vehicle body deviation and right-side vehicle body straightening according to different angles of the vehicle body; when L isH＞LWWhen the vehicle body deviates to the right; when L isH＜LWIn time, two cases are distinguished: when theta is larger than 0 DEG, the vehicle body deviates to the left, and when theta is equal to 0 DEG, the vehicle body is straight; among the four vertex coordinates of the minimum circumscribed rectangle of the vehicle body in the plan view, the vertex with the largest ordinate value is defined as P0 (x)P0,yP0) If the ordinate values of the two vertexes are maximum, the vertex having the maximum ordinate value and the minimum abscissa value is defined as P0 (x)P0,yP0) And the other three vertexes are represented by P0 (x)P0,yP0) As a starting point, sequentially defined as P1 (x) in clockwise orderP1,yP1)、P2(xP2,yP2)、P3(xP3,yP3). In addition, when the vehicle body is in a straight state, the four vertexes corresponding to the minimum circumscribed rectangle of the vehicle body are the same as the left-side time of the vehicle body, so that the straight state of the vehicle body is the same as the left-side time of the vehicle body.
When the vehicle is in a vehicle body left-offset state, referring to fig. 5, the center coordinates of the left front wheel, the right front wheel, the left rear wheel and the right rear wheel of the vehicle are PW3 (x)PW3,yPW3)、PW0(xPW0,yPW0)、PW2(xPW2,yPW2)、PW1(xPW1,yPW1) (ii) a Using the coordinates P0 (x) of four vertexes of the minimum bounding rectangle of the car body in the top viewP0,yP0)、P1(xP1,yP1)、P2(xP2,yP2)、P3(xP3,yP3) And the Ratio of the distance from the center of the front wheel to the front bumper of the vehicle to the center of the rear wheel to the total length of the vehicle body in the side view obtained in the step 2) isF、RatioBAnd calculating the center coordinates of the front wheel and the rear wheel under the plan view as follows:
Figure BDA0002227150510000101
when the vehicle is in a state of right-leaning of the vehicle body, referring to fig. 6, the center coordinates of the left front wheel, the right front wheel, the left rear wheel and the right rear wheel of the wheels are respectively PW0 (x)PW0,yPW0)、PW1(xPW1,yPW1)、PW3(xPW3,yPW3)、PW2(xPW2,yPW2) (ii) a Using the coordinates P0 (x) of four vertexes of the minimum bounding rectangle of the car body in the top viewP0,yP0)、P1(xP1,yP1)、P2(xP2,yP2)、P3(xP3,yP3) And the Ratio of the distance from the center of the front wheel to the front bumper of the vehicle to the center of the rear wheel to the total length of the vehicle body in the side view obtained in the step 2) isF、RatioBAnd calculating the center coordinates of the front wheel and the rear wheel under the plan view as follows:
Figure BDA0002227150510000102
and 5: when the vehicle is in the left-hand position, the vertex coordinate P0 (x) of the minimum bounding rectangle of the vehicle body in the plan view obtained in the step 4 is used according to the distance formula between the two pointsP0,yP0)、P1(xP1,yP1) Calculating the total length L of the vehicle body in plan viewDComprises the following steps:
Figure BDA0002227150510000103
when the vehicle is in a state of right-hand body deviation, the vertex coordinate P1 (x) of the minimum bounding rectangle of the vehicle body in the plan view obtained in the step 4 is used according to the distance formula between the two pointsP1,yP1)、P2(xP2,yP2) Calculating the total length L of the vehicle body in plan viewDComprises the following steps:
Figure BDA0002227150510000111
the Ratio of the diameter of the wheel in the side view to the total length of the vehicle body in the side view, which is obtained in the step 2, is usedWDetermining the wheel diameter D in plan viewDComprises the following steps:
DD＝RatioW·LD
the above-mentioned embodiments are merely preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, so that the changes in the shape and principle of the present invention should be covered within the protection scope of the present invention.

Claims (5)

  1. The method for identifying the parking state wheel position in the interactive area of the AGV intelligent parking system is characterized by comprising the following steps of:
    1) acquiring video information containing a complete vehicle body side image through an auxiliary camera, transmitting the video information to a computer, and further preprocessing a video frame through the computer;
    2) preprocessing a video frame by a computer in the step 1), and extracting the wheel outline and the minimum external rectangle of the vehicle body of the vehicle in a side view by adopting a feature extraction method, so as to obtain the central coordinates and the wheel diameters of the front wheel and the rear wheel in the side view; the total length of the vehicle body in the side view is obtained through a distance formula between two points, the distances from the centers of the front and rear wheels to a front bumper of the vehicle respectively in the side view are obtained through a distance formula from the points to a straight line, the ratio of the distance from the centers of the front and rear wheels to the front bumper of the vehicle respectively in the side view to the total length of the vehicle body in the side view is further obtained, and the ratio of the diameter of the vehicle wheel in the side view to the total length of the vehicle body in the side view is obtained; adopting a pixel coordinate system as a common coordinate system where all coordinate values in the image are located, wherein the origin of the pixel coordinate system is at the upper left corner of the image, the horizontal direction is the x axis, the horizontal coordinate value is increased to the right, the longitudinal direction is the y axis, and the vertical coordinate value is increased to the lower side;
    3) acquiring video information containing a complete interaction area of the AGV intelligent parking system through a main camera arranged right above the interaction area of the AGV intelligent parking system, transmitting the video information to a computer, and further preprocessing a video frame through the computer;
    4) preprocessing the video frame by the computer in the step 3), and then acquiring a foreground image of the vehicle under the top view by adopting a moving target detection algorithm, so as to obtain a minimum circumscribed rectangle of the vehicle body under the top view; under the parking state, the center coordinates of the front wheel and the rear wheel under the top view are calculated by utilizing the coordinates of four vertexes of the minimum circumscribed rectangle of the vehicle body under the top view and the ratio of the distance from the center of the front wheel and the center of the rear wheel under the side view to a front bumper of the vehicle obtained in the step 2) to the total length of the vehicle body under the side view; adopting a pixel coordinate system as a common coordinate system where all coordinate values in the image are located, wherein the origin of the pixel coordinate system is at the upper left corner of the image, the horizontal direction is the x axis, the horizontal coordinate value is increased to the right, the longitudinal direction is the y axis, and the vertical coordinate value is increased to the lower side;
    5) and (3) calculating the total length of the vehicle body in the top view by utilizing coordinates of four vertexes of the minimum circumscribed rectangle of the vehicle body in the top view, and calculating the diameter of the vehicle wheel in the top view by combining the ratio of the diameter of the vehicle wheel in the side view obtained in the step 2) to the total length of the vehicle body in the side view, wherein the method specifically comprises the following steps:
    under a pixel coordinate system, the x axis rotates anticlockwise, and a first side of a minimum circumscribed rectangle of the car body under the condition of first encountering a top view is defined as LHThe other with LHThe mutually perpendicular sides are defined as LWDefining θ as x-axis and LHThe value range of theta is [0 degrees and 90 degrees ]; in the parking state, the vehicle body is divided into three states of left deviation, right deviation and body straightening according to different angles of the vehicle body; when L isH＞LWWhen the vehicle body deviates to the right; when L isH＜LWIn time, two cases are distinguished: when theta is larger than 0 DEG, the vehicle body deviates to the left, and when theta is equal to 0 DEG, the vehicle body is straight; among the four vertex coordinates of the minimum circumscribed rectangle of the vehicle body in the plan view, the vertex with the largest ordinate value is defined as P0 (x)P0，yP0) If the ordinate values of the two vertexes are maximum, the vertex having the maximum ordinate value and the minimum abscissa value is defined as P0 (x)P0，yP0) And the other three vertexes are represented by P0 (x)P0，yP0) As a starting point, sequentially defined as P1 (x) in clockwise orderP1，yP1)、P2(xP2，yP2)、P3(xP3，yP3)；
    When the vehicle is in a state that the vehicle body is deviated from the left and the vehicle body is straight, according to a distance formula between two points, the vertex coordinate P0 (x) of the minimum circumscribed rectangle of the vehicle body in the plan view is utilizedP0，yP0)、P1(xP1，yP1) Calculating the total length L of the vehicle body in plan viewDComprises the following steps:
    Figure FDA0003411604170000021
    when the vehicle is in a right-leaning state, according to a distance formula between two points, the vertex coordinate P1 (x) of the minimum circumscribed rectangle of the vehicle body in a plan view is utilizedP1，yP1)、P2(xP2，yP2) Calculating the total length L of the vehicle body in plan viewDComprises the following steps:
    Figure FDA0003411604170000022
    the Ratio of the wheel diameter under the side view and the total length of the vehicle body under the side view, which is obtained in the step 2), is reusedWDetermining the wheel diameter D in plan viewDComprises the following steps:
    DD＝RatioW·LD。
  2. 2. the method for identifying parking status wheel positions in an interactive area of an AGV parking system according to claim 1, wherein: in step 1), the auxiliary camera is arranged in any area of the whole AGV intelligent parking system, wherein the area can collect video information containing complete images of the side faces of the vehicle body, and the vehicle information collected by the auxiliary camera is required to be in one-to-one correlation with the information collected by the main camera.
  3. 3. The method for identifying parking status wheel positions in an interactive area of an AGV parking system according to claim 1, wherein: in the step 2), the wheels of the vehicle in the side view are extracted by adopting a characteristic extraction methodThe contour and the minimum circumscribed rectangle of the vehicle body are obtained, and the central coordinates PWF (x) of the front and rear wheels in the side view are obtainedPWF，yPWF)、PWB(xPWB，yPWB) And wheel diameter D in side viewCAnd four vertex coordinates of a minimum bounding rectangle of the vehicle body in the side view, wherein the vertex with the maximum longitudinal coordinate value is defined as PC0 (x)PC0，yPC0) If the ordinate values of the two vertexes are the maximum, the vertex having the maximum ordinate value and the minimum abscissa value is defined as PC0 (x)PC0，yPC0) And the remaining three vertices are PC0 (x)PC0，yPC0) As a starting point, sequentially defined as PC1 (x) in clockwise orderPC1，yPC1)、PC2(xPC2，yPC2)、PC3(xPC3，yPC3)；
    According to the distance formula between two points, the vertex coordinate PC0 (x) of the minimum bounding rectangle of the car body in the side view is utilizedPC0，yPC0)、PC1(xPC1，yPC1) Determining the total length L of the vehicle body in the side viewCComprises the following steps:
    Figure FDA0003411604170000031
    according to the distance formula from the point to the straight line, the central coordinates PWF (x) of the front wheel and the rear wheel in the side view are utilizedPWF，yPWF)、PWB(xPWB，yPWB) The vertex coordinates PC0 (x) of the minimum bounding rectangle of the vehicle body in the side viewPC0，yPC0)、PC3(xPC3，yPC3) Determining the distance L from the center of the front and rear wheels to the front bumper of the vehicle in the side viewF、LBComprises the following steps:
    Figure FDA0003411604170000032
    by means of LC、LF、LBDetermining the distance from the center of the front and rear wheels to the front bumper of the vehicle in the side view and the side viewRatio of total length of lower bodyF、RatioBComprises the following steps:
    Figure FDA0003411604170000041
    by means of LC、DCAnd determining the Ratio of the diameter of the wheel under the side view to the total length of the vehicle body under the side viewWComprises the following steps:
    RatioW＝DC/LC。
  4. 4. the method for identifying parking status wheel positions in an interactive area of an AGV parking system according to claim 1, wherein: in the step 3), the main camera is installed right above the interaction area of the AGV intelligent parking system, collects video information containing the complete interaction area of the AGV intelligent parking system from a overlooking angle, and transmits the video information to the computer.
  5. 5. The method for identifying parking status wheel positions in an interactive area of an AGV parking system according to claim 1, wherein: in the step 4), under the pixel coordinate system, the x axis rotates anticlockwise, and the first side of the minimum circumscribed rectangle of the vehicle body under the first encountering plan view is defined as LHThe other with LHThe mutually perpendicular sides are defined as LWDefining θ as x-axis and LHThe value range of theta is [0 degrees and 90 degrees ]; in the parking state, the vehicle body is divided into three states of left deviation, right deviation and body straightening according to different angles of the vehicle body; when L isH＞LWWhen the vehicle body deviates to the right; when L isH＜LWIn time, two cases are distinguished: when theta is larger than 0 DEG, the vehicle body deviates to the left, and when theta is equal to 0 DEG, the vehicle body is straight; among the four vertex coordinates of the minimum circumscribed rectangle of the vehicle body in the plan view, the vertex with the largest ordinate value is defined as P0 (x)P0，yP0) If the ordinate values of the two vertexes are maximum, the vertex having the maximum ordinate value and the minimum abscissa value is defined as P0 (x)P0，yP0) And the other three vertexes are represented by P0(xP0，yP0) As a starting point, sequentially defined as P1 (x) in clockwise orderP1，yP1)、P2(xP2，yP2)、P3(xP3，yP3)；
    When the vehicle is in a left inclined state and a right straight state, the center coordinates of a left front wheel, a right front wheel, a left rear wheel and a right rear wheel of the vehicle are respectively PW3 (x)PW3，yPW3)、PW0(xPW0，yPW0)、PW2(xPW2，yPW2)、PW1(xPW1，yPW1) (ii) a Using the coordinates P0 (x) of four vertexes of the minimum bounding rectangle of the car body in the top viewP0，yP0)、P1(xP1，yP1)、P2(xP2，yP2)、P3(xP3，yP3) And the Ratio of the distance from the center of the front wheel to the front bumper of the vehicle to the center of the rear wheel to the total length of the vehicle body in the side view obtained in the step 2) isF、RatioBAnd calculating the center coordinates of the front wheel and the rear wheel under the plan view as follows:
    Figure FDA0003411604170000051
    when the vehicle is in a right-leaning state, the center coordinates of the left front wheel, the right front wheel, the left rear wheel and the right rear wheel of the wheels are respectively PWO (x)PW0，yPW0)、PW1(xPW1，yPW1)、PW3(xPW3，yPW3)、PW2(xPW2，yPW2) (ii) a Using the coordinates P0 (x) of four vertexes of the minimum bounding rectangle of the car body in the top viewP0，yP0)、P1(xP1，yP1)、P2(xP2，yP2)、P3(xP3，yP3) And the Ratio of the distance from the center of the front wheel to the front bumper of the vehicle to the center of the rear wheel to the total length of the vehicle body in the side view obtained in the step 2) isF、RatioBAnd calculating the center coordinates of the front wheel and the rear wheel under the plan view as follows:
    Figure FDA0003411604170000052
CN201910955512.6A 2019-10-09 2019-10-09 Recognition method for parking state wheel positions in interaction area of AGV intelligent parking system Expired - Fee Related CN110796023B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910955512.6A CN110796023B (en) 2019-10-09 2019-10-09 Recognition method for parking state wheel positions in interaction area of AGV intelligent parking system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910955512.6A CN110796023B (en) 2019-10-09 2019-10-09 Recognition method for parking state wheel positions in interaction area of AGV intelligent parking system

Publications (2)

Publication Number Publication Date
CN110796023A CN110796023A (en) 2020-02-14
CN110796023B true CN110796023B (en) 2022-03-29

Family

ID=69438859

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910955512.6A Expired - Fee Related CN110796023B (en) 2019-10-09 2019-10-09 Recognition method for parking state wheel positions in interaction area of AGV intelligent parking system

Country Status (1)

Country Link
CN (1) CN110796023B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114753686B (en) * 2022-05-23 2023-08-15 广西交科集团有限公司 Variable-size parking device of self-calibration camera based on vanishing point detection

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103072528A (en) * 2013-01-30 2013-05-01 深圳市汉华安道科技有限责任公司 Vehicle and panoramic parking method and system thereof
CN105863351A (en) * 2016-05-26 2016-08-17 山东建筑大学 Autonomous parking system and method based on intelligent automobile transporters
CN105869432A (en) * 2016-03-29 2016-08-17 江苏大学 Method for identifying parking lot scene based on multiple-sensor fusion
CN107403454A (en) * 2017-08-03 2017-11-28 武汉纺织大学 A kind of sky parking's vehicle position parameter and dimensional parameters measuring system and method
CN109917417A (en) * 2019-03-14 2019-06-21 珠海丽亭智能科技有限公司 A kind of vehicle appearance measurement method and device and equipment based on photographic device
CN110077399A (en) * 2019-04-09 2019-08-02 魔视智能科技(上海)有限公司 A kind of vehicle collision avoidance method merged based on roadmarking, wheel detection
CN110176151A (en) * 2019-06-17 2019-08-27 北京精英路通科技有限公司 A kind of method, apparatus, medium and the equipment of determining parking behavior

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103072528A (en) * 2013-01-30 2013-05-01 深圳市汉华安道科技有限责任公司 Vehicle and panoramic parking method and system thereof
CN105869432A (en) * 2016-03-29 2016-08-17 江苏大学 Method for identifying parking lot scene based on multiple-sensor fusion
CN105863351A (en) * 2016-05-26 2016-08-17 山东建筑大学 Autonomous parking system and method based on intelligent automobile transporters
CN107403454A (en) * 2017-08-03 2017-11-28 武汉纺织大学 A kind of sky parking's vehicle position parameter and dimensional parameters measuring system and method
CN109917417A (en) * 2019-03-14 2019-06-21 珠海丽亭智能科技有限公司 A kind of vehicle appearance measurement method and device and equipment based on photographic device
CN110077399A (en) * 2019-04-09 2019-08-02 魔视智能科技(上海)有限公司 A kind of vehicle collision avoidance method merged based on roadmarking, wheel detection
CN110176151A (en) * 2019-06-17 2019-08-27 北京精英路通科技有限公司 A kind of method, apparatus, medium and the equipment of determining parking behavior

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
"A Vision-Based Autonmated Guided Vehicle System with Marker Recognition for Indooe Use";Jeisung Lee et al.;《sensors》;20130807;全文 *
"Research on Automatic Parking Systems Based on Parking Scene Recognition";SHIDIAN MA et al.;《IEEE Access》;20171018;全文 *
"基于环视系统的自动泊车研究";李宴 等;《2018中国汽车工程学会年会论文集》;20181231;全文 *

Also Published As

Publication number Publication date
CN110796023A (en) 2020-02-14

Similar Documents

Publication Publication Date Title
US10977504B2 (en) Vehicle-mounted image target objection recognition device
CN109386155B (en) Alignment method of unmanned parking transfer robot for automatic parking lot
CN110443225B (en) Virtual and real lane line identification method and device based on feature pixel statistics
US6906620B2 (en) Obstacle detection device and method therefor
CN111967360B (en) Target vehicle posture detection method based on wheels
CN113657224B (en) Method, device and equipment for determining object state in vehicle-road coordination
CN112967283B (en) Target identification method, system, equipment and storage medium based on binocular camera
CN108596058A (en) Running disorder object distance measuring method based on computer vision
CN107239746B (en) Obstacle identification and tracking method for road rescue safety monitoring
CN103593671B (en) The wide-range lane line visible detection method worked in coordination with based on three video cameras
CN111178150B (en) Lane line detection method, system and storage medium
CN111681283B (en) Monocular stereoscopic vision-based relative pose calculation method applied to wireless charging alignment
CN109827516B (en) Method for measuring distance through wheel
CN106813569A (en) A kind of automobile tire 3-D positioning method based on line-structured light
Li et al. Judgment and optimization of video image recognition in obstacle detection in intelligent vehicle
CN108154114B (en) Lane line detection method
CN110796023B (en) Recognition method for parking state wheel positions in interaction area of AGV intelligent parking system
CN114445636A (en) Train bottom item mapping method
CN116091603A (en) Box workpiece pose measurement method based on point characteristics
Cai et al. Robust road lane detection from shape and color feature fusion for vehicle self-localization
CN110633699B (en) Visual detection method for parking behavior of interaction area of AGV intelligent parking system
CN113486837B (en) Automatic driving control method for low-pass obstacle
CN115540750A (en) Monocular vision based object size detection method and system and storable medium
CN112241717B (en) Front vehicle detection method, and training acquisition method and device of front vehicle detection model
CN114429469A (en) Heading machine body pose determination method and system based on three-laser-spot target

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20220329

CF01 Termination of patent right due to non-payment of annual fee