CN110940319A - Height limit detection method and device based on machine vision recognition - Google Patents

Height limit detection method and device based on machine vision recognition Download PDF

Info

Publication number
CN110940319A
CN110940319A CN201910998023.9A CN201910998023A CN110940319A CN 110940319 A CN110940319 A CN 110940319A CN 201910998023 A CN201910998023 A CN 201910998023A CN 110940319 A CN110940319 A CN 110940319A
Authority
CN
China
Prior art keywords
height
image
limiting rod
vehicle
camera
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.)
Withdrawn
Application number
CN201910998023.9A
Other languages
Chinese (zh)
Inventor
刘展海
何伟
刘展航
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.)
Guangdong Interactive Electronic Network Media Co Ltd
Original Assignee
Guangdong Interactive Electronic Network Media Co Ltd
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 Guangdong Interactive Electronic Network Media Co Ltd filed Critical Guangdong Interactive Electronic Network Media Co Ltd
Priority to CN201910998023.9A priority Critical patent/CN110940319A/en
Publication of CN110940319A publication Critical patent/CN110940319A/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C11/00Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
    • G01C11/04Interpretation of pictures
    • G01C11/06Interpretation of pictures by comparison of two or more pictures of the same area

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

The invention discloses a height limit detection method and a device based on machine vision recognition, wherein the detection method comprises the following steps: adjusting the focal length of a camera of the camera mechanism, and calibrating an image coordinate system of the camera; collecting a vehicle image; acquiring height information of the target vehicle according to the longitudinal scale corresponding to the field depth scale mark of the target vehicle in the vehicle image; acquiring a height limiting rod image in front of the running target vehicle; according to the longitudinal scale corresponding to the field depth scale line of the height-limiting rod image where the target height-limiting rod is located in the height-limiting rod image, height information of the target height-limiting rod is obtained; comparing the height information of the target height limiting rod with the height information of the target vehicle; obtaining a height difference; the detection method can realize the detection of the overall height of the vehicle and the height of the height limiting rod, and judges whether the vehicle is ultrahigh or not according to the analysis result by comparing and detecting the heights of the vehicle and the height limiting rod.

Description

Height limit detection method and device based on machine vision recognition
Technical Field
The invention relates to the field of machine vision recognition, in particular to a height limit detection method and device based on machine vision recognition.
Background
With the improvement of the scientific and technological and economic levels of China, the types and the number of vehicles are more and more, and some roads are provided with height limiting frames or height limiting rods for protecting the roads and ensuring the personal safety of drivers; the height limiting frame is mainly used for door-shaped frameworks at the entrance of bridges, culverts and tunnels to play a role in protection. Due to the appearance of various vehicle types, road planning is also rapidly updated, the heights of height limiting rods are different, the height limiting rods are used for protecting the safety of passengers and cargos of a driver, but the heights of most height limiting rods are not erected according to standards, the driver bumps the height limiting rods under the conditions that the height of the vehicle is not clear, the vehicle is not driven intensively and the like, the safety of the driver, the passengers and the cargos is threatened, and serious economic loss is caused. Except for height limiting frames, roads such as bridges, culverts, tunnels and the like have certain height limitation, and in order to avoid accidents as much as possible, a height limiting detection method based on machine vision recognition is provided.
Disclosure of Invention
The invention aims to provide a height limit detection method and a height limit detection device based on machine vision recognition, the detection method can realize the height detection of the whole vehicle and the height detection of a height limit rod, the height of the detected vehicle and the height of the height limit rod are compared for analysis, and whether the height of the detected vehicle is ultrahigh or not is judged according to the analysis result; the device realizes monocular vision ranging through monocular acquisition of corresponding coordinates of a target object in an image coordinate system, so that the detection of the overall height of the vehicle and the height of the height limiting rod is realized, the safety of the vehicle is ensured, the structure is simple, and the operation speed is high.
The technical scheme is as follows:
a height limit detection method based on machine vision recognition comprises the following steps:
(11) adjusting the focal length of a camera mechanism, and calibrating an image coordinate system of the camera; the image coordinate system comprises longitudinal scale marks, transverse scale marks and field depth scale marks, wherein a lower left angular point of the image coordinate system is a coordinate origin, the coordinate origin is vertically upward and is a longitudinal scale mark, the coordinate origin is horizontally rightward and is a transverse scale mark, the coordinate origin is obliquely upward and is a field depth scale mark, and an included angle between the field depth scale mark and the transverse scale mark is 12 degrees;
(12) collecting a vehicle image;
(13) acquiring height information of the target vehicle according to the longitudinal scale corresponding to the field depth scale mark of the vehicle image where the target vehicle is located in the vehicle image;
(14) the camera shooting mechanism is installed in a target vehicle, and the camera shooting direction of the camera shooting mechanism faces to the front of the target vehicle in running;
(15) acquiring a height limiting rod image in front of the running target vehicle;
(16) acquiring height information of the target height limiting rod according to the longitudinal scale corresponding to the field depth scale mark of the height limiting rod image where the target height limiting rod is located in the height limiting rod image;
(17) comparing the height information of the target height limiting rod with the height information of the target vehicle;
(18) obtaining the height difference between the target vehicle and the target height limiting rod;
(19) and outputting the height difference.
The step (13) further comprises the steps of:
(131) preprocessing the vehicle image;
(132) converting the vehicle image into a vehicle image to be analyzed;
(133) the vehicle image to be analyzed is digitally described, and vehicle position information of a target vehicle in the vehicle image to be analyzed is completed;
(134) determining vehicle depth of field scales corresponding to vehicle depth of field scale lines of a vehicle image where the target vehicle is located according to the vehicle position information, and obtaining the distance from the target vehicle to a shooting point according to the vehicle depth of field scales;
(135) and determining the vehicle longitudinal scale information of the target vehicle corresponding to the vehicle depth of field scale according to the vehicle position information, so as to obtain the height information of the target vehicle.
When said step (131) is carried out, said pre-processing comprises at least an image smoothing process, an image segmentation process and/or a mathematical morphology process.
When the step (133) is implemented, the step of digitally describing the vehicle image to be analyzed is a step of mapping the vehicle image to be analyzed from a two-dimensional space to a three-dimensional space.
Said step (16) further comprises the steps of:
(161) preprocessing the height limiting rod image;
(162) converting the height-limiting rod image into a height-limiting rod image to be analyzed;
(163) digitally describing the height limiting rod image to be analyzed to complete the position information of the height limiting rod of the target height limiting rod in the height limiting rod image to be analyzed;
(164) determining a depth of field scale of the height-limiting rod corresponding to a depth of field scale line of the height-limiting rod image of the target height-limiting rod according to the position information of the height-limiting rod, and obtaining the distance from the target height-limiting rod to a shooting point according to the depth of field scale of the height-limiting rod;
(165) determining the longitudinal scale information of the target height limiting rod corresponding to the depth of field scale of the height limiting rod according to the position information of the height limiting rod, so as to obtain the height information of the target height limiting rod;
(166) and determining the transverse scale information of the target height limiting rod at the height limiting rod corresponding to the depth of field scale of the height limiting rod according to the position information of the height limiting rod, so as to obtain the width information of the target height limiting rod.
When the step (163) is implemented, the digital description of the height-limiting rod image to be analyzed is to map the height-limiting rod image to be analyzed from a two-dimensional space to a three-dimensional space.
The utility model provides a limit for height detection device based on machine vision discernment, includes mounting bracket, camera shooting mechanism passes through mounting bracket movable mounting is in the target vehicle, camera shooting mechanism's the direction of making a video recording the place ahead that the target vehicle travel, camera shooting mechanism includes the camera, the printing has the image coordinate system on the front end lens of camera, the image coordinate system includes vertical scale mark, horizontal scale mark, degree of depth scale mark, the lower left corner point of image coordinate system is the origin of coordinates, the origin of coordinates makes progress vertically and is vertical scale mark, origin of coordinates level is horizontal scale mark right, the origin of coordinates slant upwards is the degree of depth scale mark, degree of depth scale mark with the contained angle of horizontal scale mark is 12 degrees.
The camera shooting mechanism comprises a shell, a circuit board and a touch display screen, the camera is installed on the rear panel of the shell, the touch display screen is installed on the front panel of the shell, the circuit board is installed in the shell, and the camera and the touch display screen are respectively electrically connected with the circuit board.
The camera includes first camera, second camera, first camera, second camera are installed respectively on the back face of casing, first camera, second camera are vertical neatly arranged.
The circuit board comprises an image acquisition module, an image processing module and a distance calculation module, wherein the camera, the image acquisition module, the image processing module and the distance calculation module are electrically connected in sequence.
The camera shooting mechanism further comprises a power switch, a volume button and a loudspeaker, wherein the power switch and the volume button are respectively installed on the side face of the shell, the loudspeaker is installed in the shell, and the power switch, the volume button and the loudspeaker are respectively electrically connected with the circuit board.
The camera shooting mechanism further comprises a battery, the battery is installed in the shell, and the battery is electrically connected with the circuit board.
The mounting bracket includes the fixed bolster and is used for the centre gripping the clamping device of mechanism of making a video recording, clamping device includes the objective table and locates the tensile clamping mechanism of objective table top, clamping device's back rotate connect in the top of fixed bolster.
The fixed bolster includes base, bracing piece, the lower extreme of bracing piece with the base is connected, clamping device's back rotate connect in the upper end of bracing piece.
The mounting bracket further comprises a screw instrument, and the back surface of the clamping device is rotatably connected to the upper end of the supporting rod through the screw instrument.
The fixed support further comprises a direction switching key, the direction switching key is installed on the supporting rod, and the direction switching key is electrically connected with the screw instrument.
The following illustrates the advantages or principles of the invention:
1. the invention provides a height-limiting detection method based on machine vision identification, wherein an image secondary system is printed on a camera adopting the detection method, the image coordinate system comprises a longitudinal scale mark, a transverse scale mark and a depth-of-field scale mark, the distance marked by the depth-of-field scale mark corresponds to the distance from a shooting point to a target object, the distance marked by the longitudinal scale mark corresponds to the height of the target object, the distance marked by the transverse scale mark corresponds to the width of the target object, the method obtains the height, width and depth information of an object to be detected by a monocular vision corresponding point marking method, utilizes an image coordinate system to combine a projection model, solves the corresponding relation between the image coordinate system and an actual imaging angle, determines the corresponding coordinates of the target object in a world coordinate system and an image coordinate system, thereby obtaining the height, width and depth information of the object to be detected, has high operation speed, the real-time requirement in the vehicle running process is met; the detection method can realize the detection of the overall height of the vehicle and the height of the height limiting rod, and judges whether the vehicle is ultrahigh or not according to the analysis result by comparing and detecting the heights of the vehicle and the height limiting rod.
2. The position information of the target object is obtained by preprocessing a target object image, converting the target object image into a target object image to be analyzed, then digitally describing the target object image to be analyzed, describing the target object image pixels of the vehicle to be analyzed by adopting a two-dimensional matrix, representing the spatial positions of pixel points by rows and columns of the two-dimensional matrix, and mapping the target object image to be analyzed from the two-dimensional space to a three-dimensional space.
3. The invention provides a height limit detection device based on machine vision identification, which comprises a mounting frame and a camera mechanism, wherein the camera mechanism comprises a camera, an image coordinate system is printed on the camera, the image coordinate system comprises a longitudinal scale mark, a transverse scale mark and a depth of field scale mark, when the device is used, a user firstly adjusts the focal length of the camera mechanism, calibrates the image coordinate system of the camera, then uses the camera mechanism to collect a vehicle image, the vehicle image combines a projection model by using the image coordinate system to solve the corresponding relation between the image coordinate system and an actual imaging angle, determines the corresponding coordinates of a target vehicle in a world coordinate system and an image coordinate system so as to obtain the height information of the target vehicle, then, the user movably installs the camera mechanism in the target vehicle through the mounting frame, and enables the camera direction of the camera mechanism to face the front of the target vehicle, acquiring a height-limiting rod image in front of the target vehicle in running by using a camera mechanism, solving a corresponding relation between an image coordinate system and an actual imaging angle by using the height-limiting rod image and combining a projection model with the image coordinate system, determining corresponding coordinates of the target height-limiting rod in a world coordinate system and the image coordinate system so as to obtain height information of the target height-limiting rod, finally comparing the height information of the target height-limiting rod with the height information of the target vehicle so as to obtain a height difference between the target vehicle and the target height-limiting rod, and judging whether the target vehicle runs safely through the target height-limiting rod by using the height difference; the device realizes monocular vision ranging through monocular acquisition of corresponding coordinates of a target object in an image coordinate system, so that the detection of the overall height of the vehicle and the height of the height limiting rod is realized, the safety of the vehicle is ensured, the structure is simple, and the operation speed is high.
4. The camera shooting mechanism comprises a shell, a circuit board and a touch display screen, wherein the camera and the touch display screen are respectively arranged on the front surface and the back surface of the shell, so that the camera shooting operation of the camera shooting mechanism is facilitated.
5. The camera comprises the first camera and the second camera, and the two cameras are arranged, so that the calibration operation of an image coordinate system is facilitated.
6. The circuit board comprises an image acquisition module, an image processing module and a distance calculation module, wherein the image acquisition module is used for sending a shot single image to the image processing module; the image processing module is used for preprocessing the single image, converting the single image into an image to be analyzed and sending the image to the distance calculation module; after the distance calculation module carries out digital description on the image to be analyzed to complete the position information of the target object, the image coordinate system is utilized to combine with the projection model, the corresponding relation between the image coordinate system and the actual imaging angle is solved, and the corresponding coordinates of the target object in the world coordinate system and the image coordinate system are determined, so that the height, the width and the depth information of the target object are obtained.
7. The camera shooting mechanism also comprises a power switch, a volume button and a loudspeaker, wherein the power switch is used for turning on and off the camera shooting mechanism, the loudspeaker is arranged and used for carrying out voice broadcasting according to a judgment result and reminding a driver to make a judgment response in time, and the volume button is used for adjusting the volume of the loudspeaker.
8. The camera mechanism of the invention further comprises a battery, and the battery is arranged to provide power for the camera mechanism.
9. The mounting frame comprises a fixing support and a clamping device, wherein the clamping device is used for clamping the camera shooting mechanism, the fixing support is conveniently installed in a vehicle, the back surface of the clamping device is rotatably connected to the top end of the fixing support, and the horizontal adjustment of the camera shooting mechanism installed on the clamping device is facilitated.
10. The fixing support comprises a base and a supporting rod, wherein the base is arranged at the lower end of the fixing support, so that the fixing support is more convenient to mount in a vehicle.
11. The clamping device is rotatably connected to the upper end of the supporting rod through the screw gauge, and automatic balance of the camera shooting mechanism arranged on the clamping device is achieved.
12. The direction switching key is arranged on the fixed support, so that the clamping direction of the clamping device can be conveniently switched.
Drawings
Fig. 1 is a schematic plan view of an image coordinate system of a height limit detection device based on machine vision recognition according to an embodiment of the present invention.
Fig. 2 is a front view of a camera mechanism of a height limit detection device based on machine vision recognition according to an embodiment of the present invention.
Fig. 3 is a rear view internal structure diagram of the camera mechanism of the height limit detection device based on machine vision recognition according to the embodiment of the invention.
Fig. 4 is a side view of a camera mechanism of a height limit detection device based on machine vision recognition according to an embodiment of the present invention.
Fig. 5 is a front view of a mounting frame of a height limiting detection device based on machine vision recognition according to an embodiment of the present invention.
Fig. 6 is a front view of the height limit detection device mounting frame clamping mechanism based on machine vision recognition in the state of pulling apart according to the embodiment of the invention.
Fig. 7 is a side view of a mounting bracket of a height limiting detection device based on machine vision recognition according to an embodiment of the invention.
Description of reference numerals:
10. the device comprises a mounting rack, 11, a fixed support, 111, a base, 112, a support rod, 113, a direction switching key, 12, a clamping device, 121, an object stage, 122, a clamping mechanism, 13, a screw instrument, 20, a camera mechanism, 21, a camera, 211, a first camera, 212, a second camera, 22, an image coordinate system, 221, a longitudinal scale mark, 222, a transverse scale mark, 223, a depth of field scale mark, 224, a coordinate origin, 23, a shell, 24, a circuit board, 241, an image acquisition module, 242, an image processing module, 243, a distance calculation module, 25, a touch display screen, 26, a power switch, 27 and a volume key.
Detailed Description
The following provides a detailed description of embodiments of the invention.
A height limit detection method based on machine vision recognition comprises the following steps:
(11) adjusting the focal length of a camera 21 of the camera mechanism 20, and calibrating an image coordinate system 22 of the camera 21; as shown in fig. 1, the image coordinate system 22 includes a longitudinal scale mark 221, a transverse scale mark 222, and a depth of field scale mark 223, a lower left corner point of the image coordinate system 22 is a coordinate origin 224, the coordinate origin 224 is the longitudinal scale mark 221 vertically upward, the coordinate origin 224 is the transverse scale mark 222 horizontally rightward, the coordinate origin 224 is the depth of field scale mark 223 obliquely upward, and an included angle between the depth of field scale mark 223 and the transverse scale mark 222 is 12 degrees;
(12) collecting a vehicle image;
(13) acquiring height information of the target vehicle according to the longitudinal scale corresponding to the field depth scale mark 223 of the vehicle image where the target vehicle is located in the vehicle image;
(14) mounting the camera 20 in the target vehicle, and making the camera direction of the camera 20 face the front of the target vehicle;
(15) acquiring a height limiting rod image in front of the running target vehicle;
(16) acquiring height information of the target height limiting rod according to the longitudinal scale corresponding to the field depth scale mark 223 of the height limiting rod image where the target height limiting rod is located in the height limiting rod image;
(17) comparing the height information of the target height limiting rod with the height information of the target vehicle;
(18) obtaining the height difference between the target vehicle and the target height limiting rod;
(19) and outputting the height difference.
The invention provides a height limit detection method based on machine vision identification, wherein an image secondary system is printed on a camera 21 adopting the detection method, an image coordinate system 22 comprises a longitudinal scale mark 221, a transverse scale mark 222 and a depth of field scale mark 223, the distance marked by the depth of field scale mark 223 corresponds to the distance from a shooting point to a target object, the distance marked by the longitudinal scale mark 221 corresponds to the height of the target object, the distance marked by the transverse scale mark 222 corresponds to the width of the target object, the height, the width and the depth information of an object to be detected are obtained by a monocular vision corresponding point marking method, the corresponding relation between the image coordinate system 22 and an actual imaging angle is solved by utilizing the image coordinate system 22 in combination with a projection model, and the corresponding coordinates of the target object in a world coordinate system and the image coordinate system 22 are determined, so that the height, the width, the height, the depth and the depth of the object to be detected are obtained, The depth information and the operation speed are high, and the real-time requirement in the vehicle running process is met; the detection method can realize the detection of the overall height of the vehicle and the height of the height limiting rod, and judges whether the vehicle is ultrahigh or not according to the analysis result by comparing and detecting the heights of the vehicle and the height limiting rod.
Wherein, step (13) further comprises the following steps:
(131) preprocessing the vehicle image;
(132) converting the vehicle image into a vehicle image to be analyzed;
(133) the vehicle image to be analyzed is digitally described, and vehicle position information of the target vehicle in the vehicle image to be analyzed is completed;
(134) determining vehicle depth of field scales corresponding to vehicle depth of field scale lines 223 of the vehicle image of the target vehicle according to the vehicle position information, and obtaining the distance from the target vehicle to the shooting point according to the vehicle depth of field scales;
(135) and determining the longitudinal scale information of the target vehicle corresponding to the depth of field scale of the vehicle according to the position information of the vehicle, so as to obtain the height information of the target vehicle.
When step (131) is performed, the preprocessing includes at least image smoothing processing, image segmentation processing and/or mathematical morphology processing.
When the step (133) is implemented, the step of digitally describing the vehicle image to be analyzed is to map the vehicle image to be analyzed from a two-dimensional space to a three-dimensional space.
The step (16) further comprises the steps of:
(161) preprocessing the height limiting rod image;
(162) converting the height-limiting rod image into a height-limiting rod image to be analyzed;
(163) digitally describing the height-limiting rod image to be analyzed to complete the position information of the height-limiting rod of the target height-limiting rod in the height-limiting rod image to be analyzed;
(164) determining the depth of field scale of the height-limiting rod corresponding to the depth of field scale line 223 of the height-limiting rod of the image of the height-limiting rod where the target height-limiting rod is located according to the position information of the height-limiting rod, and obtaining the distance from the target height-limiting rod to a shooting point according to the depth of field scale of the height-limiting rod;
(165) determining the longitudinal scale information of the target height limiting rod in the height limiting rod corresponding to the depth of field scale of the height limiting rod according to the position information of the height limiting rod, so as to obtain the height information of the target height limiting rod;
(166) and determining the transverse scale information of the target height limiting rod at the height limiting rod corresponding to the field depth scale of the height limiting rod according to the position information of the height limiting rod, thereby obtaining the width information of the target height limiting rod.
When the step (163) is implemented, the step of digitally describing the height-limiting rod image to be analyzed is to map the height-limiting rod image to be analyzed from a two-dimensional space to a three-dimensional space.
The position information of the target object is obtained by preprocessing a target object image, converting the target object image into a target object image to be analyzed, then digitally describing the target object image to be analyzed, describing the target object image pixels of the vehicle to be analyzed by adopting a two-dimensional matrix, representing the spatial positions of pixel points by rows and columns of the two-dimensional matrix, and mapping the target object image to be analyzed from the two-dimensional space to a three-dimensional space.
As shown in fig. 1 to 7, a height limit detection device based on machine vision recognition includes an installation frame 10 and a camera mechanism 20, the camera mechanism 20 is movably installed in a target vehicle through the installation frame 10, a camera direction of the camera mechanism 20 faces a front direction of the target vehicle, the camera mechanism 20 includes a camera 21, an image coordinate system 22 is printed on a front lens of the camera 21, as shown in fig. 1, the image coordinate system 22 includes a longitudinal scale mark 221, a transverse scale mark 222, and a depth of field scale mark 223, a lower left corner of the image coordinate system 22 is a coordinate origin 224, the coordinate origin 224 is vertically upward the longitudinal scale mark 221, the coordinate origin 224 is horizontally rightward the transverse scale mark 222, the coordinate origin 224 is obliquely upward the depth of field scale mark 223, and an included angle between the depth of field scale mark 223 and the transverse scale mark 222 is 12 degrees.
The invention provides a height limit detection device based on machine vision identification, which comprises a mounting rack 10 and a camera mechanism 20, wherein the camera mechanism 20 comprises a camera 21, an image coordinate system 22 is printed on the camera 21, the image coordinate system 22 comprises a longitudinal scale mark 221, a transverse scale mark 222 and a depth of field scale mark 223, when the height limit detection device is used, a user firstly adjusts the focal length of the camera 21 of the camera mechanism 20, calibrates the image coordinate system 22 of the camera 21, then uses the camera mechanism 20 to collect a vehicle image, the vehicle image uses the image coordinate system 22 to be combined with a projection model, finds out the corresponding relation between the image coordinate system 22 and an actual imaging angle, determines the corresponding coordinates of a target vehicle in a world coordinate system and the image coordinate system 22 so as to obtain the height information of the target vehicle, then, the user movably mounts the camera mechanism 20 in the target vehicle through the mounting rack 10, and enables the camera direction of the camera mechanism 20 to face the front of the target vehicle, then, a camera mechanism 20 is used for acquiring a height-limiting rod image in front of the target vehicle in running, the height-limiting rod image is combined with a projection model by using an image coordinate system 22 to solve the corresponding relation between the image coordinate system 22 and an actual imaging angle, and the corresponding coordinates of the target height-limiting rod in a world coordinate system and the image coordinate system 22 are determined so as to obtain the height information of the target height-limiting rod, finally, the height information of the target height-limiting rod is compared with the height information of the target vehicle so as to obtain the height difference between the target vehicle and the target height-limiting rod, and the height difference is used for judging whether the target vehicle runs safely through the target height-limiting rod; the device realizes monocular vision ranging through monocular acquisition of corresponding coordinates of a target object in an image coordinate system 22, so that the detection of the overall height of the vehicle and the height of the height limiting rod is realized, the safety of the vehicle is ensured, the structure is simple, and the operation speed is high.
The camera mechanism 20 includes a housing 23, a circuit board 24, a touch display screen 25, a power switch 26, a volume button 27, a speaker, and a battery, the camera 21 is installed on a rear panel of the housing 23, the touch display screen 25 is installed on a front panel of the housing 23, the circuit board 24 is installed in the housing 23, and the camera 21 and the touch display screen 25 are electrically connected to the circuit board 24. The camera 21 and the touch display screen 25 are respectively arranged on the front and the back of the shell 23, so that the camera shooting operation of the camera shooting mechanism 20 is facilitated.
The camera 21 includes a first camera 211 and a second camera 212, the first camera 211 and the second camera 212 are respectively mounted on the rear panel of the housing 23, and the first camera 211 and the second camera 212 are vertically and neatly arranged. The arrangement of the two cameras 21 facilitates the calibration operation of the image coordinate system 22.
The circuit board 24 includes an image acquisition module 241, an image processing module 242, and a distance calculation module 243, and the camera 21, the image acquisition module 241, the image processing module 242, and the distance calculation module 243 are electrically connected in sequence. The image acquisition module 241 is configured to send the single shot image to the image processing module 242; the image processing module 242 preprocesses the single image, converts the single image into an image to be analyzed, and sends the image to the distance calculating module 243; after the distance calculation module 243 digitally describes the image to be analyzed to complete the position information of the target object, the image coordinate system 22 is used in combination with the projection model to solve the corresponding relationship between the image coordinate system 22 and the actual imaging angle, and the corresponding coordinates of the target object in the world coordinate system and the image coordinate system 22 are determined, so as to obtain the height, width and depth information of the target object.
The power switch 26 and the volume key 27 are respectively arranged on the side surface of the shell 23, the loudspeaker is arranged in the shell 23, and the power switch 26, the volume key 27 and the loudspeaker are respectively electrically connected with the circuit board 24. The power switch 26 is used for turning on and off the camera mechanism 20, the loudspeaker is arranged for voice broadcasting according to the judgment result, a driver can be timely reminded to make judgment response, and the volume key 27 is used for adjusting the volume of the loudspeaker.
The battery is mounted in the housing 23 and the battery is electrically connected to the circuit board 24. The battery is provided to supply power to the camera mechanism 20.
The mounting frame 10 includes a fixed bracket 11 and a clamping device 12 for clamping the camera mechanism 20, the clamping device 12 includes an object stage 121 and a stretchable clamping mechanism 122 disposed above the object stage 121, and a back surface of the clamping device 12 is rotatably connected to a top end of the fixed bracket 11. The fixed bolster 11 is conveniently installed in the vehicle, and the back of clamping device 12 rotates to be connected on the top of fixed bolster 11, the level adjustment of the mechanism of making a video recording 20 of conveniently installing on clamping device 12.
The fixing bracket 11 comprises a base 111 and a supporting rod 112, wherein the lower end of the supporting rod 112 is connected with the base 111, and the back of the clamping device 12 is rotatably connected with the upper end of the supporting rod 112. The base 111 is provided at the lower end of the fixing bracket 11, so that the fixing bracket 11 can be conveniently installed in a vehicle.
In another embodiment, the mounting bracket 10 of the present invention further comprises a screw 13, and the back surface of the clamping device 12 is rotatably connected to the upper end of the supporting rod 112 through the screw 13. The clamping device 12 is rotatably connected to the upper end of the support rod 112 through the screw 13, so that the automatic balance of the camera mechanism 20 mounted on the clamping device 12 is realized.
The fixing bracket 11 further includes a direction switch button 113, the direction switch button 113 is mounted on the supporting rod 112, and the direction switch button 113 is electrically connected to the screw instrument 13. The direction switching button 113 is disposed on the fixing bracket 11 to facilitate switching of the clamping direction of the clamping device 12.
The above are merely specific embodiments of the present invention, and the scope of the present invention is not limited thereby; any alterations and modifications without departing from the spirit of the invention are within the scope of the invention.

Claims (10)

1. A height limit detection method based on machine vision recognition is characterized by comprising the following steps:
(11) adjusting the focal length of a camera mechanism, and calibrating an image coordinate system of the camera; the image coordinate system comprises longitudinal scale marks, transverse scale marks and field depth scale marks, wherein a lower left angular point of the image coordinate system is a coordinate origin, the coordinate origin is vertically upward and is a longitudinal scale mark, the coordinate origin is horizontally rightward and is a transverse scale mark, the coordinate origin is obliquely upward and is a field depth scale mark, and an included angle between the field depth scale mark and the transverse scale mark is 12 degrees;
(12) collecting a vehicle image;
(13) acquiring height information of the target vehicle according to the longitudinal scale corresponding to the field depth scale mark of the vehicle image where the target vehicle is located in the vehicle image;
(14) the camera shooting mechanism is installed in a target vehicle, and the camera shooting direction of the camera shooting mechanism faces to the front of the target vehicle in running;
(15) acquiring a height limiting rod image in front of the running target vehicle;
(16) acquiring height information of the target height limiting rod according to the longitudinal scale corresponding to the field depth scale mark of the height limiting rod image where the target height limiting rod is located in the height limiting rod image;
(17) comparing the height information of the target height limiting rod with the height information of the target vehicle;
(18) obtaining the height difference between the target vehicle and the target height limiting rod;
(19) and outputting the height difference.
2. The machine-vision-recognition-based height limit detection method according to claim 1, wherein the step (13) further comprises the steps of:
(131) preprocessing the vehicle image;
(132) converting the vehicle image into a vehicle image to be analyzed;
(133) the vehicle image to be analyzed is digitally described, and vehicle position information of a target vehicle in the vehicle image to be analyzed is completed;
(134) determining vehicle depth of field scales corresponding to vehicle depth of field scale lines of a vehicle image where the target vehicle is located according to the vehicle position information, and obtaining the distance from the target vehicle to a shooting point according to the vehicle depth of field scales;
(135) and determining the vehicle longitudinal scale information of the target vehicle corresponding to the vehicle depth of field scale according to the vehicle position information, so as to obtain the height information of the target vehicle.
3. The machine-vision-recognition-based height-limiting detection method of claim 2, wherein the preprocessing at least comprises an image smoothing process, an image segmentation process and/or a mathematical morphology process when the step (131) is performed.
4. The machine-vision-recognition-based height-limiting detection method of claim 2, wherein the step (133) is performed by digitally describing the vehicle image to be analyzed by mapping the vehicle image to be analyzed from a two-dimensional space to a three-dimensional space.
5. The machine-vision-recognition-based height limit detection method according to any one of claims 1 to 4, wherein the step (16) further comprises the steps of:
(161) preprocessing the height limiting rod image;
(162) converting the height-limiting rod image into a height-limiting rod image to be analyzed;
(163) digitally describing the height limiting rod image to be analyzed to complete the position information of the height limiting rod of the target height limiting rod in the height limiting rod image to be analyzed;
(164) determining a depth of field scale of the height-limiting rod corresponding to a depth of field scale line of the height-limiting rod image of the target height-limiting rod according to the position information of the height-limiting rod, and obtaining the distance from the target height-limiting rod to a shooting point according to the depth of field scale of the height-limiting rod;
(165) determining the longitudinal scale information of the target height limiting rod corresponding to the depth of field scale of the height limiting rod according to the position information of the height limiting rod, so as to obtain the height information of the target height limiting rod;
(166) and determining the transverse scale information of the target height limiting rod at the height limiting rod corresponding to the depth of field scale of the height limiting rod according to the position information of the height limiting rod, so as to obtain the width information of the target height limiting rod.
6. The machine-vision-recognition-based height-limiting detection method of claim 5, wherein the step (163) is performed by digitally describing the height-limiting rod image to be analyzed by mapping the height-limiting rod image to be analyzed from a two-dimensional space to a three-dimensional space.
7. The utility model provides a limit for height detection device based on machine vision discernment, a serial communication port, including mounting bracket, camera shooting mechanism passes through mounting bracket movable mounting is in the target vehicle, camera shooting mechanism's the direction of making a video recording the place ahead that the target vehicle travel, camera shooting mechanism includes the camera, the printing has the image coordinate system on the front end lens of camera, the image coordinate system includes vertical scale mark, horizontal scale mark, degree of depth scale mark, the lower left corner point of image coordinate system is the origin of coordinates, the origin of coordinates makes progress vertically and is vertical scale mark, origin of coordinates level is horizontal scale mark right, the origin of coordinates slant upwards is the degree of depth scale mark, the degree of depth scale mark with the contained angle of horizontal scale mark is 12 degrees.
8. The height limit detection device based on machine vision identification as claimed in claim 7, wherein the camera mechanism includes a housing, a circuit board, and a touch display screen, the camera is mounted on a rear panel of the housing, the touch display screen is mounted on a front panel of the housing, the circuit board is mounted in the housing, and the camera and the touch display screen are electrically connected to the circuit board respectively; the camera comprises a first camera and a second camera, the first camera and the second camera are respectively arranged on the rear plate surface of the shell, and the first camera and the second camera are longitudinally and orderly arranged; the circuit board comprises an image acquisition module, an image processing module and a distance calculation module, and the camera, the image acquisition module, the image processing module and the distance calculation module are electrically connected in sequence; the camera shooting mechanism further comprises a power switch, a volume button and a loudspeaker, wherein the power switch and the volume button are respectively arranged on the side surface of the shell, the loudspeaker is arranged in the shell, and the power switch, the volume button and the loudspeaker are respectively and electrically connected with the circuit board; the camera shooting mechanism further comprises a battery, the battery is installed in the shell, and the battery is electrically connected with the circuit board.
9. The height limit detection device based on machine vision identification as claimed in claim 7, wherein the mounting bracket comprises a fixed bracket and a clamping device for clamping the camera mechanism, the clamping device comprises an object stage and a stretchable clamping mechanism arranged above the object stage, and the back surface of the clamping device is rotatably connected to the top end of the fixed bracket; the fixed bolster includes base, bracing piece, the lower extreme of bracing piece with the base is connected, clamping device's back rotate connect in the upper end of bracing piece.
10. The height limit detection device based on machine vision identification as claimed in claim 7, characterized in that, the mounting bracket further comprises a screw instrument, and the back of the clamping device is rotatably connected to the upper end of the support rod through the screw instrument. The fixed support further comprises a direction switching key, the direction switching key is installed on the supporting rod, and the direction switching key is electrically connected with the screw instrument.
CN201910998023.9A 2019-10-21 2019-10-21 Height limit detection method and device based on machine vision recognition Withdrawn CN110940319A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910998023.9A CN110940319A (en) 2019-10-21 2019-10-21 Height limit detection method and device based on machine vision recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910998023.9A CN110940319A (en) 2019-10-21 2019-10-21 Height limit detection method and device based on machine vision recognition

Publications (1)

Publication Number Publication Date
CN110940319A true CN110940319A (en) 2020-03-31

Family

ID=69905941

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910998023.9A Withdrawn CN110940319A (en) 2019-10-21 2019-10-21 Height limit detection method and device based on machine vision recognition

Country Status (1)

Country Link
CN (1) CN110940319A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111539279A (en) * 2020-04-14 2020-08-14 杭州鸿泉物联网技术股份有限公司 Road height limit height detection method, device, equipment and storage medium
CN112014845A (en) * 2020-08-28 2020-12-01 安徽江淮汽车集团股份有限公司 Vehicle obstacle positioning method, device, equipment and storage medium
CN112287879A (en) * 2020-11-18 2021-01-29 河南以太坊贸易有限公司 Road height limiting method, device and equipment and road height limiting control system
CN113851000A (en) * 2021-09-10 2021-12-28 泰州蝶金软件有限公司 Command analysis system based on cloud computing

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111539279A (en) * 2020-04-14 2020-08-14 杭州鸿泉物联网技术股份有限公司 Road height limit height detection method, device, equipment and storage medium
CN111539279B (en) * 2020-04-14 2023-08-15 杭州鸿泉物联网技术股份有限公司 Road height limit detection method, device, equipment and storage medium
CN112014845A (en) * 2020-08-28 2020-12-01 安徽江淮汽车集团股份有限公司 Vehicle obstacle positioning method, device, equipment and storage medium
CN112014845B (en) * 2020-08-28 2024-01-30 安徽江淮汽车集团股份有限公司 Vehicle obstacle positioning method, device, equipment and storage medium
CN112287879A (en) * 2020-11-18 2021-01-29 河南以太坊贸易有限公司 Road height limiting method, device and equipment and road height limiting control system
CN113851000A (en) * 2021-09-10 2021-12-28 泰州蝶金软件有限公司 Command analysis system based on cloud computing

Similar Documents

Publication Publication Date Title
CN110940319A (en) Height limit detection method and device based on machine vision recognition
JP5615441B2 (en) Image processing apparatus and image processing method
CN101808236B (en) Vehicle periphery displaying apparatus
US20170140542A1 (en) Vehicular image processing apparatus and vehicular image processing system
US20060114320A1 (en) Position detecting apparatus and method of correcting data therein
US20080181591A1 (en) Camera posture estimation device, vehicle, and camera posture estimation method
US7616806B2 (en) Position detecting apparatus and method of correcting data therein
JP6938107B1 (en) How to measure road surface damage
CN109883433B (en) Vehicle positioning method in structured environment based on 360-degree panoramic view
JP2008085710A (en) Driving support system
JP2012105090A (en) Parking support apparatus
CN113359140A (en) Method for aligning a vehicle service system relative to a vehicle
US7567688B2 (en) Apparatus for and method of extracting image
CN110562140A (en) Multi-camera implementation method and system of transparent A column
JP5915268B2 (en) Parameter calculation method, information processing apparatus, and program
CN101719041A (en) Device and method for installing and debugging camera
US9990556B2 (en) Image processing apparatus, imaging apparatus, movable object, program, and region setting method
KR101697229B1 (en) Automatic calibration apparatus based on lane information for the vehicle image registration and the method thereof
JP4867709B2 (en) Display distortion measuring apparatus and display distortion measuring method
JP2008210084A (en) Road marking recognition device
CN112484743B (en) Vehicle-mounted HUD fusion live-action navigation display method and system thereof
CN114155300A (en) Projection effect detection method and device for vehicle-mounted HUD system
CN104590115A (en) Driving safety auxiliary system and method
CN210570655U (en) Height limit detection device based on machine vision recognition
JP4752319B2 (en) Image recognition apparatus and image recognition method

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20200331

WW01 Invention patent application withdrawn after publication