Binocular ranging system calibration method based on electronic map
Through the electronic map-based method, the disparity value is calculated using the distance and correction parameters on the target reference object and the electronic map to generate the ranging parameters, which solves the problem of the optical axis not parallel and the deployment situation of the binocular ranging system after transportation, and achieves rapid calibration and high-precision ranging.
Patent Information
- Application Number
- CN202411705195.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-26
- Publication Date
- 2025-03-14
AI Technical Summary
The binocular distance measuring system no longer parallel due to bumps after long-distance transportation, and the failure requires on-site correction. In some deployment occasions, drone flight is prohibited, resulting in the inability to quickly calibrate and deploy.
The binocular ranging system calibration method based on electronic map is adopted, and the target reference objects are selected, the observation angle is adjusted, the correction parameters are calculated, and the disparity value is calculated based on the target reference objects distance and correction parameters on the electronic map are calculated, and the distance measurement parameters are generated, and the system calibration is completed.
The rapid calibration and deployment of the binocular ranging system is realized, the system adaptability and ranging accuracy are improved, and the dependence on drones is avoided.
Smart Images
Figure CN119618071A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of binocular stereo vision, and in particular to a binocular distance measurement system calibration method based on an electronic map. Background Art
[0002] Binocular ranging is a computer vision method that uses the principle of triangulation and digital image processing technology to obtain the matching points of two images, thereby obtaining depth information. The binocular ranging system is calibrated at the factory so that the optical axes of the two cameras are parallel. The ranging parameters are generated by calibrating targets at different distances, thereby achieving high ranging accuracy.
[0003] However, when the ranging system is transported over long distances, the bumps during transportation will cause the optical axes of the two cameras to no longer be parallel, and the deployment site does not have the calibration equipment to calibrate the optical axes. At the same time, the change of the optical axis also makes the ranging parameters invalid. Therefore, the optical axis of the binocular ranging system must be recalibrated on site to regenerate the ranging parameters. For some special deployment occasions, such as airports, drones are strictly prohibited from flying. Therefore, a distance parameter generation method that does not rely on drones must be found. The binocular ranging system can be quickly calibrated and generate ranging parameters, thereby achieving the purpose of rapid deployment. Summary of the invention
[0004] The present invention provides a binocular distance measurement system calibration method based on an electronic map, so as to solve the problem that the existing binocular distance measurement system relies on the calibration of an unmanned aerial vehicle.
[0005] In order to achieve the above object, the present invention is implemented by the following technical solutions: The present invention provides a binocular distance measurement system calibration method based on an electronic map, comprising the following steps: The steps include: Step 1: Select a first target reference object, adjust the observation angle of the binocular ranging system, and obtain multiple sets of images of the first target reference object at a predetermined position in the image; Step 2: Calculating correction parameters of the binocular ranging system based on multiple sets of images of the first target reference object; Step 3: Selecting a number of second target reference objects on the electronic map according to a predetermined constraint, obtaining a plurality of sets of images of the second target reference objects, and calculating the disparity value of the binocular ranging system by combining the distance from the second target reference object on the electronic map to the binocular ranging system and the correction parameter; Step 4: Obtain the calibration value of the ranging parameter of the binocular ranging system according to the disparity value, and complete the calibration of the binocular ranging system.
[0006] Further, the adjusting the observation angle of the binocular ranging system includes: adjusting the pitch and azimuth of the binocular ranging system; The predetermined positions include the first target reference object being located at the upper left, upper right, lower left, lower right and the middle of the image.
[0007] Furthermore, the distance between the first target reference object and the binocular ranging system is the detectable distance of the binocular ranging system. .
[0008] Furthermore, the step 2 comprises the following steps: Step 201: Based on the multiple groups of images of the first target reference object, combined with the centroid coordinates of the first target reference object in the image, respectively construct a target set of the first target reference object in the first field of view and the second field of view of the binocular ranging system, and pair the centroid coordinates of the target set in the first field of view and the target set in the second field of view to obtain multiple groups of target pairs; Step 202: Calculate the correction parameters of the binocular ranging system according to the multiple target pairs combined with the affine transformation model and the coordinate transformation relationship of the multiple target pairs in the binocular ranging system.
[0009] Further, in step 202, the coordinate transformation relationship of the plurality of target pairs in combination with the affine transformation model and the plurality of target pairs in the binocular ranging system is expressed as: ; in, and Represents the column coordinates and row coordinates of the first target reference object in the first field of view image of the binocular ranging system; and Represents the column coordinates and row coordinates of the first target reference object in the second field of view image of the binocular ranging system; represents an affine transformation model; The affine transformation model It is expressed as: ; in, to Represents the affine transformation model parameter value; The correction parameter is expressed by the following formula: ; in, represents the correction parameter; , indicating that there is A second target reference; Column coordinates representing the coordinates of the center of mass of the first target reference object in the second field of view; row coordinates representing the coordinates of the centroid of the first target reference object in the second field of view; Column coordinates representing the coordinates of the center of mass of the first target reference object in the first field of view; The row coordinates representing the coordinates of the center of mass of the first target reference object in the first field of view.
[0010] Furthermore, the step 3 comprises the following steps: Step 301: selecting a number of second target reference objects in an electronic map according to a predetermined constraint, and acquiring a plurality of groups of images of the second target reference objects; Step 302: Based on the multiple groups of images of the second target reference object, combined with the centroid coordinates of the second target reference object in the image, respectively construct a target set of the second target reference object in the first field of view and the second field of view of the binocular ranging system, and correct the target set of the second field of view in combination with the distance of the second target reference object and the affine transformation model to obtain a corrected target set of the second field of view; Step 303: Based on the target set in the corrected second field of view, the disparity value of the binocular ranging system is calculated in combination with the inverse function relationship between the disparity and the distance value.
[0011] Furthermore, the correction expression for the second field of view target set by combining the distance of the second target reference object and the affine transformation model is as follows: ; in, The column coordinates representing the centroid coordinates of the target set after correction of the target set in the second field of view; The row coordinates representing the centroid coordinates of the target set after correction of the target set in the second field of view; Column coordinates representing the coordinates of the center of mass of the second target reference object in the second field of view; row coordinates representing the coordinates of the center of mass of the second target reference object in the second field of view; The inverse function relationship between the parallax and the distance value is as follows: ; in, is the distance value, i.e., the distance between the second target reference object and the binocular ranging system; is the disparity value; , , These are all ranging parameters.
[0012] Further, in step 303, when the first field of view and the second field of view in the binocular ranging system are installed horizontally, the expression of the parallax value is as follows: ; When the first field of view and the second field of view in the binocular ranging system are installed vertically, the expression of the parallax value is as follows: .
[0013] Furthermore, the predetermined constraints include: the number of second target reference objects is greater than or equal to 6, and the distance from the binocular ranging system is between 100 meters and 1200 meters.
[0014] Further, in step 4, obtaining the calibration value of the ranging parameter of the binocular ranging system according to the disparity value includes: obtaining the calibration value of the ranging parameter of the binocular ranging system according to the disparity value in combination with numerical optimization technology; The expression of the calibration value is: .
[0015] Beneficial effects: The present invention provides a binocular ranging system calibration method based on an electronic map. Typical buildings at different distances are found on the electronic map, and their disparity values in a left second field of view image are obtained based on an acquired image registration algorithm, thereby generating ranging parameters. The generated ranging parameters have high ranging accuracy, and the binocular ranging system can be quickly calibrated without a drone, thereby improving the adaptability of the system and reducing the deployment requirements of the binocular ranging system. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 It is a flow chart of a binocular distance measurement system calibration method based on an electronic map according to a preferred embodiment of the present invention; Figure 2 Schematic diagram of a first target reference object at different positions in an image according to a preferred embodiment of the present invention; Figure (a) is a schematic diagram of the first target reference object being located at a position A slightly to the left of the center of the left eye image; Figure (b) is a schematic diagram of the first target reference object being at position A in the right eye image; Figure (c) is a schematic diagram showing that the first target reference object is located at position B slightly to the right of the center of the left eye image; Figure (d) is a schematic diagram of the first target reference object being at position B in the right eye image. DETAILED DESCRIPTION
[0017] The technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0018] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the usual meanings understood by persons with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "one" do not indicate quantity restrictions, but indicate the existence of at least one. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship also changes accordingly.
[0019] See also Figure 1 The present application embodiment provides a binocular ranging system calibration method based on an electronic map, comprising: Step 1: Select a first target reference object, adjust the observation angle of the binocular ranging system, and obtain multiple sets of images of the first target reference object at a predetermined position in the image; Specifically, when the first target reference object is selected, the distance between the first target reference object and the binocular ranging system is the detectable distance of the binocular ranging system. ; Adjusting the observation angle of the binocular ranging system includes: adjusting the pitch and azimuth of the binocular ranging system; The predetermined positions include the first target reference object being located at the upper left, upper right, lower left, lower right and the middle of the image.
[0020] Step 2: Calculate the correction parameters of the binocular ranging system based on multiple sets of images of the first target reference object; Step 201: Based on multiple groups of images of the first target reference object, combined with the centroid coordinates of the first target reference object in the image, respectively construct a target set of the first target reference object in the first field of view and the second field of view of the binocular ranging system, and pair the centroid coordinates of the target set in the first field of view and the target set in the second field of view to obtain multiple groups of target pairs; Then the center of mass coordinates of the first target reference object in the binocular ranging system are , is the column coordinate, is the row coordinate; then the target set formed by the first target reference object in the first field of view is , and the target set formed by matching the first target reference object one by one in the second field of view is ,like Form a target pair.
[0021] Step 202: Calculate the correction parameters of the binocular ranging system according to the multiple target pairs combined with the affine transformation model and the coordinate transformation relationship of the multiple target pairs in the binocular ranging system.
[0022] The affine transformation model is used to correct the optical axis. It is expressed as: ; in, to Represents the affine transformation model parameter value; Then the coordinate transformation relationship of multiple target pairs combined with the affine transformation model and multiple target pairs in the binocular ranging system is expressed as: ; in, and Represents the row and column coordinates of the first target reference object in the left eye image of the binocular ranging system; and Represents the row and column coordinates of the first target reference object in the right eye image of the binocular ranging system; represents an affine transformation model; The correction parameters are expressed by the following formula: ; in, represents the correction parameter; , indicating that there is A second target reference; Column coordinates representing the coordinates of the center of mass of the first target reference object in the second field of view; row coordinates representing the coordinates of the centroid of the first target reference object in the second field of view; Column coordinates representing the coordinates of the center of mass of the first target reference object in the first field of view; The row coordinates representing the coordinates of the center of mass of the first target reference object in the first field of view.
[0023] Step 3: Selecting a number of second target reference objects on the electronic map according to predetermined constraints, obtaining multiple sets of images of the second target reference objects, and calculating the disparity value of the binocular ranging system by combining the distance from the second target reference object on the electronic map to the binocular ranging system and the correction parameters; Step 301: Selecting a number of second target reference objects in an electronic map according to a predetermined constraint, and acquiring a plurality of groups of images of the second target reference objects; When selecting the second target reference objects, the number of the second target reference objects is greater than or equal to 6, and the distance between the second target reference objects and the binocular ranging system is between 100 meters and 1200 meters.
[0024] Step 302: Based on the multiple groups of images of the second target reference object, combined with the centroid coordinates of the second target reference object in the image, respectively construct a target set of the second target reference object in the first field of view and the second field of view of the binocular ranging system, and correct the target set of the second field of view in combination with the distance of the second target reference object and the affine transformation model to obtain a corrected target set of the second field of view; Step 303: Based on the target set in the corrected second field of view, the disparity value of the binocular ranging system is calculated in combination with the inverse function relationship between the disparity and the distance value.
[0025] ; in, The column coordinates representing the centroid coordinates of the target set after correction of the target set in the second field of view; The row coordinates representing the centroid coordinates of the target set after correction of the target set in the second field of view; Column coordinates representing the coordinates of the center of mass of the second target reference object in the second field of view; row coordinates representing the coordinates of the center of mass of the second target reference object in the second field of view; The inverse function relationship between parallax and distance value is as follows: ; in, is the distance value, i.e., the distance between the second target reference object and the binocular ranging system; is the disparity value; in, is the ranging parameter.
[0026] Without considering the installation relationship between the first field of view and the second field of view, the parallax The calculation formula can be used as follows: ; When the first field of view and the second field of view in the binocular ranging system are installed horizontally, the calculation formula of the parallax value can be used: ; When the first field of view and the second field of view in the binocular ranging system are installed vertically, the calculation formula of the parallax value can be used: .
[0027] Step 4: Obtain the calibration value of the ranging parameter of the binocular ranging system according to the disparity value, and complete the calibration of the binocular ranging system.
[0028] Specifically, the calibration value of the ranging parameter of the binocular ranging system is obtained according to the parallax value combined with the numerical optimization technology; The expression of the calibration value is: .
[0029] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.
Claims
1. A binocular distance measurement system calibration method based on electronic map, characterized in that: The steps include: Step 1: Select a first target reference object, adjust the observation angle of the binocular ranging system, and obtain multiple sets of images of the first target reference object at a predetermined position in the image; Step 2: Calculating correction parameters of the binocular ranging system based on multiple sets of images of the first target reference object; Step 3: Selecting a number of second target reference objects on the electronic map according to a predetermined constraint, obtaining a plurality of sets of images of the second target reference objects, and calculating the disparity value of the binocular ranging system by combining the distance from the second target reference object on the electronic map to the binocular ranging system and the correction parameter; Step 4: Obtain the calibration value of the ranging parameter of the binocular ranging system according to the disparity value, and complete the calibration of the binocular ranging system.
2. The method for calibrating a binocular distance measurement system based on an electronic map according to claim 1, characterized in that: The adjusting the observation angle of the binocular ranging system comprises: adjusting the pitch and azimuth of the binocular ranging system; The predetermined positions include the first target reference object being located at the upper left, upper right, lower left, lower right and the middle of the image.
3. The method for calibrating a binocular ranging system based on an electronic map according to any one of claims 1 to 2, characterized in that: The distance between the first target reference object and the binocular ranging system is the detectable distance of the binocular ranging system. .
4. The method for calibrating a binocular distance measurement system based on an electronic map according to claim 1, characterized in that: The step 2 comprises the following steps: Step 201: Based on the multiple groups of images of the first target reference object, combined with the centroid coordinates of the first target reference object in the image, respectively construct a target set of the first target reference object in the first field of view and the second field of view of the binocular ranging system, and pair the centroid coordinates of the target set in the first field of view and the target set in the second field of view to obtain multiple groups of target pairs; Step 202: Calculate the correction parameters of the binocular ranging system according to the multiple target pairs combined with the affine transformation model and the coordinate transformation relationship of the multiple target pairs in the binocular ranging system.
5. The method for calibrating a binocular distance measurement system based on an electronic map according to claim 4, characterized in that: In step 202, the coordinate transformation relationship of the plurality of target pairs in combination with the affine transformation model and the plurality of target pairs in the binocular ranging system is expressed as: ； in, and Represents the column coordinates and row coordinates of the first target reference object in the first field of view image of the binocular ranging system; and Represents the column coordinates and row coordinates of the first target reference object in the second field of view image of the binocular ranging system; represents an affine transformation model; The affine transformation model It is expressed as: ； in, to Represents the affine transformation model parameter value; The correction parameter is expressed by the following formula: ； in, represents the correction parameter; , indicating that there is A second target reference; row coordinates representing the coordinates of the centroid of the first target reference object in the second field of view; Column coordinates representing the coordinates of the center of mass of the first target reference object in the second field of view; row coordinates representing the coordinates of the center of mass of the first target reference object in the first field of view; The column coordinates representing the coordinates of the center of mass of the first target reference object in the first field of view.
6. The method for calibrating a binocular distance measurement system based on an electronic map according to claim 1, characterized in that: The step 3 comprises the following steps: Step 301: selecting a number of second target reference objects in an electronic map according to a predetermined constraint, and acquiring a plurality of groups of images of the second target reference objects; Step 302: Based on the multiple groups of images of the second target reference object, combined with the centroid coordinates of the second target reference object in the image, respectively construct a target set of the second target reference object in the first field of view and the second field of view of the binocular ranging system, and correct the target set of the second field of view in combination with the distance of the second target reference object and the affine transformation model to obtain a corrected target set of the second field of view; Step 303: Based on the target set in the corrected second field of view, the disparity value of the binocular ranging system is calculated in combination with the inverse function relationship between the disparity and the distance value.
7. The method for calibrating a binocular distance measurement system based on an electronic map according to claim 6, characterized in that: The expression for correcting the second field of view target set by combining the distance of the second target reference object and the affine transformation model is as follows: ； in, The column coordinates representing the centroid coordinates of the target set after correction of the target set in the second field of view; The row coordinates representing the centroid coordinates of the target set after correction of the target set in the second field of view; Column coordinates representing the coordinates of the center of mass of the second target reference object in the second field of view; row coordinates representing the coordinates of the center of mass of the second target reference object in the second field of view; The inverse function relationship between the parallax and the distance value is as follows: ； in, is the distance value, i.e., the distance between the second target reference object and the binocular ranging system; is the disparity value; , , These are all ranging parameters.
8. The method for calibrating a binocular distance measurement system based on an electronic map according to claim 6, characterized in that: In step 303, when the first field of view and the second field of view in the binocular ranging system are installed horizontally, the expression of the parallax value is as follows: ； When the first field of view and the second field of view in the binocular ranging system are installed vertically, the expression of the parallax value is as follows: 。 9. The method for calibrating a binocular distance measurement system based on an electronic map according to claim 6, characterized in that: The predetermined constraints include: the number of second target reference objects is greater than or equal to 6, and the distance from the binocular ranging system is between 100 meters and 1200 meters.
10. The method for calibrating a binocular distance measurement system based on an electronic map according to claim 1, characterized in that: In step 4, the step of obtaining the calibration value of the ranging parameter of the binocular ranging system according to the disparity value includes: obtaining the calibration value of the ranging parameter of the binocular ranging system according to the disparity value in combination with the numerical optimization technology; The expression of the calibration value is: 。