CN112241988B - Image processing method, device, computer equipment and storage medium - Google Patents
Image processing method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN112241988B CN112241988B CN201910778472.2A CN201910778472A CN112241988B CN 112241988 B CN112241988 B CN 112241988B CN 201910778472 A CN201910778472 A CN 201910778472A CN 112241988 B CN112241988 B CN 112241988B
- Authority
- CN
- China
- Prior art keywords
- mark
- image
- coordinate system
- point cloud
- small
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 24
- 238000004590 computer program Methods 0.000 claims description 15
- 238000003384 imaging method Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 4
- 239000003550 marker Substances 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The application relates to an image processing method, an image processing device, a computer device and a storage medium. The method comprises the following steps: acquiring an image corresponding to a preset calibration target, and identifying marks in the image to obtain a large mark and a small mark in the marks; further establishing a topological coordinate system of the image according to the large identifier; and determining the image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark. By adopting the method, the external parameter calibration between the three-dimensional laser radar and the monocular camera can be accurately carried out on the specific calibration target.
Description
Technical Field
The present application relates to the field of external parameter calibration technologies, and in particular, to an image processing method, an image processing device, a computer device, and a storage medium.
Background
Calibration is a fundamental problem in the fields of computer vision and robots, and mainly refers to whether the accuracy (precision) of a used instrument meets the standard or not by using a standard metering instrument, and is generally used for instruments with higher precision, and calibration can be considered as calibration.
In a mobile platform equipped with two types of sensors, namely a laser radar and a monocular camera, when an autonomous vehicle is in the same state, calibration between the laser radar and the monocular camera is required. According to the characteristics of two types of sensors to be calibrated, namely a laser radar and a monocular camera, a corresponding calibration device is required to be designed so as to achieve the aim of external parameter calibration. The data of the calibrating device acquired by the laser radar and the monocular camera are respectively a three-dimensional point cloud and a two-dimensional image, and effective information extraction is needed to obtain parameters for external parameter calibration.
At present, the external parameter calibration precision between the three-dimensional laser radar and the monocular camera is lower.
Disclosure of Invention
In view of the foregoing, it is desirable to provide an image processing method, apparatus, computer device, and storage medium capable of improving the accuracy of external parameter calibration.
An image processing method, the method comprising:
Acquiring an image corresponding to a preset calibration target;
Identifying the marks in the image to obtain a large mark and a small mark in the marks;
establishing a topological coordinate system of the image according to the large identifier;
And determining the image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark.
In one embodiment, the method comprises:
Acquiring a two-dimensional image corresponding to a preset calibration target;
identifying the marks in the two-dimensional image to obtain a large mark and a small mark in the marks;
Establishing a topological coordinate system of the two-dimensional image according to the large identifier;
and determining two-dimensional image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark.
In one embodiment, the identifying the identifier in the two-dimensional image to obtain the large identifier and the small identifier in the identifier includes:
Classifying the marks according to the imaging areas of the marks in the two-dimensional image to obtain a large mark and a small mark.
In one embodiment, the establishing a topological coordinate system of the two-dimensional image according to the large identifier includes:
establishing a basic coordinate system by taking any one angle of the two-dimensional image as a circle center so as to determine the image coordinate of the large mark;
and establishing a topological coordinate system of the two-dimensional image according to the topological relation between the large-identification image coordinates.
In one embodiment, the determining, according to the topological coordinate system, the large identifier and the small identifier, the two-dimensional image coordinate corresponding to the preset calibration target includes:
determining the image coordinates of the small mark according to the topological coordinate system and the image coordinates of the large mark;
And obtaining the two-dimensional image coordinates corresponding to the preset calibration target according to the image coordinates of the large mark and the image coordinates of the small mark.
In one embodiment, the determining the image coordinates of the small logo according to the topological coordinate system and the image coordinates of the large logo includes:
And searching the small mark along the horizontal axis and the vertical axis of the topological coordinate system by taking the pixel distance between the adjacent large mark image coordinates as an interval, and determining the small mark image coordinates.
In one embodiment, the method comprises:
Acquiring a three-dimensional point cloud image corresponding to a preset calibration target;
Identifying the marks in the three-dimensional point cloud image to obtain a large mark and a small mark in the marks;
establishing a topological coordinate system of the three-dimensional point cloud image according to the large identifier;
and determining the three-dimensional point cloud image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark.
In one embodiment, the identifying the identifier in the three-dimensional point cloud image, and obtaining the large identifier and the small identifier in the identifier includes:
Acquiring an initial mark and an initial mark curvature;
determining target marks and target mark curvatures according to the initial marks and the initial mark curvatures;
And determining a large mark and a small mark according to the target mark and the curvature of the target mark.
In one embodiment, the obtaining the initial identity and the initial identity curvature comprises:
performing plane fitting on the three-dimensional point cloud to obtain the initial calibration target;
calculating the direction vector of the initial calibration target, and determining the direction of the initial calibration target;
and acquiring a three-dimensional point cloud of which the direction of the initial calibration target is in a first preset range, and calculating the curvature of an initial mark corresponding to the three-dimensional point cloud in the first preset range, wherein a graph formed by the three-dimensional point cloud of which the direction of the initial calibration target is in the first preset range is the initial mark.
In one embodiment, the determining the target mark and the target mark curvature according to the initial mark and the initial mark curvature includes:
rotating the initial mark to obtain a rotated mark;
Searching a three-dimensional point cloud in a second preset range of the rotated marks, and if the number of the three-dimensional point cloud and the rotated marks reaches a preset threshold, taking the initial mark as a target mark, wherein the curvature of the initial mark is the curvature of the target mark.
In one embodiment, the establishing the topological coordinate system of the three-dimensional point cloud image according to the large identifier includes:
Acquiring the large identifier;
Determining a three-dimensional point cloud corresponding to the large identifier according to the position relation of the large identifier;
Fitting the three-dimensional point cloud corresponding to the large identifier to obtain the three-dimensional point cloud image coordinates of the large identifier;
and establishing a topological coordinate system of the three-dimensional point cloud image according to the topological relation between the coordinates of the three-dimensional point cloud image with the large mark.
In one embodiment, the establishing the topological coordinate system of the three-dimensional point cloud image according to the topological relation between the coordinates of the three-dimensional point cloud image with the large identifier includes:
Obtaining a residual identifier;
Screening the residual identifiers to obtain screened identifiers;
and fitting the screened marks to determine small marks.
An image processing apparatus, the apparatus comprising:
The image acquisition module is used for acquiring an image corresponding to a preset calibration target;
the identification module is used for identifying the identifications in the images to obtain large identifications and small identifications in the identifications;
the coordinate system establishing module is used for establishing a topological coordinate system of the image according to the large identifier;
and the image coordinate determining module is used for determining the image coordinate corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark.
A vehicle comprising an image processing apparatus as described above.
A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method according to any of the preceding claims when executing the computer program.
A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any of the preceding claims.
The image processing method, the device, the computer equipment and the storage medium are used for obtaining the large mark and the small mark in the marks by obtaining the images corresponding to the preset calibration targets and identifying the marks in the images; further establishing a topological coordinate system of the image according to the large identifier; and determining the image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark. By the method, the external parameter calibration between the three-dimensional laser radar and the monocular camera can be accurately performed on the specific calibration target.
Drawings
FIG. 1 is a diagram of an application environment for an image processing method in one embodiment;
FIG. 2 is a flow chart of an image processing method in one embodiment;
FIG. 3 (a) is a schematic front view of a predetermined calibration target;
FIG. 3 (b) is a schematic view of the preset calibration target rotated 30 about the x-axis;
FIG. 3 (c) is a schematic view of the preset calibration target rotated 30 about the z-axis;
FIG. 3 (d) is a schematic view of the preset calibration target rotated 30 about the y-axis;
FIG. 4 is a flow chart of a two-dimensional image processing method in one embodiment;
FIG. 5 is a schematic representation of a pre-set calibration target topology in one embodiment;
FIG. 6 is a flow diagram of a three-dimensional point cloud image processing method in one embodiment;
FIG. 7 is a block diagram showing the structure of an image processing apparatus in one embodiment;
FIG. 8 is a schematic diagram of an application of an image processing apparatus in a vehicle in one embodiment;
Fig. 9 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The image processing method provided by the application can be applied to an application environment shown in fig. 1. Wherein the terminal 50 communicates with the server 60 via a network. The terminal 50 obtains the image corresponding to the preset calibration target, and transmits the image corresponding to the preset calibration target to the server 60 through the network. The server 60 identifies the identifiers in the image to obtain a large identifier and a small identifier in the identifiers; further establishing a topological coordinate system of the image according to the large identifier; and determining the image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark. The terminal 50 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices, and the server 60 may be implemented as a stand-alone server or a server cluster composed of a plurality of servers.
In one embodiment, as shown in fig. 2, an image processing method is provided, and the method is applied to the server 104 in fig. 1 for illustration, and includes the following steps:
step S10: acquiring an image corresponding to a preset calibration target;
step S20: identifying the marks in the image to obtain a large mark and a small mark in the marks;
step S30: establishing a topological coordinate system of the image according to the large identifier;
Step S40: and determining the image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark.
In the steps S10-S40, the preset calibration target in the application consists of a calibration plate and a mark, the mark is fixed on the calibration plate through a screw, and the shape, the arrangement mode and the size of the mark are set according to the requirement. The large mark and the small mark are only different in size, and the spherical mark is adopted by the application, so the large mark can be called as a large mark ball, and the small mark can be called as a small mark ball.
Taking fig. 3 as an example, the preset calibration target consists of a calibration plate and identification balls, wherein the identification hemispheres are fixed on the calibration plate through screws, the target size is 1600mm multiplied by 1400mm, the surface is provided with 42 identification balls in total of 7 rows and 6 columns, the diameters of 4 large identification balls are 200mm, the diameters of the rest identification balls are 100mm, the hemispheres are uniformly distributed, and the center distances of the balls are 200mm. Fig. 3 (a) is a front view of the preset calibration target, fig. 3 (b) is rotated 30 ° around the x-axis, fig. 3 (c) is rotated 30 ° around the z-axis, and fig. 3 (d) is rotated 30 ° around the y-axis.
Further, the images corresponding to the preset calibration targets can be two-dimensional images, three-dimensional images and other multidimensional images, and the laser radar and the monocular camera are adopted to respectively collect preset calibration target data so as to obtain a three-dimensional point cloud image and a two-dimensional image. And respectively processing the two-dimensional image and the three-dimensional point cloud image, and determining the image coordinates corresponding to the preset calibration target.
The image processing method, the device, the computer equipment and the storage medium are used for obtaining the large mark and the small mark in the marks by obtaining the images corresponding to the preset calibration targets and identifying the marks in the images; further establishing a topological coordinate system of the image according to the large identifier; and determining the image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark. By the method, the external parameter calibration between the three-dimensional laser radar and the monocular camera can be accurately performed on the specific calibration target.
In one embodiment, as shown in fig. 4, the method includes:
Step S11: acquiring a two-dimensional image corresponding to a preset calibration target;
step S21: identifying the marks in the two-dimensional image to obtain a large mark and a small mark in the marks;
step S31: establishing a topological coordinate system of the two-dimensional image according to the large identifier;
Step S41: and determining two-dimensional image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark.
In steps S11-S41, the two-dimensional image is processed to determine the image coordinates corresponding to the preset calibration target.
According to the consistency of the directions of the spheres, the spheres shot by the cameras through different angles are round; meanwhile, the two-dimensional image corresponding to the circle is an ellipse. Since the target uses spheres, the shape of the resulting image is between circular and elliptical. Referring to fig. 5, it can be seen that the target is deformed corresponding to the circle in the image when rotating, and the way of extracting the center of the circle in the rotating scene is no longer applicable. The application uses the centroid of the marked hemispherical image area as the imaging point corresponding to the sphere center of the marked sphere.
In one embodiment, the step S21 includes:
Step S211: classifying the marks according to the imaging areas of the marks in the two-dimensional image to obtain a large mark and a small mark.
Specifically, although two-dimensional images of the preset calibration target obtained by the monocular camera may have different angles, both large marks and small marks of the preset calibration target in the obtained two-dimensional images are deformed. But can also be distinguished by large and small marks through the imaging area in the two-dimensional image.
In one embodiment, the step S31 includes:
Step S311: establishing a basic coordinate system by taking any one angle of the two-dimensional image as a circle center so as to determine the image coordinate of the large mark;
Step S312: and establishing a topological coordinate system of the two-dimensional image according to the topological relation between the large-identification image coordinates.
In steps S311-S312, referring to fig. 5, since the imaging area of the large marker balls is larger than that of the remaining marker balls, the image coordinates O i(ui,vi corresponding to the centers of the four large marker balls may be first determined, where i=1, 2,3,4. According to the topological relation of the four large identification balls, the point corresponding to the point O 2 with the largest v coordinate, the point corresponding to the point O 4 with the largest u coordinate and the point corresponding to the point O 1 with the smallest u coordinate in the four image coordinates are known, and the rest points are points O 3. With the point O 2 as an origin, a straight line parallel to the connecting line of the point O 1 and the point O 3 is an x t axis, and a y t axis is perpendicular to an x t axis, so that a target topological coordinate system is established.
In one embodiment, the step S41 includes:
Step S411: determining the image coordinates of the small mark according to the topological coordinate system and the image coordinates of the large mark;
Step S412: and obtaining the two-dimensional image coordinates corresponding to the preset calibration target according to the image coordinates of the large mark and the image coordinates of the small mark.
In steps S411-S412, based on the established topological coordinate system and the image coordinates of the large markers, the small markers may be found along the topological coordinate system to obtain the image coordinates of the small markers.
In one embodiment, the step S411 includes:
Step S4111: and searching the small mark along the horizontal axis and the vertical axis of the topological coordinate system by taking the pixel distance between the adjacent large mark image coordinates as an interval, and determining the small mark image coordinates.
Specifically, since the marker hemispheres are uniformly distributed, the remaining marker sphere centers in the x t axis and y t axis directions are found with the pixel spacing of adjacent large marker spheres in the x t axis and y t axis as a threshold. The image coordinates corresponding to the target can be obtained, and the two-dimensional image processing corresponding to the preset calibration target is completed.
In one embodiment, referring to fig. 6, the method comprises:
Step S12: acquiring a three-dimensional point cloud image corresponding to a preset calibration target;
step S22: identifying the marks in the three-dimensional point cloud image to obtain a large mark and a small mark in the marks;
step S32: establishing a topological coordinate system of the three-dimensional point cloud image according to the large identifier;
step S42: and determining the three-dimensional point cloud image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark.
In steps S12-S42, the three-dimensional point cloud image is processed to determine the image coordinates corresponding to the preset calibration target. The processing of the three-dimensional point cloud image needs to manually extract the three-dimensional point cloud of the corresponding area of the preset calibration target, and the three-dimensional point cloud needs to be classified because the three-dimensional point cloud consists of the identification ball point cloud and the target plate.
In one embodiment, the step S22 includes:
Step S221: acquiring an initial mark and an initial mark curvature;
step S222: determining target marks and target mark curvatures according to the initial marks and the initial mark curvatures;
Step S223: and determining a large mark and a small mark according to the target mark and the curvature of the target mark.
In steps S221-S223, the initial identifier refers to an identifier in a three-dimensional point cloud image obtained by scanning a preset calibration target through a laser radar, namely an initial large identifier and a small identifier. And the target mark refers to marks obtained by processing the initial mark, namely a target large mark and a target small mark.
In one embodiment, the step S221 includes:
step S2211: performing plane fitting on the three-dimensional point cloud to obtain the initial calibration target;
Step S2212: calculating the direction vector of the initial calibration target, and determining the direction of the initial calibration target;
Step S2213: and acquiring a three-dimensional point cloud of which the direction of the initial calibration target is in a first preset range, and calculating the curvature of an initial mark corresponding to the three-dimensional point cloud in the first preset range, wherein a graph formed by the three-dimensional point cloud of which the direction of the initial calibration target is in the first preset range is the initial mark.
In steps S2211-S2213, the first preset range refers to a range set according to the study object, and may be represented by [ a, b ], where a is smaller than b.
In one embodiment, the step S222 includes:
step S2221: rotating the initial mark to obtain a rotated mark;
Step S2222: searching a three-dimensional point cloud in a second preset range of the rotated marks, and if the number of the three-dimensional point cloud and the rotated marks reaches a preset threshold, taking the initial mark as a target mark, wherein the curvature of the initial mark is the curvature of the target mark.
In steps S2221-S2222, the second preset range refers to a range set according to the subject, and may be represented by [ c, d ], where c is smaller than d. The preset threshold value refers to a value set by the system, and can be 3 or 5. And if the number of the three-dimensional point cloud and the rotated marks is lower than a preset threshold value, ignoring the initial marks and continuing to search the three-dimensional point cloud.
In one embodiment, the step S32 includes:
Step S321: acquiring the large identifier;
Step S322: determining a three-dimensional point cloud corresponding to the large identifier according to the position relation of the large identifier;
step S323: fitting the three-dimensional point cloud corresponding to the large identifier to obtain the three-dimensional point cloud image coordinates of the large identifier;
Step S324: and establishing a topological coordinate system of the three-dimensional point cloud image according to the topological relation between the coordinates of the three-dimensional point cloud image with the large mark.
In steps S321-S324, the large marker balls are processed first because of the different curvatures of the marker balls, see fig. 5, the centers of the three large marker balls are located on the same straight line, and the included angle and distance with the fourth large marker ball are known. According to the constraint condition, point clouds corresponding to four large-identification hemispheres can be positioned in the sphere to be determined.
In one embodiment, the step S324 includes:
step S325: obtaining a residual identifier;
step S326: screening the residual identifiers to obtain screened identifiers;
step S327: and fitting the screened marks to determine small marks.
In steps S325 to S327, the remaining identifications refer to identifications that have not been determined through calculation. Since all the large identifiers are identified in the steps, the small identifiers are needed to be identified later so as to determine the small identifiers.
It should be understood that, although the steps in the flowchart of fig. 2 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or other steps.
In one embodiment, as shown in fig. 7, there is provided an image processing apparatus including: an image acquisition module 100, an identification recognition module 200, a coordinate system creation module 300, and an image coordinate determination module 400, wherein:
the image acquisition module 100 is used for acquiring an image corresponding to a preset calibration target;
The identification recognition module 200 is used for recognizing the identifications in the image to obtain large identifications and small identifications in the identifications;
a coordinate system establishing module 300, configured to establish a topological coordinate system of the image according to the large identifier;
The image coordinate determining module 400 is configured to determine, according to the topological coordinate system, the large identifier, and the small identifier, an image coordinate corresponding to the preset calibration target.
In one embodiment, the method comprises:
the two-dimensional image acquisition module 101 is used for acquiring a two-dimensional image corresponding to a preset calibration target;
The first identifier identifying module 201 is configured to identify identifiers in the two-dimensional image, so as to obtain a large identifier and a small identifier in the identifiers;
a first coordinate system establishing module 301, configured to establish a topological coordinate system of the two-dimensional image according to the large identifier;
The two-dimensional image coordinate determining module 401 is configured to determine two-dimensional image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large identifier and the small identifier.
In one embodiment, the first identification module 201 includes:
and the classification module 2011 is used for classifying the marks according to the imaging area of the marks in the two-dimensional image to obtain a large mark and a small mark.
In one embodiment, the first coordinate system establishment module 301 includes:
A basic coordinate establishing module 3011, configured to establish a basic coordinate system with any one angle of the two-dimensional image as a center of a circle, so as to determine an image coordinate of the large identifier;
The two-dimensional image topological coordinate establishing module 3012 is used for establishing a topological coordinate system of the two-dimensional image according to the topological relation among the image coordinates of the large identifier.
In one embodiment, the two-dimensional image coordinate determining module 401 includes:
a small identifier coordinate determining module 4011, configured to determine, according to the topological coordinate system and the image coordinates of the large identifier, the image coordinates of the small identifier;
the two-dimensional image coordinate determining module 4012 is configured to obtain, according to the image coordinate of the large identifier and the image coordinate of the small identifier, a two-dimensional image coordinate corresponding to the preset calibration target.
In one embodiment, the small-identification coordinate determination module 4011 comprises:
The small logo coordinate calculating module 4011a is configured to search the small logo along the horizontal axis and the vertical axis of the topological coordinate system with the pixel distance between the adjacent large logo image coordinates as an interval, and determine the small logo image coordinates.
In one embodiment, the method comprises:
The three-dimensional image acquisition module 102 is used for acquiring a three-dimensional point cloud image corresponding to a preset calibration target;
the second identifier identifying module 202 is configured to identify identifiers in the three-dimensional point cloud image, so as to obtain a large identifier and a small identifier in the identifiers;
A second coordinate system establishing module 302, configured to establish a topological coordinate system of the three-dimensional point cloud image according to the large identifier;
And the three-dimensional image coordinate determining module 402 is configured to determine three-dimensional point cloud image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large identifier and the small identifier.
In one embodiment, the second identification module 202 includes:
An initial identifier acquisition module 2021, configured to acquire an initial identifier and an initial identifier curvature;
a target identifier acquisition module 2022 for determining a target identifier and a target identifier curvature according to the initial identifier and the initial identifier curvature;
the label determining module 2023 is configured to determine a large label and a small label according to the target label and the target label curvature.
In one embodiment, the initial identity acquisition module 2021 includes:
the fitting module 2021a is configured to perform plane fitting on the three-dimensional point cloud to obtain the initial calibration target;
A vector calculation module 2021b, configured to calculate a direction vector of the initial calibration target, and determine a direction of the initial calibration target;
the curvature calculating module 2021c is configured to obtain a three-dimensional point cloud with a direction of the initial calibration target within a first preset range, calculate a curvature of an initial identifier corresponding to the three-dimensional point cloud within the first preset range, where a graph formed by the three-dimensional point cloud with the direction of the initial calibration target within the first preset range is the initial identifier.
In one embodiment, the target identification acquisition module 2022 includes:
A rotation module 2022a, configured to rotate the initial identifier to obtain a rotated identifier;
the target identifier determining module 2022b is configured to search for a three-dimensional point cloud within the second preset range of the rotated identifiers, and if the number of the three-dimensional point cloud and the rotated identifiers reaches a preset threshold, the initial identifier is a target identifier, where a curvature of the initial identifier is a curvature of the target identifier.
In one embodiment, the second coordinate system establishment module 302 includes:
A large identifier obtaining module 3021, configured to obtain the large identifier;
A large identifier corresponding point cloud determining module 3022, configured to determine a three-dimensional point cloud corresponding to the large identifier according to a position relationship of the large identifier;
the large-identifier coordinate determining module 3022 is configured to fit the three-dimensional point cloud corresponding to the large identifier to obtain three-dimensional point cloud image coordinates of the large identifier;
the three-dimensional topological coordinate system determining module 3023 is configured to establish a topological coordinate system of the three-dimensional point cloud image according to the topological relation between the coordinates of the three-dimensional point cloud image with the large identifier.
In one embodiment, the three-dimensional topological coordinate system determination module 3023 then comprises:
A residual identifier obtaining module 3024, configured to obtain a residual identifier;
A screening module 3025, configured to screen the remaining identifiers to obtain screened identifiers;
And the small identifier coordinate calculation module 3026 is configured to fit the screened identifier to determine a small identifier.
For specific limitations of the image processing apparatus, reference may be made to the above limitations of the image processing method, and no further description is given here. The respective modules in the above-described image processing apparatus may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, in connection with fig. 8, a vehicle includes an image processing apparatus as described above. The image processing device is applied to a vehicle and comprises a laser radar arranged at the roof position and a monocular camera arranged behind the front windshield of the vehicle, wherein the laser radar and the monocular camera are respectively connected with an in-vehicle industrial personal computer through data lines. The preset calibration target is arranged in front of the vehicle, the distance between the preset calibration target and the vehicle is set according to the identifiable degree of the laser radar and the monocular camera, and the specific distance is not limited.
And respectively acquiring images corresponding to the preset calibration targets through a laser radar and a monocular camera, transmitting the images to an in-car industrial personal computer, and identifying the images by the industrial personal computer so as to determine image coordinates corresponding to the preset calibration targets.
In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 9. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is for storing relevant data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement an image processing method.
It will be appreciated by persons skilled in the art that the architecture shown in fig. 9 is merely a block diagram of some of the architecture relevant to the present inventive arrangements and is not limiting as to the computer device to which the present inventive arrangements are applicable, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:
Acquiring an image corresponding to a preset calibration target;
Identifying the marks in the image to obtain a large mark and a small mark in the marks;
establishing a topological coordinate system of the image according to the large identifier;
And determining the image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
Acquiring an image corresponding to a preset calibration target;
Identifying the marks in the image to obtain a large mark and a small mark in the marks;
establishing a topological coordinate system of the image according to the large identifier;
And determining the image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (8)
1. An image processing method, the method comprising:
Acquiring an image corresponding to a preset calibration target;
Identifying the marks in the image to obtain a large mark and a small mark in the marks;
establishing a topological coordinate system of the image according to the large identifier;
determining image coordinates corresponding to the preset calibration targets according to the topological coordinate system, the large marks and the small marks;
Wherein the method further comprises:
Acquiring a two-dimensional image corresponding to a preset calibration target;
identifying the marks in the two-dimensional image to obtain a large mark and a small mark in the marks;
Establishing a topological coordinate system of the two-dimensional image according to the large identifier;
Determining two-dimensional image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark;
the identifying the identification in the two-dimensional image to obtain the large identification and the small identification in the identification comprises the following steps:
Classifying the marks according to the imaging area of the marks in the two-dimensional image to obtain a large mark and a small mark;
wherein, the establishing the topological coordinate system of the two-dimensional image according to the large identifier comprises:
establishing a basic coordinate system by taking any one angle of the two-dimensional image as a circle center so as to determine the image coordinate of the large mark;
Establishing a topological coordinate system of the two-dimensional image according to the topological relation between the large-identification image coordinates;
Wherein, according to the topological coordinate system, the large mark and the small mark, determining the two-dimensional image coordinate corresponding to the preset calibration target comprises:
determining the image coordinates of the small mark according to the topological coordinate system and the image coordinates of the large mark;
Obtaining two-dimensional image coordinates corresponding to the preset calibration target according to the image coordinates of the large mark and the image coordinates of the small mark;
wherein, the determining the image coordinates of the small mark according to the topological coordinate system and the image coordinates of the large mark comprises:
And searching the small mark along the horizontal axis and the vertical axis of the topological coordinate system by taking the pixel distance between the adjacent large mark image coordinates as an interval, and determining the small mark image coordinates.
2. The method according to claim 1, characterized in that the method comprises:
Acquiring a three-dimensional point cloud image corresponding to a preset calibration target;
Identifying the marks in the three-dimensional point cloud image to obtain a large mark and a small mark in the marks;
establishing a topological coordinate system of the three-dimensional point cloud image according to the large identifier;
and determining the three-dimensional point cloud image coordinates corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark.
3. The method of claim 2, wherein the identifying the markers in the three-dimensional point cloud image to obtain the large markers and the small markers in the markers comprises:
Acquiring an initial mark and an initial mark curvature;
determining target marks and target mark curvatures according to the initial marks and the initial mark curvatures;
determining a large mark and a small mark according to the target mark and the curvature of the target mark;
wherein, the obtaining the initial mark and the initial mark curvature comprises:
performing plane fitting on the three-dimensional point cloud to obtain the initial calibration target;
calculating the direction vector of the initial calibration target, and determining the direction of the initial calibration target;
acquiring a three-dimensional point cloud of which the direction of the initial calibration target is in a first preset range, and calculating the curvature of an initial mark corresponding to the three-dimensional point cloud in the first preset range, wherein a graph formed by the three-dimensional point cloud of which the direction of the initial calibration target is in the first preset range is the initial mark;
wherein, the determining the target mark and the target mark curvature according to the initial mark and the initial mark curvature comprises:
rotating the initial mark to obtain a rotated mark;
Searching a three-dimensional point cloud in a second preset range of the rotated marks, and if the number of the three-dimensional point cloud and the rotated marks reaches a preset threshold, taking the initial mark as a target mark, wherein the curvature of the initial mark is the curvature of the target mark.
4. A method according to claim 3, wherein said establishing a topological coordinate system of said three-dimensional point cloud image from said large identifier comprises:
Acquiring the large identifier;
Determining a three-dimensional point cloud corresponding to the large identifier according to the position relation of the large identifier;
Fitting the three-dimensional point cloud corresponding to the large identifier to obtain the three-dimensional point cloud image coordinates of the large identifier;
establishing a topological coordinate system of the three-dimensional point cloud image according to the topological relation between the coordinates of the three-dimensional point cloud image with the large mark;
wherein, the establishing the topological coordinate system of the three-dimensional point cloud image according to the topological relation between the three-dimensional point cloud image coordinates of the large identifier comprises:
Obtaining a residual identifier;
Screening the residual identifiers to obtain screened identifiers;
and fitting the screened marks to determine small marks.
5. An image processing apparatus, characterized in that the image processing apparatus is based on the image processing method according to any one of claims 1 to 4, the apparatus comprising:
The image acquisition module is used for acquiring an image corresponding to a preset calibration target;
the identification module is used for identifying the identifications in the images to obtain large identifications and small identifications in the identifications;
the coordinate system establishing module is used for establishing a topological coordinate system of the image according to the large identifier;
and the image coordinate determining module is used for determining the image coordinate corresponding to the preset calibration target according to the topological coordinate system, the large mark and the small mark.
6. A vehicle characterized in that it comprises the image processing apparatus according to claim 5.
7. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 4 when the computer program is executed.
8. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910778472.2A CN112241988B (en) | 2019-08-22 | 2019-08-22 | Image processing method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910778472.2A CN112241988B (en) | 2019-08-22 | 2019-08-22 | Image processing method, device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112241988A CN112241988A (en) | 2021-01-19 |
| CN112241988B true CN112241988B (en) | 2024-09-06 |
Family
ID=74168075
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910778472.2A Active CN112241988B (en) | 2019-08-22 | 2019-08-22 | Image processing method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112241988B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114943777A (en) * | 2022-06-20 | 2022-08-26 | 商汤国际私人有限公司 | Method, device and electronic device for calibrating external parameters of image acquisition equipment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102376089A (en) * | 2010-12-09 | 2012-03-14 | 深圳大学 | Target correction method and system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8854594B2 (en) * | 2010-08-31 | 2014-10-07 | Cast Group Of Companies Inc. | System and method for tracking |
| KR101748077B1 (en) * | 2016-04-05 | 2017-06-15 | 국방과학연구소 | Database Construction Method for Target Identification Using Bistatic Radar and Target Identification Apparatus of Bistatic Radar Using the Database and Target Identification Method of Bistatic Radar Using The Target Identification Apparatus |
| CN108932475B (en) * | 2018-05-31 | 2021-11-16 | 中国科学院西安光学精密机械研究所 | Three-dimensional target identification system and method based on laser radar and monocular vision |
| CN109118545B (en) * | 2018-07-26 | 2021-04-16 | 深圳市易尚展示股份有限公司 | Three-dimensional imaging system calibration method and system based on rotating shaft and binocular camera |
-
2019
- 2019-08-22 CN CN201910778472.2A patent/CN112241988B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102376089A (en) * | 2010-12-09 | 2012-03-14 | 深圳大学 | Target correction method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112241988A (en) | 2021-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11392146B2 (en) | Method for detecting target object, detection apparatus and robot | |
| CN110869974A (en) | Point cloud processing method, point cloud processing device and storage medium | |
| WO2016199605A1 (en) | Image processing device, method, and program | |
| CN113156407B (en) | Vehicle-mounted laser radar external parameter joint calibration method, system, medium and device | |
| US11045953B2 (en) | Relocalization method and robot using the same | |
| US12273646B2 (en) | Method for vehicle hinge point calibration and corresponding calibration apparatus, computer device, and storage medium | |
| CN112465877B (en) | Kalman filtering visual tracking stabilization method based on motion state estimation | |
| CN112308930A (en) | Camera external parameter calibration method, system and device | |
| Ding et al. | A robust detection method of control points for calibration and measurement with defocused images | |
| CN106558069A (en) | A kind of method for tracking target and system based under video monitoring | |
| US10902610B2 (en) | Moving object controller, landmark, and moving object control method | |
| CN112419405A (en) | Target tracking joint display method, security system and electronic equipment | |
| CN112241988B (en) | Image processing method, device, computer equipment and storage medium | |
| Bastanlar | A simplified two-view geometry based external calibration method for omnidirectional and PTZ camera pairs | |
| CN110750094B (en) | Method, device and system for determining posture change information of movable device | |
| CN116817929B (en) | Method and system for simultaneously positioning multiple targets on ground plane by unmanned aerial vehicle | |
| CN114119652A (en) | Method and device for three-dimensional reconstruction and electronic equipment | |
| CN113227708A (en) | Method and device for determining pitch angle and terminal equipment | |
| CN111223139A (en) | Target positioning method and terminal equipment | |
| CN111489433A (en) | Vehicle damage positioning method and device, electronic equipment and readable storage medium | |
| CN111383262A (en) | Occlusion detection method, system, electronic terminal and storage medium | |
| CN112446928B (en) | External parameter determining system and method for shooting device | |
| CN113378738A (en) | Comparison method and device, equipment and storage medium | |
| CN114187344A (en) | Map construction method, device and equipment | |
| Liu et al. | Outdoor camera calibration method for a GPS & camera based surveillance system |
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 | ||
| CB02 | Change of applicant information |
Address after: 100176 floor 10, building 1, zone 2, yard 9, Taihe 3rd Street, Beijing Economic and Technological Development Zone, Daxing District, Beijing Applicant after: Beijing National New Energy Vehicle Technology Innovation Center Co.,Ltd. Address before: 100176 1705, block a, building 1, No. 10, Ronghua Middle Road, economic and Technological Development Zone, Daxing District, Beijing Applicant before: BEIJING NEW ENERGY VEHICLE TECHNOLOGY INNOVATION CENTER Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |