CN112308899B - Trailer angle identification method and device - Google Patents
Trailer angle identification method and device Download PDFInfo
- Publication number
- CN112308899B CN112308899B CN202011240105.6A CN202011240105A CN112308899B CN 112308899 B CN112308899 B CN 112308899B CN 202011240105 A CN202011240105 A CN 202011240105A CN 112308899 B CN112308899 B CN 112308899B
- Authority
- CN
- China
- Prior art keywords
- dimensional code
- trailer
- image
- angle
- position information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000001514 detection method Methods 0.000 claims abstract description 47
- 238000007781 pre-processing Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 9
- 238000009434 installation Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000012512 characterization method Methods 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1408—Methods for optical code recognition the method being specifically adapted for the type of code
- G06K7/1417—2D bar codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1439—Methods for optical code recognition including a method step for retrieval of the optical code
- G06K7/1443—Methods for optical code recognition including a method step for retrieval of the optical code locating of the code in an image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Toxicology (AREA)
- Electromagnetism (AREA)
- Health & Medical Sciences (AREA)
- Geometry (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a trailer angle identification method and device, which are used for acquiring a trailer image acquired by a camera, identifying position information of each two-dimensional code image in a world coordinate system and a corresponding two-dimensional code ID from the trailer image, determining that the corresponding two-dimensional code image is a left end two-dimensional code image or a right end two-dimensional code image of a trailer head according to each two-dimensional code ID, calculating a target vector angle between the two-dimensional code IDs by using an inverse trigonometric function based on the position information of the left end two-dimensional code image and the position information of the corresponding two-dimensional code ID and the corresponding two-dimensional code ID, and finding a target trailer angle corresponding to the target vector angle from the pre-stored corresponding relation between the vector angle and the trailer angle. According to the invention, the trailer angle is determined according to the position information of the two-dimensional code images arranged at the two ends of the trailer head, so that the hardware cost is low, the complicated operation of arranging the detection device is omitted, and the identification efficiency of the trailer angle is improved.
Description
Technical Field
The invention relates to the technical field of automatic driving of trucks, in particular to a trailer angle identification method and device.
Background
Trucks, also known as trucks, generally referred to as vans, refer to vehicles that are primarily used to transport cargo, and sometimes also vehicles that may tow other vehicles, belonging to the category of commercial vehicles. The truck mainly comprises a truck head and a trailer, and when the truck turns at an intersection, the truck head and the trailer form an included angle, which is called a trailer angle, and the angle theta is shown in fig. 1. When the rotation control and the speed control are carried out on the truck, the swing posture of the trailer is obtained according to the change condition of the included angle between the truck head and the trailer, so that the trailer is prevented from being damaged due to collision with the road edge or other non-road obstacles. Thus, trailer angle is important for truck automatic turn control, and trailer angle identification is essential during truck automatic driving.
The conventional solution is to install a detection device on a truck, for example, to install an electronic sensor when identifying the angle of the trailer, and to arrange a wire harness on the trailer, and to consider the power supply problem of the electronic sensor. For another example, an electronically controlled angle detection device is installed, in which case an arcuate saddle or the like would need to be installed. Therefore, the detection mode of the installation detection device is adopted in the prior art, so that not only is the hardware cost high, but also the complicated operation of the installation detection device is increased, and the recognition efficiency of the trailer angle is low.
Disclosure of Invention
In view of the above, the invention discloses a trailer angle identification method and device, which are used for solving the problems of high hardware cost and low trailer angle identification efficiency in the traditional scheme.
A trailer angle identification method, comprising:
acquiring a trailer image acquired by a camera, wherein the trailer image comprises two-dimensional code images stuck to two ends of a trailer head;
identifying the position information of each two-dimensional code image in a world coordinate system and the corresponding two-dimensional code ID from the trailer image;
determining a left two-dimensional code image and a right two-dimensional code image according to each two-dimensional code ID;
Calculating a target vector angle between the two-dimensional code IDs by using an inverse trigonometric function based on the position information of the left-end two-dimensional code image and the corresponding two-dimensional code ID and the position information of the right-end two-dimensional code image and the corresponding two-dimensional code ID;
and searching a target trailer angle corresponding to the target vector angle from the pre-stored corresponding relation between the vector angle and the trailer angle.
Optionally, the identifying, from the trailer image, the position information of each two-dimensional code image in the world coordinate system and the corresponding two-dimensional code ID includes:
Inputting the trailer image into a pre-established object detection model, and predicting frame information of the area where all the two-dimensional code images in the trailer image are located;
Extracting a plurality of two-dimensional code areas from the trailer image according to the frame information;
And identifying each two-dimensional code area to obtain the position information of the two-dimensional code image in each two-dimensional code area in a world coordinate system and the corresponding two-dimensional code ID.
Optionally, the determining the left two-dimensional code image and the right two-dimensional code image according to each two-dimensional code ID includes:
determining that the corresponding two-dimensional code image is positioned in a left-end two-dimensional code area or a right-end two-dimensional code area of the trailer head according to each two-dimensional code ID;
And determining the two-dimensional code image with the highest recognition confidence from the left two-dimensional code area as the left two-dimensional code image, and determining the two-dimensional code image with the highest recognition confidence from the right two-dimensional code area as the right two-dimensional code image based on the recognition confidence of each two-dimensional code image.
Optionally, the identifying, from the trailer image, the position information of each two-dimensional code image in the world coordinate system and the corresponding two-dimensional code ID further includes:
performing image preprocessing on the trailer image to enhance the feature representation of the two-dimensional code area and obtaining an enhanced feature two-dimensional code;
acquiring pixel point coordinates and two-dimension code IDs of four corner points of a two-dimension code box in the enhanced feature two-dimension code;
and acquiring the position information of the two-dimensional code by adopting a gesture evaluation PNP algorithm to the pixel point coordinates of the four corner points of the two-dimensional code block.
Optionally, the acquiring the coordinates of pixel points and the two-dimensional code ID of four corner points of the two-dimensional code block in the enhanced feature two-dimensional code specifically includes:
According to the mapping relation between the image pixels before distortion and the image pixels after distortion, carrying out distortion correction processing on the enhancement feature two-dimensional code to obtain a non-distortion two-dimensional code;
and adopting a ApriTag detection algorithm of aruco libraries for the undistorted two-dimensional code region to obtain pixel point coordinates and two-dimensional code IDs of four corner points of the two-dimensional code block.
Optionally, the image preprocessing includes: high pass filtering, white balancing, and contrast enhancement.
Optionally, two polygonal devices are respectively stuck at two ends of the trailer head, and two-dimensional code images are respectively arranged in front of and on the side surfaces of the polygonal devices.
A trailer angle identification apparatus comprising:
The acquisition unit is used for acquiring a trailer image acquired by the camera, wherein the trailer image comprises two-dimensional code images stuck to two ends of a trailer head;
The identification unit is used for identifying the position information of each two-dimensional code image in the world coordinate system and the corresponding two-dimensional code ID from the trailer image;
The determining unit is used for determining a left two-dimensional code image and a right two-dimensional code image according to each two-dimensional code ID;
The computing unit is used for computing a target vector angle between the two-dimensional code IDs by using an inverse trigonometric function based on the position information of the left-end two-dimensional code image and the corresponding two-dimensional code ID and the position information of the right-end two-dimensional code image and the corresponding two-dimensional code ID;
And the searching unit is used for searching the target trailer angle corresponding to the target vector angle from the prestored corresponding relation between the vector angle and the trailer angle.
Optionally, the identification unit is specifically configured to:
Inputting the trailer image into a pre-established object detection model, and predicting frame information of the area where all the two-dimensional code images in the trailer image are located;
Extracting a plurality of two-dimensional code areas from the trailer image according to the frame information;
And identifying each two-dimensional code area to obtain the position information of the two-dimensional code image in each two-dimensional code area in a world coordinate system and the corresponding two-dimensional code ID.
Optionally, the determining unit is specifically configured to:
determining that the corresponding two-dimensional code image is positioned in a left-end two-dimensional code area or a right-end two-dimensional code area of the trailer head according to each two-dimensional code ID;
And determining the two-dimensional code image with the highest recognition confidence from the left two-dimensional code area as the left two-dimensional code image, and determining the two-dimensional code image with the highest recognition confidence from the right two-dimensional code area as the right two-dimensional code image based on the recognition confidence of each two-dimensional code image.
According to the technical scheme, the trailer angle identification method and device are disclosed, the trailer images acquired by the camera are acquired, the trailer images comprise two-dimensional code images stuck to two ends of a trailer head, the position information of each two-dimensional code image in a world coordinate system and the corresponding two-dimensional code ID are identified from the trailer images, the corresponding two-dimensional code image is determined to be a left-end two-dimensional code image or a right-end two-dimensional code image of the trailer head according to each two-dimensional code ID, the target vector angle between the two-dimensional code IDs is calculated by using an inverse trigonometric function based on the position information of the left-end two-dimensional code image and the position information of the corresponding two-dimensional code ID and the corresponding two-dimensional code ID, and the target trailer angle corresponding to the target vector angle is found from the corresponding relation between the prestored vector angle and the trailer angle. According to the invention, the trailer angle is determined according to the position information of the two-dimensional code images arranged at the two ends of the trailer head, so that the hardware cost is low, the complicated operation of arranging the detection device is omitted, and the identification efficiency of the trailer angle is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the disclosed drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic view of a truck turn.
FIG. 2 is a top view of a three-state overlay of an inner axle type trailer according to an embodiment of the present invention;
FIG. 3 is a top view of three states of an outer shaft type trailer according to an embodiment of the present invention;
FIG. 4 is a flowchart of a trailer angle identification method according to an embodiment of the present invention;
Fig. 5 is an installation side view of a camera and a two-dimensional code image according to an embodiment of the present invention;
fig. 6 is a schematic diagram illustrating pasting of a two-dimensional code image on a trailer according to an embodiment of the present invention;
Fig. 7 is a top view of a camera and a two-dimensional code image according to an embodiment of the present invention;
fig. 8 is a schematic diagram of a two-dimensional code corner detection result according to an embodiment of the present invention;
fig. 9 is a schematic structural diagram of a trailer angle recognition device according to an embodiment of the present invention.
Detailed Description
In order to facilitate understanding of the technical problems to be solved by the present invention, the following describes an application scenario of the present invention, which is specifically as follows:
The inner axle trailer refers to a state that a rotating shaft of the truck is arranged on the inner side of the trailer, see a top view of an overlapping diagram of three states of the inner axle type trailer shown in fig. 2 in detail, a solid frame is a state that an included angle theta formed between a truck head and the trailer is 0 DEG, two virtual frames respectively represent the states that the included angle theta formed between the truck head and the trailer is 90 DEG and-90 DEG, a rectangle in fig. 2 represents a pasting position of a two-dimensional code (ApriTag) image, two-dimensional code images are respectively pasted on two ends of the truck head, the two-dimensional code images are pasted facing the camera, and the vector angle of a connecting line between the two-dimensional code images is estimated by detecting the coordinate positions of the two-dimensional code images, so that the angle of the trailer is calculated.
The outer axle trailer refers to a state that a rotating shaft of the truck is arranged on the outer side of the trailer, and a detailed view is shown in a top view of three states of the outer axle type trailer shown in fig. 3, a solid frame is a state that an included angle theta formed between a truck head and the trailer is 0 DEG, two virtual frames respectively represent the states that the included angle theta formed between the truck head and the trailer is 90 DEG and-90 DEG, a rectangle in fig. 2 represents a pasting position of a two-dimensional code image, and the two-dimensional code image is pasted facing the camera.
Compared with the outer axle trailer, the inner axle trailer needs to consider the condition that the included angle between the truck head and the trailer is large angle. When the included angle between the truck head and the trailer is large, the visual angle of the camera and the two-dimensional code image stuck at the front end of the truck are large, so that the condition that the two-dimensional code image acquired by the camera is not complete and the left and right positions are exchanged easily occurs, and the detection of the two-dimensional code image is invalid.
It should be noted that the present invention mainly solves the situation of the inner axle trailer, and at the same time, the trailer angle identification method disclosed by the present invention is also suitable for the situation of the outer axle trailer.
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The embodiment of the invention discloses a trailer angle identification method and device, which are used for acquiring trailer images acquired by a camera, wherein the trailer images comprise two-dimensional code images stuck at two ends of a trailer head, the position information of each two-dimensional code image in a world coordinate system and a corresponding two-dimensional code ID are identified from the trailer images, the corresponding two-dimensional code image is determined to be a left-end two-dimensional code image or a right-end two-dimensional code image of the trailer head according to each two-dimensional code ID, the target vector angle between the two-dimensional code IDs is calculated by utilizing an inverse trigonometric function based on the position information of the left-end two-dimensional code image and the position information of the corresponding two-dimensional code ID and the corresponding two-dimensional code ID, and the target trailer angle corresponding to the target vector angle is found from the corresponding relation between the prestored vector angle and the trailer angle. According to the invention, the trailer angle is determined according to the position information of the two-dimensional code images arranged at the two ends of the trailer head, so that the hardware cost is low, the complicated operation of arranging the detection device is omitted, and the identification efficiency of the trailer angle is improved.
Referring to fig. 4, a flow chart of a trailer angle identification method disclosed in an embodiment of the invention includes:
S101, acquiring a trailer image acquired by a camera;
The trailer image comprises two-dimensional code images stuck at two ends of the trailer head.
Specifically, a polygonal device is respectively installed at two ends of the trailer head, and the trailer image comprises two-dimension code images adhered to the front and side surfaces of the two polygonal devices.
Specifically, referring to the installation side view of the camera and the two-dimensional code image shown in fig. 5, the camera is installed at the rear of the vehicle head, the two-dimensional code image is pasted at the front end of the trailer, and the pasting mode of the two-dimensional code image is shown in fig. 6.
Referring to a schematic diagram of pasting a two-dimensional code image on a trailer shown in fig. 6, when an angle of an included angle formed between the trailer and a truck head is close to 90 degrees or-90 degrees, a normal vector of a two-dimensional code pasted on the trailer and a central visual angle of a camera are in a large angle, so that detection failure of the two-dimensional code is caused. For the large angle between the trailer and the truck head, as shown in fig. 6, the invention designs two polygonal devices which are respectively arranged at two ends of the trailer head, and a two-dimensional code image is respectively stuck on the front face (the face opposite to the head) and the two side faces of each polygonal device, and six two-dimensional code images are stuck on the two polygonal devices, so that the trailer angle detection of the trailer in the range of [ -90 degrees, 90 degrees ] is realized.
Alternatively, the polygonal means may be a cube, cuboid or trapezoid, or the like.
Step S102, identifying the position information of each two-dimensional code image in a world coordinate system and the corresponding two-dimensional code ID from the trailer image;
Specifically, inputting a trailer image into a pre-established object detection model, and predicting frame information of an area where all two-dimensional code images in the trailer image are located; extracting a plurality of two-dimensional code areas from the trailer image according to the frame information; and identifying each two-dimensional code area to obtain the position information of the two-dimensional code image in the world coordinate system and the corresponding two-dimensional code ID in each two-dimensional code area.
Wherein, object detection model is: and training the YOLO (YouOnly LookOnce) model or the SSD (Single Shot MultiBoxDetector) model by taking the trailer image set containing the two-dimensional code image as a training set.
The object detection model is a trained deep learning model and has the capability of detecting a two-dimensional code image from a trailer image. By inputting the trailer image into the object detection model, the frame information of the area where the two-dimensional code image is located in the trailer image is predicted, the frame information is the frame of the area where the two-dimensional code is located, which is automatically identified in the trailer image, and at least one two-dimensional code image exists in a two-dimensional space surrounded by the frame.
Step S103, determining a left two-dimensional code image and a right two-dimensional code image according to each two-dimensional code ID;
It should be noted that, in this embodiment, each two-dimensional code image has a unique corresponding two-dimensional code ID, and each two-dimensional code ID has a left area identifier and a right area identifier of the trailer head, so that the two-dimensional code image corresponding to each two-dimensional code ID can be determined to be the left end two-dimensional code image or the right end two-dimensional code image of the trailer head according to each two-dimensional code ID. This is because, when the angle of the trailer is large, the conventional identification method may cause that the positions of the two-dimensional codes attached to the left and right ends of the trailer on the trailer image are opposite to the physical positions, and therefore, the two-dimensional code ID includes information about whether the two-dimensional code is specifically located at the left end or the right end of the trailer, and the two-dimensional code ID is identified to determine the positions of the two-dimensional codes
Specifically, determining that a corresponding two-dimensional code image is positioned in a left-end two-dimensional code area or a right-end two-dimensional code area of the trailer head according to each two-dimensional code ID; the left two-dimensional code area and the right two-dimensional code area both contain at least one two-dimensional code image.
Based on the identification confidence of each two-dimensional code image, determining the two-dimensional code image with the highest identification confidence from a left two-dimensional code area as the left two-dimensional code image, and determining the two-dimensional code image with the highest identification confidence from a right two-dimensional code area as the right two-dimensional code image.
In the identification process, certain two-dimensional code incomplete identification phenomenon exists, so that the integrity of the identifiable image determines the identification confidence, and the more complete the two-dimensional code image is identified, the higher the identification confidence is.
Step S104, calculating a target vector angle between two-dimensional code IDs by using an inverse trigonometric function based on the position information of the left-end two-dimensional code image in the world coordinate system and the corresponding two-dimensional code ID and the position information of the right-end two-dimensional code image in the world coordinate system and the corresponding two-dimensional code ID;
Specifically, referring to the top view of the installation of the camera and the two-dimensional code image shown in fig. 7, assuming that coordinates of two-dimensional code IDs of a left-end two-dimensional code image and a right-end two-dimensional code image, which are respectively installed at two ends of a trailer head, are (x, y) and (x 1, y 1), respectively, a calculation formula of a target vector angle between the two-dimensional code IDs is as follows:
step S105, finding a target trailer angle corresponding to the target vector angle from the pre-stored corresponding relation between the vector angle and the trailer angle.
Different two-dimensional code ID combinations can be detected under different trailer angles, and the relation between the different two-dimensional code ID combinations and the trailer angles can be calibrated in advance, so that the trailer angles can be calculated according to the vector angles among the different two-dimensional code IDs and the corresponding relation among the trailer angles.
It should be noted that, if the deflection angle of the single two-dimensional code is directly identified by adopting the algorithm, and the trailer angle is determined based on the deflection angle of the single two-dimensional code, in actual use, the method is greatly affected by environmental factors and is unstable.
The inventor of the invention discovers after research that the estimation of the pose (position and deflection angle) of the two-dimensional code by the algorithm is determined by four corner points of the two-dimensional code, and in a real environment, the prediction of the four corner points of the two-dimensional code is fluctuated due to the fact that the illumination intensity is unstable, so that the prediction of the two-dimensional code and the deflection angle is influenced. Experiments prove that the fluctuation of four corner points of the two-dimensional code has larger influence on the deflection angle of a single two-dimensional code, and the angle prediction requirement of high precision (for example, 1 degree precision) cannot be met, but the influence on the position prediction is less than within the range of 0.5 cm; therefore, when two-dimensional code images are attached to two ends of a trailer and the distance is about 1.8m, the angle fluctuation error brought is about 0.16 degrees, and the error requirement is far smaller than 1 degree, namely, the relative deflection angle of the two-dimensional codes is less influenced by the fluctuation of the position of a single two-dimensional code, and the further the two-dimensional codes are away, the higher the accuracy of angle calculation is. Therefore, the trailer angle is calculated by adopting the position information of the two-dimension codes, so that the trailer angle is more stable, reliable and accurate compared with the calculation of the trailer angle based on the deflection angle information of the two-dimension codes, the obtained trailer angle has stronger robustness, and the method is suitable for angle detection with high precision requirements.
In summary, according to the trailer angle identification method disclosed by the invention, the trailer image acquired by the camera is acquired, the trailer image comprises two-dimensional code images stuck at two ends of the trailer head, the position information of each two-dimensional code image in a world coordinate system and the corresponding two-dimensional code ID are identified from the trailer image, the corresponding two-dimensional code image is determined to be a left-end two-dimensional code image or a right-end two-dimensional code image of the trailer head according to each two-dimensional code ID, the target vector angle between the two-dimensional code IDs is calculated by using an inverse trigonometric function based on the position information of the left-end two-dimensional code image and the position information of the corresponding two-dimensional code ID and the corresponding two-dimensional code ID, and the target trailer angle corresponding to the target vector angle is found from the corresponding relation between the prestored vector angle and the trailer angle. According to the invention, the trailer angle is determined according to the position information of the two-dimensional code images arranged at the two ends of the trailer head, so that the hardware cost is low, the complicated operation of arranging the detection device is omitted, and the identification efficiency of the trailer angle is improved.
It is particularly required to explain that in the actual situation, due to the direct sunlight condition, the pixel difference between the black area and the white area of the two-dimensional code becomes small, so that the identification of four corner points of the two-dimensional code is inaccurate or invalid; the detection problem in the scene can be effectively avoided by adding measures such as filtering and contrast, however, false detection is increased if the full image of the trailer is filtered and compared. Therefore, the two-dimensional code area is detected from the trailer image, and then the characteristic characterization of the two-dimensional code is enhanced through image preprocessing (preprocessing: high-pass filtering- > white balance- > contrast enhancement) on the two-dimensional code area, so that the reliability and stability of two-dimensional code detection and corner detection are improved.
Therefore, to further optimize the above embodiment, identifying the position information of each two-dimensional code image in the world coordinate system and the corresponding two-dimensional code ID from the trailer image may further include:
performing image preprocessing on the trailer image to enhance the feature representation of the two-dimension code area and obtaining an enhanced feature two-dimension code; acquiring pixel point coordinates and two-dimension code IDs of four corner points of a two-dimension code box in the enhanced feature two-dimension code; and (3) carrying out attitude evaluation on pixel point coordinates of four corner points of the two-dimensional code box to obtain the position information of the two-dimensional code by adopting a PNP algorithm.
Wherein, the image preprocessing includes: high pass filtering- > white balancing- > contrast enhancement.
Specifically, in practical application, high-pass filtering is performed on a two-dimensional code area in a trailer image to obtain a first two-dimensional code area, and then white balance processing is performed on the first two-dimensional code area to obtain a second two-dimensional code area; and carrying out contrast enhancement treatment on the second two-dimensional code region to obtain an enhanced characteristic two-dimensional code region.
Optionally, the contrast enhancement processing may be performed on the two-dimensional code region using the saturate _cast function of OpenCV.
According to the invention, the reliability and stability of trailer angle identification are improved by preprocessing the two-dimensional code image to enhance the characteristic representation of the two-dimensional code region.
In order to further optimize the above embodiment, the process of obtaining the coordinates of the pixel points of the four corner points of the two-dimensional code block in the enhanced feature two-dimensional code and the two-dimensional code ID may specifically include:
according to the mapping relation between the image pixels before distortion and the image pixels after distortion, carrying out distortion correction processing on the enhancement feature two-dimensional code to obtain a non-distortion two-dimensional code;
And adopting a ApriTag detection algorithm of aruco libraries for the undistorted two-dimensional code region to obtain pixel point coordinates of four corner points of the two-dimensional code box and the two-dimensional code ID.
In the two-dimensional code corner detection result schematic diagram shown in fig. 8, four corners of the black square frame are four corners of the two-dimensional code square frame.
The mapping relation between the image pixels before distortion and the image pixels after distortion exists in the form of a mapping table, and the mapping table is obtained by using an OpenCV-based de-distortion function initUndistortRectifyMap.
It should be noted that, in practical application, the actual size of the two-dimensional code is known, and according to the pixel Point coordinates of the four corner points of the two-dimensional code frame, a PNP (PESPECTIVE-N-Point) algorithm is evaluated by using a gesture, so that the position information (x, y, z) of the two-dimensional code in the world coordinate system can be obtained, and the vertical height z is negligible because the vertical height z is not related to the angle calculation.
In summary, according to the trailer angle identification method disclosed by the invention, the trailer image acquired by the camera is acquired, the trailer image comprises two-dimensional code images stuck to two ends of the trailer head, the characteristic representation of the enhanced two-dimensional code area is carried out on the trailer image, the enhanced characteristic two-dimensional code is obtained, the pixel point coordinates and the two-dimensional code ID of four corner points of a two-dimensional code box in the enhanced characteristic two-dimensional code are obtained, the position information of the two-dimensional code in a world coordinate system is obtained by adopting a gesture evaluation PNP algorithm on the pixel point coordinates of the four corner points of the two-dimensional code box, the two-dimensional code ID of each two-dimensional code image is identified from the trailer image, the corresponding two-dimensional code image is determined as the left-end two-dimensional code image or the right-end two-dimensional code image of the trailer head according to each two-dimensional code ID, the position information of the left-end two-dimensional code image in the world coordinate system, the corresponding two-dimensional code ID and the corresponding two-dimensional code ID of the right-end two-dimensional code image in the world coordinate system are calculated by utilizing an inverse trigonometric function, and the target vector angle between the two-dimensional code IDs is found from the pre-stored corresponding relation of the vector angle and the trailer angle, and the target vector angle corresponding to the world angle is found. According to the invention, the trailer angle is determined according to the position information of the two-dimensional code images arranged at the two ends of the trailer head, so that the hardware cost is low, the complicated operation of arranging the detection device is omitted, and the identification efficiency of the trailer angle is improved.
In addition, the problem of detection failure caused by the fact that the visual angle of the camera and the two-dimensional code image stuck to the front end of the truck are in a large angle is solved by sticking two-dimensional code images to the two ends of the trailer head respectively, reliability and stability of trailer angle identification are improved by preprocessing the two-dimensional code images to enhance the characteristic representation of the two-dimensional code area, and compared with a traditional scheme of installing the detection device, the hardware cost of the polygonal device is low.
Corresponding to the embodiment of the method, the invention also discloses a trailer angle identification device.
Referring to fig. 9, a schematic structural diagram of a trailer angle identification apparatus according to an embodiment of the present invention is disclosed, where the apparatus includes:
an acquisition unit 201, configured to acquire a trailer image acquired by a camera;
The trailer image comprises two-dimensional code images stuck at two ends of the trailer head.
Specifically, a polygonal device is respectively installed at two ends of the trailer head, and the trailer image comprises two-dimension code images adhered to the front and side surfaces of the two polygonal devices.
The installation schematic diagrams of the camera and the two-dimensional code image can be seen in fig. 5 and 6.
An identifying unit 202, configured to identify, from the trailer images, position information of each two-dimensional code image in a world coordinate system and a corresponding two-dimensional code ID;
the identification unit 202 may specifically be configured to:
Inputting a trailer image into a pre-established object detection model, and predicting frame information of an area where all two-dimensional code images in the trailer image are located; extracting a plurality of two-dimensional code areas from the trailer image according to the frame information; and identifying each two-dimensional code area to obtain the position information of the two-dimensional code image in the world coordinate system and the corresponding two-dimensional code ID in each two-dimensional code area.
Wherein, object detection model is: and training the YOLO (YouOnly LookOnce) model or the SSD (Single Shot MultiBoxDetector) model by taking the trailer image set containing the two-dimensional code image as a training set.
The object detection model is a trained deep learning model and has the capability of detecting a two-dimensional code image from a trailer image. And (3) inputting the trailer image into the object detection model, and predicting the frame information of the area where the two-dimensional code image is located in the trailer image.
A determining unit 203, configured to determine a left-end two-dimensional code image and a right-end two-dimensional code image according to each two-dimensional code ID;
It should be noted that, in this embodiment, each two-dimensional code image has a unique corresponding two-dimensional code ID, and each two-dimensional code ID has a left area identifier and a right area identifier of the trailer head, so that the two-dimensional code image corresponding to each two-dimensional code ID can be determined to be the left end two-dimensional code image or the right end two-dimensional code image of the trailer head according to each two-dimensional code ID.
The determining unit 203 may specifically be configured to:
determining that the corresponding two-dimensional code image is positioned in a left-end two-dimensional code area or a right-end two-dimensional code area of the trailer head according to each two-dimensional code ID;
Based on the identification confidence of each two-dimensional code image, determining the two-dimensional code image with the highest identification confidence from the left two-dimensional code area as a left two-dimensional code image, and determining the two-dimensional code image with the highest identification confidence from the right two-dimensional code area as a right two-dimensional code image.
A calculating unit 204, configured to calculate, based on the position information of the left two-dimensional code image and the corresponding two-dimensional code ID, and the position information of the right two-dimensional code image and the corresponding two-dimensional code ID, a target vector angle between the two-dimensional code IDs using an inverse trigonometric function;
and the searching unit 205 is configured to search for a target trailer angle corresponding to the target vector angle from the pre-stored correspondence between vector angles and trailer angles.
Different ID two-dimensional code combinations can be detected under different trailer angles, and the relation between the different two-dimensional code ID combinations and the trailer angles can be calibrated in advance, so that the trailer angles can be calculated according to the vector angles among the different two-dimensional code IDs and the corresponding relation among the trailer angles.
It should be noted that, if the deflection angle of the single two-dimensional code is directly identified by adopting the algorithm, and the trailer angle is determined based on the deflection angle of the single two-dimensional code, in actual use, the method is greatly affected by environmental factors and is unstable.
In summary, the trailer angle identification device disclosed by the invention acquires the trailer image acquired by the camera, the trailer image comprises two-dimensional code images stuck at two ends of the trailer head, the position information of each two-dimensional code image in a world coordinate system and the corresponding two-dimensional code ID are identified from the trailer image, the corresponding two-dimensional code image is determined to be a left-end two-dimensional code image or a right-end two-dimensional code image of the trailer head according to each two-dimensional code ID, the target vector angle between the two-dimensional code IDs is calculated by utilizing an inverse trigonometric function based on the position information of the left-end two-dimensional code image and the position information of the corresponding two-dimensional code ID and the corresponding two-dimensional code ID, and the target trailer angle corresponding to the target vector angle is found from the corresponding relation between the prestored vector angle and the trailer angle. According to the invention, the trailer angle is determined according to the position information of the two-dimensional code images arranged at the two ends of the trailer head, so that the hardware cost is low, the complicated operation of arranging the detection device is omitted, and the identification efficiency of the trailer angle is improved.
It is particularly required to explain that in the actual situation, due to the direct sunlight condition, the pixel difference between the black area and the white area of the two-dimensional code becomes small, so that the identification of four corner points of the two-dimensional code is inaccurate or invalid; the detection problem in the scene can be effectively avoided by adding measures such as filtering and contrast, however, false detection is increased if the full image of the trailer is filtered and compared. Therefore, the two-dimensional code area is detected from the trailer image, and then the characteristic characterization of the two-dimensional code is enhanced through image preprocessing (preprocessing: high-pass filtering- > white balance- > contrast enhancement) on the two-dimensional code area, so that the reliability and stability of two-dimensional code detection and corner detection are improved.
Thus, to further optimize the above embodiment, the identification unit 202 may in particular also be used for:
performing image preprocessing on the trailer image to enhance the feature representation of the two-dimension code area and obtaining an enhanced feature two-dimension code; acquiring pixel point coordinates and two-dimension code IDs of four corner points of a two-dimension code box in the enhanced feature two-dimension code; and (3) carrying out attitude evaluation on pixel point coordinates of four corner points of the two-dimensional code box to obtain the position information of the two-dimensional code by adopting a PNP algorithm.
Wherein, the image preprocessing includes: high pass filtering- > white balancing- > contrast enhancement.
Specifically, in practical application, high-pass filtering is performed on a two-dimensional code area in a trailer image to obtain a first two-dimensional code area, and then white balance processing is performed on the first two-dimensional code area to obtain a second two-dimensional code area; and carrying out contrast enhancement processing on the second two-dimensional code region to obtain a two-dimensional code region with enhanced characteristic characterization.
Optionally, the contrast enhancement processing may be performed on the two-dimensional code region using the saturate _cast function of OpenCV.
According to the invention, the reliability and stability of trailer angle identification are improved by preprocessing the two-dimensional code image to enhance the characteristic representation of the two-dimensional code region.
To further optimize the above embodiment, the process of acquiring the coordinates of the pixel points of the four corner points of the two-dimensional code block in the enhanced feature two-dimensional code and the two-dimensional code ID by the identifying unit 202 may specifically include:
according to the mapping relation between the image pixels before distortion and the image pixels after distortion, carrying out distortion correction processing on the enhancement feature two-dimensional code to obtain a non-distortion two-dimensional code;
And adopting a ApriTag detection algorithm of aruco libraries for the undistorted two-dimensional code region to obtain pixel point coordinates of four corner points of the two-dimensional code box and the two-dimensional code ID.
In the two-dimensional code corner detection result schematic diagram shown in fig. 8, four corners of the black square frame are four corners of the two-dimensional code square frame.
In summary, the trailer angle identification device disclosed by the invention acquires the trailer image acquired by the camera, the trailer image comprises two-dimensional code images stuck at two ends of the trailer head, the trailer image is subjected to image preprocessing to enhance the feature representation of the two-dimensional code area, the enhanced feature two-dimensional code is obtained, the pixel point coordinates and the two-dimensional code ID of four corner points of a two-dimensional code box in the enhanced feature two-dimensional code are obtained, the position information of the two-dimensional code in a world coordinate system is obtained by adopting a gesture evaluation PNP algorithm on the pixel point coordinates of the four corner points of the two-dimensional code box, the two-dimensional code ID of each two-dimensional code image is identified from the trailer image, the corresponding two-dimensional code image is determined as the left-end two-dimensional code image or the right-end two-dimensional code image of the trailer head according to each two-dimensional code ID, the position information of the left-end two-dimensional code image in the world coordinate system, the corresponding two-dimensional code ID and the corresponding two-dimensional code ID of the right-end two-dimensional code image in the world coordinate system are calculated by utilizing an inverse trigonometric function, and the target vector angle between the two-dimensional code ID and the target vector angle corresponding to the trailer angle is found from the pre-stored corresponding relation of the vector angle and the target vector angle. According to the invention, the trailer angle is determined according to the position information of the two-dimensional code images arranged at the two ends of the trailer head, so that the hardware cost is low, the complicated operation of arranging the detection device is omitted, and the identification efficiency of the trailer angle is improved.
In addition, the problem of detection failure caused by the fact that the visual angle of the camera and the two-dimensional code image stuck to the front end of the truck are in a large angle is solved by sticking two-dimensional code images to the two ends of the trailer head respectively, reliability and stability of trailer angle identification are improved by preprocessing the two-dimensional code images to enhance the characteristic representation of the two-dimensional code area, and compared with a traditional scheme of installing the detection device, the hardware cost of the polygonal device is low.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A trailer angle identification method, comprising:
acquiring a trailer image acquired by a camera, wherein the trailer image comprises two-dimensional code images stuck to two ends of a trailer head;
identifying the position information of each two-dimensional code image in a world coordinate system and the corresponding two-dimensional code ID from the trailer image;
Determining that the corresponding two-dimensional code image is positioned in a left-end two-dimensional code area or a right-end two-dimensional code area of the trailer head according to each two-dimensional code ID; determining a left two-dimensional code image in the left two-dimensional code area and a right two-dimensional code image in the right two-dimensional code area based on the identification confidence of each two-dimensional code image;
Calculating a target vector angle between the two-dimensional code IDs by using an inverse trigonometric function based on the position information of the left-end two-dimensional code image and the corresponding two-dimensional code ID and the position information of the right-end two-dimensional code image and the corresponding two-dimensional code ID;
and searching a target trailer angle corresponding to the target vector angle from the pre-stored corresponding relation between the vector angle and the trailer angle.
2. The trailer angle identification method as claimed in claim 1, wherein said identifying the position information of each two-dimensional code image in the world coordinate system and the corresponding two-dimensional code ID from the trailer image comprises:
Inputting the trailer image into a pre-established object detection model, and predicting frame information of the area where all the two-dimensional code images in the trailer image are located;
Extracting a plurality of two-dimensional code areas from the trailer image according to the frame information;
And identifying each two-dimensional code area to obtain the position information of the two-dimensional code image in each two-dimensional code area in a world coordinate system and the corresponding two-dimensional code ID.
3. The trailer angle identification method as claimed in claim 1, wherein,
And determining the two-dimensional code image with the highest recognition confidence from the left two-dimensional code area as the left two-dimensional code image, and determining the two-dimensional code image with the highest recognition confidence from the right two-dimensional code area as the right two-dimensional code image based on the recognition confidence of each two-dimensional code image.
4. The trailer angle identification method as claimed in claim 2, wherein said identifying the position information of each two-dimensional code image in the world coordinate system and the corresponding two-dimensional code ID from the trailer image further comprises:
performing image preprocessing on the trailer image to enhance the feature representation of the two-dimensional code area and obtaining an enhanced feature two-dimensional code;
acquiring pixel point coordinates and two-dimension code IDs of four corner points of a two-dimension code box in the enhanced feature two-dimension code;
and acquiring the position information of the two-dimensional code by adopting a gesture evaluation PNP algorithm to the pixel point coordinates of the four corner points of the two-dimensional code block.
5. The trailer angle identification method as claimed in claim 4, wherein said obtaining pixel coordinates and two-dimensional code IDs of four corner points of a two-dimensional code block in the enhanced feature two-dimensional code specifically comprises:
According to the mapping relation between the image pixels before distortion and the image pixels after distortion, carrying out distortion correction processing on the enhancement feature two-dimensional code to obtain a non-distortion two-dimensional code;
And adopting a ApriTag detection algorithm of aruco libraries for the undistorted two-dimensional code region to obtain pixel point coordinates of four corner points of the two-dimensional code block and two-dimensional code IDs.
6. The trailer angle identification method as claimed in claim 4, wherein said image preprocessing comprises: high pass filtering, white balancing, and contrast enhancement.
7. The trailer angle identification method as claimed in claim 1, wherein two polygonal devices are respectively attached to two ends of the trailer head, and two-dimensional code images are respectively arranged on the front and side surfaces of the polygonal devices.
8. A trailer angle identification device, comprising:
The acquisition unit is used for acquiring a trailer image acquired by the camera, wherein the trailer image comprises two-dimensional code images stuck to two ends of a trailer head;
The identification unit is used for identifying the position information of each two-dimensional code image in the world coordinate system and the corresponding two-dimensional code ID from the trailer image;
The determining unit is used for determining that the corresponding two-dimensional code image is positioned in the left-end two-dimensional code area or the right-end two-dimensional code area of the trailer head according to each two-dimensional code ID; determining a left two-dimensional code image in the left two-dimensional code area and a right two-dimensional code image in the right two-dimensional code area based on the identification confidence of each two-dimensional code image;
The computing unit is used for computing a target vector angle between the two-dimensional code IDs by using an inverse trigonometric function based on the position information of the left-end two-dimensional code image and the corresponding two-dimensional code ID and the position information of the right-end two-dimensional code image and the corresponding two-dimensional code ID;
And the searching unit is used for searching the target trailer angle corresponding to the target vector angle from the prestored corresponding relation between the vector angle and the trailer angle.
9. The trailer angle identification device as claimed in claim 8, wherein the identification unit is specifically adapted to:
Inputting the trailer image into a pre-established object detection model, and predicting frame information of the area where all the two-dimensional code images in the trailer image are located;
Extracting a plurality of two-dimensional code areas from the trailer image according to the frame information;
And identifying each two-dimensional code area to obtain the position information of the two-dimensional code image in each two-dimensional code area in a world coordinate system and the corresponding two-dimensional code ID.
10. The trailer angle identification device as claimed in claim 8, wherein the determination unit is specifically adapted to:
And determining the two-dimensional code image with the highest recognition confidence from the left two-dimensional code area as the left two-dimensional code image, and determining the two-dimensional code image with the highest recognition confidence from the right two-dimensional code area as the right two-dimensional code image based on the recognition confidence of each two-dimensional code image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011240105.6A CN112308899B (en) | 2020-11-09 | 2020-11-09 | Trailer angle identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011240105.6A CN112308899B (en) | 2020-11-09 | 2020-11-09 | Trailer angle identification method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112308899A CN112308899A (en) | 2021-02-02 |
| CN112308899B true CN112308899B (en) | 2024-05-07 |
Family
ID=74325463
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011240105.6A Active CN112308899B (en) | 2020-11-09 | 2020-11-09 | Trailer angle identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112308899B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113489963B (en) * | 2021-07-08 | 2024-02-23 | 宁波宝贝第一母婴用品有限公司 | Cart installation guiding method and device |
| CN116202424B (en) * | 2023-04-28 | 2023-08-04 | 深圳一清创新科技有限公司 | Vehicle body area detection method, tractor and tractor obstacle avoidance system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105243366A (en) * | 2015-10-10 | 2016-01-13 | 北京微尘嘉业科技有限公司 | Two-dimensional code based vehicle positioning method |
| CN109872372A (en) * | 2019-03-07 | 2019-06-11 | 山东大学 | A kind of small-sized quadruped robot overall Vision localization method and system |
| CN110046529A (en) * | 2018-12-11 | 2019-07-23 | 阿里巴巴集团控股有限公司 | Two-dimensional code identification method, device and equipment |
| CN110276808A (en) * | 2019-06-11 | 2019-09-24 | 合肥工业大学 | A kind of method of one camera combination two dimensional code measurement glass plate unevenness |
| WO2019179233A1 (en) * | 2018-03-23 | 2019-09-26 | 阿里巴巴集团控股有限公司 | Qr code identifying and reading method and apparatus, and device |
| EP3566903A1 (en) * | 2018-05-09 | 2019-11-13 | Neusoft Corporation | Method and apparatus for vehicle position detection |
-
2020
- 2020-11-09 CN CN202011240105.6A patent/CN112308899B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105243366A (en) * | 2015-10-10 | 2016-01-13 | 北京微尘嘉业科技有限公司 | Two-dimensional code based vehicle positioning method |
| WO2019179233A1 (en) * | 2018-03-23 | 2019-09-26 | 阿里巴巴集团控股有限公司 | Qr code identifying and reading method and apparatus, and device |
| EP3566903A1 (en) * | 2018-05-09 | 2019-11-13 | Neusoft Corporation | Method and apparatus for vehicle position detection |
| CN110046529A (en) * | 2018-12-11 | 2019-07-23 | 阿里巴巴集团控股有限公司 | Two-dimensional code identification method, device and equipment |
| CN109872372A (en) * | 2019-03-07 | 2019-06-11 | 山东大学 | A kind of small-sized quadruped robot overall Vision localization method and system |
| CN110276808A (en) * | 2019-06-11 | 2019-09-24 | 合肥工业大学 | A kind of method of one camera combination two dimensional code measurement glass plate unevenness |
Non-Patent Citations (1)
| Title |
|---|
| 基于二维码的单目视觉测距移动机器人定位研究;罗高;袁亮;;组合机床与自动化加工技术(第10期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112308899A (en) | 2021-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109435942B (en) | Information fusion-based parking space line and parking space recognition method and device | |
| CN107577988B (en) | Method, device, storage medium and program product for realizing side vehicle positioning | |
| KR101734057B1 (en) | Drive-by calibration from static targets | |
| CN101910781B (en) | Moving state estimation device | |
| JP4973736B2 (en) | Road marking recognition device, road marking recognition method, and road marking recognition program | |
| CN106952308B (en) | Method and system for determining position of moving object | |
| US11762071B2 (en) | Multi-mode multi-sensor calibration | |
| CN109871732B (en) | Parking grid identification system and method thereof | |
| US11117517B2 (en) | Camera based auto drive auto charge | |
| CN112308899B (en) | Trailer angle identification method and device | |
| CN110176038B (en) | Method, system and storage medium for calibrating camera of vehicle | |
| CN111507327B (en) | Target detection method and device | |
| CN112257539B (en) | Method, system and storage medium for detecting position relationship between vehicle and lane line | |
| JP2018519696A5 (en) | ||
| CN110827361B (en) | Camera group calibration method and device based on global calibration frame | |
| KR20080024776A (en) | Method and apparatus for recognizing parking slot marking by using hough transformation and parking assist system using same | |
| CN111572633B (en) | Steering angle detection method, device and system | |
| CN112598922A (en) | Parking space detection method, device, equipment and storage medium | |
| CN112529935B (en) | Target vehicle motion state identification method and device | |
| US20150063630A1 (en) | Apparatus and method for detecting obstacle | |
| Hedi et al. | A system for vehicle surround view | |
| CN114943836A (en) | Trailer angle detection method and device and electronic equipment | |
| CN113147746A (en) | Method and device for detecting ramp parking space | |
| KR101714131B1 (en) | Device and method for recognizing parking stall | |
| CN112767425A (en) | Parking space detection method and device based on vision |
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 |