Systems and methods for image capture calibration
Systems and methods in accordance with various embodiments of the invention can generate term (or PolyMap) coefficients that properly calibrate any camera lens using information obtained from a calibration pattern (e.g., chessboard patterns). Term coefficients in accordance with several embodiments of the invention can be used to transform a warped image to a dewarped image by mapping image information from the warped image to the dewarped coordinates. Determining calibration coefficients in accordance with certain embodiments of the invention can include novel and inventive processes for capturing calibration images, processing the calibration images, and/or deriving inputs needed for proper calibration as described in this disclosure. Processes described in this description can provide for improvements in the field of image processing, especially in increasing the speed of dewarping processes and in the accurate capture of calibration pattern images.
1. A method for image capture calibration, the method comprising:
identifying a set of real source keypoints from a real source image of a calibration pattern;
computing a set of ideal source keypoints based on the set of real source keypoints;
identifying ideal destination keypoints based on the ideal source keypoints;
adding boundary keypoints to the set of real source keypoints and the set of ideal destination keypoints; and
determining a set of term coefficients based on a mapping of the real source keypoints to the ideal destination keypoints.
2. The method of claim 1 , wherein identifying the set of real source keypoints comprises capturing the real source image by:
determining whether a calibration pattern is present in a first image;
when the calibration pattern is present in the first image, identifying and evaluating a position of the calibration pattern within the source image;
when the position of the calibration pattern meets a set of criteria, identifying the first image as the real source image.
3. The method of claim 2 , wherein determining whether the calibration pattern is present in the first image comprises:
identifying a plurality of regions of interest (ROIs);
computing a histogram for each ROI of the plurality of ROIs; and
determining that the calibration pattern is present when a strongly separated dense bimodal distribution is detected.
4. The method of claim 2 , wherein identifying and evaluating a position of the calibration pattern comprises identifying an amount of at least one selected from the group consisting of skew, tilt, and centeredness.
5. The method of claim 1 , wherein computing the set ideal source keypoints comprises:
determining a scale for the source image;
scaling coordinates for the set of real source keypoints based on the determined scale;
determining a set of geodesics based on the scaled coordinates; and
identifying intersects of the set of geodesics as ideal source keypoints.
6. The method of claim 5 , wherein determining the scale for the source image comprises identifying a pixel distance from a center of the real source image to a 180° field of view (FOV), wherein scaling the coordinates for the set of real source keypoints comprises dividing the coordinates by the identified pixel distance.
7. The method of claim 5 , wherein determining the set of geodesics comprises identifying geodesics along two axes, wherein each geodesic passes through a real source keypoint and a pair of points on a 180° field of view (FOV) boundary.
8. The method of claim 5 , wherein identifying intersects comprises:
identifying intersects in a first quadrant of the real source image; and
reflecting the identified intersects of the first quadrant to at least one other quadrant.
9. The method of claim 1 , wherein adding boundary points comprises:
selecting a set of boundary points at a boundary of an equirectangular destination image for the destination keypoints; and
identifying a corresponding set of boundary points at a 180° field of view (FOV) boundary for the real source keypoints.
10. The method of claim 9 , wherein identifying a corresponding set of boundary points comprises performing a Reverse PolyMap operation with affine transformation.
11. A non-transitory machine readable medium containing processor instructions for image capture calibration, where execution of the instructions by a processor causes the processor to perform a process that comprises:
identifying a set of real source keypoints from a real source image of a calibration pattern;
computing a set of ideal source keypoints based on the set of real source keypoints;
identifying ideal destination keypoints based on the ideal source keypoints;
adding boundary keypoints to the set of real source keypoints and the set of ideal destination keypoints; and
determining a set of term coefficients based on a mapping of the real source keypoints to the ideal destination keypoints.
12. The non-transitory machine readable medium of claim 11 , wherein identifying the set of real source keypoints comprises capturing the real source image by:
determining whether a calibration pattern is present in a first image;
when the calibration pattern is present in the first image, identifying and evaluating a position of the calibration pattern within the source image;
when the position of the calibration pattern meets a set of criteria, identifying the first image as the real source image.
13. The non-transitory machine readable medium of claim 12 , wherein determining whether the calibration pattern is present in the first image comprises:
identifying a plurality of regions of interest (ROIs);
computing a histogram for each ROI of the plurality of ROIs; and
determining that the calibration pattern is present when a strongly separated dense bimodal distribution is detected.
14. The non-transitory machine readable medium of claim 12 , wherein identifying and evaluating a position of the calibration pattern comprises identifying an amount of at least one selected from the group consisting of skew, tilt, and centeredness.
15. The non-transitory machine readable medium of claim 11 , wherein computing the set ideal source keypoints comprises:
determining a scale for the source image;
scaling coordinates for the set of real source keypoints based on the determined scale;
determining a set of geodesics based on the scaled coordinates; and
identifying intersects of the set of geodesics as ideal source keypoints.
16. The non-transitory machine readable medium of claim 15 , wherein determining the scale for the source image comprises identifying a pixel distance from a center of the real source image to a 180° field of view (FOV), wherein scaling the coordinates for the set of real source keypoints comprises dividing the coordinates by the identified pixel distance.
17. The non-transitory machine readable medium of claim 15 , wherein determining the set of geodesics comprises identifying geodesics along two axes, wherein each geodesic passes through a real source keypoint and a pair of points on a 180° field of view (FOV) boundary.
18. The non-transitory machine readable medium of claim 15 , wherein identifying intersects comprises:
identifying intersects in a first quadrant of the real source image; and
reflecting the identified intersects of the first quadrant to at least one other quadrant.
19. The non-transitory machine readable medium of claim 11 , wherein adding boundary points comprises:
selecting a set of boundary points at a boundary of an equirectangular destination image for the destination keypoints; and
identifying a corresponding set of boundary points at a 180° field of view (FOV) boundary for the real source keypoints.
20. The non-transitory machine readable medium of claim 19 , wherein identifying a corresponding set of boundary points comprises performing a Reverse PolyMap operation with affine transformation.