Method and apparatus for automatic intrinsic camera calibration using images of a planar calibration pattern
The present disclosure relates to a high precision method, model, and apparatus for calibrating the intrinsic camera parameter values of one or more intrinsic parameters of a camera. The intrinsic parameters include focal lengths, skew, optical center, and radial distortion coefficients. The practical implementation of the solution consists of a camera capturing several images of a planar calibration pattern from different viewpoints, and a computer implemented processing pipeline. The solution does not require additional human input and is robust to misdetection of the calibration pattern in some of the images.
1. A method comprising:
detecting, by a computing unit, a plurality of calibration features in each of a plurality of images;
generating for each of the plurality of images a corresponding lattice graph based on the plurality of calibration features, the generating referencing a global planar coordinate system having an arbitrary origin;
selecting with the computing unit one or more image capture parameters from a plurality of image capture parameters associated with capturing the plurality of images, the selecting based on a robust estimator of a reprojection error over the plurality of images; and
calibrating the plurality of image capture parameters based on one or more of the one or more selected image capture parameters and two or more of the calibration features detected in two or more of the plurality of images and the corresponding lattice graphs.
2. The method of claim 1 , wherein a least two of the plurality of images depict at least one of the plurality of calibration features from different respective angles.
3. The method of claim 1 , wherein each corresponding lattice graph has a plurality of nodes corresponding to a portion of the plurality of calibration features.
4. The method of claim 1 , wherein calibrating comprises repeated calibrating a plurality values of at least one of the plurality of image capture parameters for two or more different combinations of the plurality of images.
5. The method of claim 4 , wherein calibrating further comprises a plurality of recursions each starting with a single node detected in the image, one or more n-edge vectors and one or more visited directions and continuing via a repeated selection of one or more n-closest neighbors to the one or more n-edge vectors.
6. The method of claim 1 , wherein the selecting is further based on minimizing the robust estimator of the reprojection error.
7. The method of claim 1 , wherein detecting the plurality of calibration features includes, for each image, performing a blob detection process on the image.
8. The method of claim 7 , wherein the blob detection process is Laplacian of Gaussian.
9. The method of claim 7 , wherein the blob detection process is Maximally Stable Extremum Regions.
10. The method of claim 1 , wherein the corresponding lattice graph comprises a plurality of nodes that are based on a respective center of a respective calibration feature of the plurality of calibration features.
11. The method of claim 1 , wherein the one or more image capture parameters includes at least one of focal length, skew, optical center, and/or radial distortion coefficients.
12. A computing system comprising:
a processor;
a computer memory in communication with the processor the computer memory storing a plurality of digital images comprising a plurality of image capture calibration features, a plurality of image capture parameter data values, a plurality of lattice graphs, each lattice graph corresponding to one of the plurality of digital images and having nodes that correspond to a portion of the plurality of image capture calibration features, the plurality of lattice graphs further referencing a global planar coordinate system having an arbitrary origin;
an image capture parameter selection facility executing on the processor and selecting one or more of the plurality of image capture parameter data values from the computer memory based on an estimate of a reprojection error over the plurality of digital images; and
a calibration facility executing on the processor and calibrating the plurality of image capture parameter data values based on one or more of the selected image capture parameter data values and two or more of the plurality of image capture calibration features in two or more of the plurality of digital images and the corresponding lattice graphs.
13. The system of claim 12 , wherein the plurality of digital images includes a video containing the plurality of digital images.
14. The system of claim 12 , wherein the plurality of digital images is a series of standstill images.
15. The system of claim 12 , further integrated into an image capture device.
16. The system of claim 12 , wherein each of the nodes is based on a respective center of a respective image capture calibration feature of the plurality of image capture calibration features.
17. The system of claim 12 , wherein a least two of the plurality of digital images depict at least one of the plurality of image capture calibration features from different respective angles.
18. The system of claim 12 , wherein each corresponding lattice graph has a plurality of nodes corresponding to a portion of the plurality of image capture calibration features.
19. The system of claim 12 , wherein the calibrating comprises repeated calibrating a plurality values of at least one of the plurality of image capture parameter data values for two or more different combinations of the plurality of digital images.
20. The system of claim 12 , wherein the calibrating further comprises a plurality of recursions each starting with a single node detected in the image, one or more n-edge vectors and one or more visited directions and continuing via a repeated selection of one or more n-closest neighbors to the one or more n-edge vectors.