Method and system for establishing coordinates of target region

Abstract

A method for establishing the coordinates of a target region includes: obtaining the image coordinates of an object as it passes through the target region using a camera device; obtaining reference spatial coordinates by detecting the object's passage through the target region using a coordinate generation device; and performing the following steps in a correction phase using a processing device: performing a transformation process on the reference spatial coordinates and the image coordinates to obtain an initial extrinsic parameter matrix and an initial intrinsic parameter matrix of the camera device; performing a correction process on the reference spatial coordinates, the image coordinates, the initial extrinsic parameter matrix, and the initial intrinsic parameter matrix to obtain a correction extrinsic parameter matrix, a correction intrinsic parameter matrix, and a plurality of distortion parameters; and establishing a transformation model corresponding to the camera device based on the correction extrinsic parameter matrix, the correction intrinsic parameter matrix, and the distortion parameters.

Claims

1. A method for establishing the coordinates of a target region, comprising: The system employs a first camera device and a second camera device to capture N first images and N second images respectively when a first object passes through N positions within a target area, thereby obtaining N first image coordinates of the first object at the N positions in the N first images and N second image coordinates of the first object at the N positions in the N second images; a coordinate generation device detects the first object passing through the N positions in the target area to obtain N reference spatial coordinates corresponding to the N positions; and a processing device performs the following steps in a correction stage: performing a transformation process on the N reference spatial coordinates and the N first image coordinates to obtain a first initial extrinsic parameter matrix and a first initial intrinsic parameter matrix of the first camera device; performing the transformation process on the N reference spatial coordinates and the N second image coordinates to obtain a second initial extrinsic parameter matrix and a second initial intrinsic parameter matrix of the second camera device; A correction process is performed on the N reference spatial coordinates, the N first image coordinates, the first initial extrinsic parameter matrix, and the first initial intrinsic parameter matrix to obtain a first correction extrinsic parameter matrix, a first correction intrinsic parameter matrix, and a plurality of first distortion parameters; the correction process is then performed on the N reference spatial coordinates, the N second image coordinates, the second initial extrinsic parameter matrix, and the second initial intrinsic parameter matrix to obtain a second correction extrinsic parameter matrix, a second correction intrinsic parameter matrix, and a plurality of second distortion parameters; a first transformation model corresponding to the first imaging device is established based on the first correction extrinsic parameter matrix, the first correction intrinsic parameter matrix, and the first distortion parameters; a second transformation model corresponding to the second imaging device is established based on the second correction extrinsic parameter matrix, the second correction intrinsic parameter matrix, and the second distortion parameters; and a physical spatial coordinate of a target location in the target area is established based on the first transformation model and the second transformation model.

2. The method for establishing the coordinates of the target region as described in claim 1 further includes: In one operational phase, the first and second camera devices simultaneously capture a third image and a fourth image of a second object at a first time point during the period when the object passes through the target area, respectively, to obtain the third image coordinates of the second object in the third image and the fourth image coordinates of the second object in the fourth image; and the processing device performs the following steps: performing calculations on the third image coordinates according to the first transformation model to obtain a first projection ray; performing calculations on the fourth image coordinates according to the second transformation model to obtain a second projection ray; and performing triangulation on the first and second projection rays to obtain the physical spatial coordinates of the second object at the first time point.

3. The method for establishing the coordinates of the target area as described in claim 2, wherein during the calibration phase, the coordinate generation device is set in the target area for detection.

4. The method for establishing coordinates of a target region as described in claim 3, wherein during the operation phase, the coordinate generation device is removed from the target region.

5. The method for establishing the coordinates of the target area as described in claim 1, wherein the transformation process of the N reference spatial coordinates and the N first image coordinates is performed by performing algebraic operations on the N spatial coordinates Si and the N first image coordinates M1i according to equation (1) to obtain the first initial extrinsic parameter matrix and the first initial intrinsic parameter matrix, …(1) where, , .

6. The method for establishing the coordinates of the target region as described in claim 5, wherein the correction process for the N reference spatial coordinates, the N first image coordinates, the first initial extrinsic matrix, and the first initial intrinsic matrix includes: The first distortion model, which includes a plurality of first initial distortion parameters, the first initial extrinsic parameter matrix, and the first initial intrinsic parameter matrix are used to transform the N reference spatial coordinates to obtain N first estimated image coordinates; and the N first estimated image coordinates and the N first image coordinates are iteratively calculated according to an error function to obtain the first correction extrinsic parameter matrix, the first correction intrinsic parameter matrix, and the first distortion parameters.

7. The method for establishing the coordinates of the target region as described in claim 6, wherein iterative calculations are performed on the N first estimated image coordinates and the N first image coordinates according to the error function to obtain the first correction extrinsic matrix, the first correction intrinsic matrix, and the first distortion parameters, including: The error function iteratively updates the first initial extrinsic parameter matrix, the first initial intrinsic parameter matrix, and the N first estimated image coordinates by fixing the values ​​of the first initial distortion parameters. In response to an error value of the error function being less than a first threshold, the updated first initial extrinsic parameter matrix and the first initial intrinsic parameter matrix are respectively set as the first correction extrinsic parameter matrix and the first correction intrinsic parameter matrix. The N reference spatial coordinates are transformed according to the first distortion model, the first correction extrinsic parameter matrix, and the first correction intrinsic parameter matrix to obtain N first adjusted estimated image coordinates. The error function iteratively updates the first initial distortion parameters and the N first adjusted estimated image coordinates by updating the N first adjusted image coordinates. In response to an error value of the error function being less than a second threshold, the updated first initial distortion parameters are respectively set as the first distortion parameters.

8. The method for establishing the coordinates of the target region as described in claim 1, wherein the transformation process of the N reference spatial coordinates and the N second image coordinates is performed by performing algebraic operations on the N spatial coordinates Si and the N second image coordinates M2i according to equation (2) to obtain the second initial extrinsic parameter matrix and the second initial intrinsic parameter matrix, …(2) where, , .

9. The method for establishing the coordinates of the target region as described in claim 8, wherein the correction processing of the N reference spatial coordinates, the N second image coordinates, the second initial extrinsic matrix, and the second initial intrinsic matrix includes: The N reference spatial coordinates are transformed according to a second distortion model including a plurality of second initial distortion parameters, the second initial extrinsic parameter matrix, and the second initial intrinsic parameter matrix to obtain N second estimated image coordinates; and the N second estimated image coordinates and the N second image coordinates are iteratively calculated according to an error function to obtain the second correction extrinsic parameter matrix, the second correction intrinsic parameter matrix, and the second distortion parameters.

10. The method for establishing the coordinates of the target region as described in claim 9, wherein iterative calculations are performed on the N second estimated image coordinates and the N second image coordinates according to the error function to obtain the second correction extrinsic matrix, the second correction intrinsic matrix, and the second distortion parameters, including: Based on the error function, with the values ​​of the second initial distortion parameters fixed, iterative calculations are performed on the N second estimated image coordinates and the N second image coordinates to update the second initial extrinsic parameter matrix, the second initial intrinsic parameter matrix, and the N second estimated image coordinates; in response to an error value of the error function being less than a third threshold, the updated second initial extrinsic parameter matrix and the second initial intrinsic parameter matrix are respectively set as the second correction extrinsic parameter matrix and the second correction intrinsic parameter matrix; the N reference spatial coordinates are transformed according to the second distortion model, the second correction extrinsic parameter matrix, and the second correction intrinsic parameter matrix to obtain N second adjusted estimated image coordinates; iterative calculations are performed on the N second adjusted estimated image coordinates and the N second image coordinates based on the error function to update the second initial distortion parameters and the N second adjusted estimated image coordinates; and in response to an error value of the error function being less than a fourth threshold, the updated second initial distortion parameters are respectively set as the second distortion parameters.

11. The method for establishing the coordinates of a target region as described in claim 1, wherein the first transformation model is configured to perform the following processing on the image coordinates of a first test point in an image captured by the first camera device of a first test point: correcting the image coordinates of the first test point according to the first distortion parameters; performing a coordinate transformation operation of the first correction intrinsic parameter matrix on the corrected image coordinates of the first test point to obtain a first normalized test point coordinate; and performing a coordinate transformation operation of the first correction extrinsic parameter matrix on the first normalized test point coordinate to obtain a projection ray of the first test point.

12. The method for establishing the coordinates of the target region as described in claim 1, wherein the second transformation model is configured to perform the following processing on the image coordinates of a second test point in an image captured by the second camera device of a second test point: correcting the image coordinates of the second test point according to the second distortion parameters; performing a coordinate transformation operation of the second correction intrinsic parameter matrix on the corrected image coordinates of the second test point to obtain a second normalized test point coordinate; and performing a coordinate transformation operation of the second correction extrinsic parameter matrix on the second normalized test point coordinate to obtain a projection ray of the second test point.

13. A coordinate system for establishing a target region, comprising: A first camera device is configured to capture N first images as an object passes through N positions within a target area, to obtain N first image coordinates of the object at the N positions in the N first images; a second camera device is configured to capture N second images as the object passes through the N positions within the target area, to obtain N second image coordinates of the object at the N positions in the N second images; a coordinate generation device is disposed at a position corresponding to the target area and configured to detect the object passing through the N positions within the target area and obtain N reference spatial coordinates corresponding to the N positions; and a processing device is electrically connected to the first camera device, the second camera device, and the coordinate generation device, and configured to perform the following steps: performing a transformation process on the N reference spatial coordinates and the N first image coordinates to obtain a first initial extrinsic parameter matrix and a first initial intrinsic parameter matrix of the first camera device; The transformation process is performed on the N reference spatial coordinates and the N second image coordinates to obtain a second initial extrinsic parameter matrix and a second initial intrinsic parameter matrix of the second camera device; a correction process is performed on the N reference spatial coordinates, the N first image coordinates, the first initial extrinsic parameter matrix, and the first initial intrinsic parameter matrix to obtain a first correction extrinsic parameter matrix, a first correction intrinsic parameter matrix, and a plurality of first distortion parameters; the correction process is performed on the N reference spatial coordinates, the N second image coordinates, the second initial extrinsic parameter matrix, and the second initial intrinsic parameter matrix to obtain a second correction extrinsic parameter matrix, a second correction intrinsic parameter matrix, and a plurality of second distortion parameters; a first transformation model corresponding to the first camera device is established based on the first correction extrinsic parameter matrix, the first correction intrinsic parameter matrix, and the first distortion parameters. A second transformation model corresponding to the second camera device is established based on the second correction extrinsic parameter matrix, the second correction intrinsic parameter matrix, and the second distortion parameters; and a physical spatial coordinate of a target location of the target region is established based on the first transformation model and the second transformation model.

14. The target region coordinate establishment system as described in claim 13, wherein the first transformation model is configured to perform the following processing on a first target image coordinate in an image captured by the first camera device of the target location: correcting the first target image coordinate according to the first distortion parameters; performing a coordinate transformation operation of the first correction intrinsic parameter matrix on the corrected first target image coordinate to obtain a first normalized target coordinate; and performing a coordinate transformation operation of the first correction extrinsic parameter matrix on the first normalized target coordinate to obtain a first target projection ray.

15. The target region coordinate establishment system as described in claim 14, wherein the second transformation model is configured to perform the following processing on a second target image coordinate in an image captured by the second camera device of the target location: correcting the second target image coordinate according to the second distortion parameters; performing a coordinate transformation operation of the second correction intrinsic parameter matrix on the corrected second target image coordinate to obtain a second normalized target coordinate; and performing a coordinate transformation operation of the second correction extrinsic parameter matrix on the second normalized target coordinate to obtain a second target projection ray.

16. The target region coordinate establishment system as described in claim 15, wherein the image corresponding to the first target image coordinate and the image corresponding to the second target image coordinate are captured simultaneously by the first camera device and the second camera device, respectively, wherein the physical spatial coordinates of the target location are obtained based on the first target projection ray and the second target projection ray.

Images