Image processing devices that utilize built-in dynamic camera models to support rapid determination of camera intrinsics and methods of operating same
Image processing devices and methods according to embodiments of the invention advantageously utilize computationally efficient and high speed operations based on dynamic camera models to support rapid determination of multiple camera intrinsics. These built-in dynamic models can utilize varying focus distance values as inputs (i.e., operands) to generate respective intrinsic parameters as outputs, as a stream of images are captured and processed in real time to possibly determine pose and position of a three-dimensional (3D) object during 3 image scanning.
1. A method of operating a digital camera, comprising:
generating a plurality of image frames of an object at respective focus distances relative to the digital camera; and
determining for each of the plurality of image frames and within the digital camera, at least one intrinsic characteristic of the digital camera using a model of the digital camera that maps the focus distance of the corresponding image frame to the at least one intrinsic characteristic; and
wherein the model of the digital camera is stored in memory within the digital camera so that said determining can be performed in real time as the plurality of image frames are generated;
wherein the model comprises a first model that maps focus distances associated with the plurality of image frames to effective focal lengths of the digital camera, and a second model that maps the effective focal lengths to the at least one intrinsic characteristic;
wherein the first model is in the form of a first linear relationship as: y 1 =m 1 x 1 +b 1 , where y 1 designates the effective focal length, x 1 designates the focus distance, m 1 designates a slope of the first linear relationship and b 1 is a first constant; and
wherein the second model is in the form of a second linear relationship as: y 2 =m 2 y 1 b 2 where y 2 designates the distortion parameter, v 1 designates the effective focal length, m 2 designates a slope of the second linear relationship and b 2 is a second constant.
2. The method of claim 1 , wherein the effective focal lengths determined for the plurality of image frames are unequal.
3. The method of claim 1 , wherein the digital camera comprises a lens having a nominal focal length; wherein the effective focal lengths determined for at least some of the plurality of image frames are unequal to the nominal focal length; and wherein at least one of the first and second models is a linear model.
4. The method of claim 1 , wherein each of the plurality of image frames is generated when a fixed or zoom lens within the digital camera is set to a nominal focal length; and wherein the effective focal lengths determined for at least some of the plurality of image frames are unequal to the nominal focal length.
5. The method of claim 1 , wherein the at least one intrinsic characteristic is selected from a group consisting of a radial distortion parameter, a tangential/skewing distortion parameter and a principal point parameter.
6. An image processing device, comprising:
a microprocessor; and
a memory having image processing program code therein that is readable by the microprocessor and executable therein, said image processing program code comprising microprocessor-readable program code configured to cause the image processing device to:
generate a plurality of image frames of an object at respective focus distances relative to the image processing device; and
determine, for each of the plurality of image frames, at least one intrinsic characteristic of the image processing device using a model of the image processing device that maps the focus distance of the corresponding image frame to the at least one intrinsic characteristic, said model comprising a first model that maps focus distances associated with the plurality of image frames to effective focal lengths of the digital camera, and a second model that maps the effective focal lengths to the at least one intrinsic characteristic;
wherein the first model is in the form of a first linear relationship as: y 1 =m 1 x 1 +where v 1 designates the effective focal length, x 1 designates the focus distance, m 1 designates a slope of the first linear relationship and b 1 is a first constant; and
wherein the second model is in the form of a second linear relationship as: y 2 =m 2 y 1 +b 2 , where y 2 designates the distortion parameter, y 1 designates the effective focal length, m 2 designates a slope of the second linear relationship and b 2 is a second constant.
7. The device of claim 6 , wherein each of the plurality of image frames is generated when a lens within the image processing device is set to a nominal focal length; and wherein the effective focal lengths determined for at least some of the plurality of image frames are unequal to the nominal focal length.
8. The device of claim 6 , wherein the at least one intrinsic characteristic is selected from a group consisting of a radial distortion parameter, a tangential distortion parameter and a principal point parameter; and wherein at least one of the first and second models is a linear model.
9. A handheld cellular device, comprising:
a body having a digital camera therein with a fixed or variable focal length lens, an image sensor optically coupled to the lens, a microprocessor and at least one memory having image processing program code therein that is readable by the microprocessor and executable therein, said microprocessor and said image processing program code collectively configured to:
generate a plurality of image frames of an object at respective focus distances relative to the handheld cellular device; and
determine, for each of the plurality of image frames, at least one intrinsic characteristic of the digital camera using a model of the digital camera that maps the focus distance of the corresponding image frame to the at least one intrinsic characteristic, said model comprising a first model that maps focus distances associated with the plurality of image frames to effective focal lengths of the digital camera, and a second model that maps the effective focal lengths to the at least one intrinsic characteristic;
wherein the first model is in the form of a first linear relationship as: v 1 =m 1 x 1 +b 1 , where y 1 designates the effective focal length, x 1 designates the focus distance, m 1 designates a slope of the first linear relationship and b 1 is a first constant; and
wherein the second model is in the form of a second linear relationship as: y 2 =m 2 y 1 b 2 where y 2 designates the distortion parameter, v 1 designates the effective focal length, m 2 designates a slope of the second linear relationship and b 2 is a second constant.
10. The device of claim 9 , wherein each of the plurality of image frames is generated when the lens is set to a nominal focal length; and wherein the effective focal lengths determined for at least some of the plurality of image frames are unequal to the nominal focal length.
11. The device of claim 9 , wherein the model of the digital camera is stored in memory within the handheld cellular device so that operations to determine the at least one intrinsic characteristic can be performed in real time as the plurality of image frames are generated; and wherein at least one of the first and second models is a linear model.