Magna Electronics Inc.

Daan He

Nikhil Gupta

Hilda Faraji

Ghanshyam Rathi

A vehicular vision system includes a camera disposed at a vehicle and operable to capture multiple frames of image data during a driving maneuver of the vehicle. A control includes an image processor that processes frames of captured image data to determine feature points in an image frame when the vehicle is operated within a first range of steering angles, and to determine motion trajectories of those feature points in subsequent image frames for the respective range of steering angles. The control determines a horizon line based on the determined motion trajectories. Responsive to determination that the determined horizon line is non-parallel to the horizontal axis of the image plane, at least one of pitch, roll or yaw of the camera is adjusted. Image data captured by the camera is processed at the control for object detection.

Vehicular vision system that dynamically calibrates a vehicular camera

A method for dynamically calibrating a vehicular camera includes disposing a camera at a vehicle and operating the camera to capture multiple frames of image data while the vehicle is in motion and is steered within at least two ranges of steering angles. Feature points are determined in an image frame when the vehicle is steered within a respective range of steering angles, and motion trajectories of those feature points are tracked in subsequent image frames for the respective range of steering angles. A horizon line is determined based on the tracked feature points. Responsive to determination that the determined horizon line is non-parallel to the horizontal axis of the image plane, at least one of pitch, roll or yaw of the camera is adjusted. Image data captured by the camera is processed at the control for object detection.

Method for dynamically calibrating a vehicular camera

A method for calibrating a vehicular camera includes providing at least a front or rear camera and a side camera with overlapping fields of view, and calibrating the front or rear camera, capturing a calibrated frame of image data with the front or rear camera, and capturing a sideward frame of image data with a side camera. At least one feature is determined present in the overlapping region of the calibrated frame, and pixel positions of the determined feature are predicted for the side camera. Misalignment of the side camera is determined based on a comparison of the predicted pixel positions of the determined feature to the pixel positions of the determined feature in the sideward frame of image data captured by the side camera. Processing of image data captured by the side camera is adjusted to accommodate the determined misalignment.

Method and system for calibrating vehicular cameras

A method for dynamically ascertaining alignment of a vehicular camera relative to a vehicle to which the camera is attached includes determining a plurality of steering angle ranges for the vehicle, each of which is a range of steering angles that approximates straight vehicle motion over less than two seconds of travel time of the vehicle while the vehicle is in motion and turning. Image data captured by a camera of the vehicle is processed to determine a central vanishing point when the vehicle is in motion and moving straight. A plurality of feature points are selected in the image frames for each steering angle range, and a vanishing point for a plurality of tracked motion trajectories for each steering angle range is determined. An alignment of the camera is determined based at least in part on the determined central vanishing point and a determined vanishing line.

Method and system for dynamically ascertaining alignment of vehicular cameras

A method for dynamically calibrating without manual intervention a forward viewing vehicular camera with respect to its three rotational degrees of freedom includes disposing a camera at a vehicle and operating the camera to acquire multiple frames of image data as the vehicle is moving and steered through a plurality of steering angles. Feature points are determined in an image frame when the vehicle is steered within a respective range of steering angles and motion trajectories of those feature points are tracked in subsequent image frames for the respective range of steering angles. A vanishing point is established in the image plane for the feature points for the respective range of steering angles. Based on established vanishing points, a vanishing line is determined. When the vanishing line is determined to be non-horizontal, at least one of pitch, roll or yaw of the camera is adjusted.

Method for dynamically calibrating vehicular cameras

A method for dynamically correcting misalignment of a vehicular camera includes fixedly disposing a camera at a vehicle and operating the camera to acquire multiple frames of image data while the vehicle is moving generally in a straight line. A plurality of sets of feature points are selected in an image frame, with each set including a first feature point and a second feature point. For each set of feature points, a motion trajectory of that set's feature points is tracked in subsequent image frames. For each tracked first and second feature points, a vanishing point is established in the image plane. Based on the established vanishing point, a vanishing line is determined in the image plane. When the vanishing line is determined to be non-horizontal in the image plane, at least one of pitch, roll or yaw of the camera is adjusted to correct rotational misalignment of the camera.

A method of dynamically calibrating a given camera relative to a reference camera of a vehicle includes identifying an overlapping region in an image frame provided by the given camera and an image frame provided by the reference camera and selecting at least a portion of an object in the overlapped region of the reference image frame. Expected pixel positions of the selected object portion in the given image frame is determined based on the location of the selected object portion in the reference image frame, and pixel positions of the selected object portion are located as detected in the given image frame. An alignment of the given camera is determined based on a comparison of the pixel positions of the selected object portion in the given image frame to the expected pixel positions of the selected object portion in the given image frame.

Method and system for dynamically calibrating vehicular cameras

A method of dynamically ascertaining the alignment of a vehicular camera. The method involves acquiring a sequence of images provided by the camera whilst the vehicle is in motion. For each range of steering angles, the method (i) selects a plurality of feature points in the images, (ii) tracks a motion trajectory for each selected feature point, and (iii) determines a vanishing point in the image plane based on the tracked motion trajectories. The method determines a vanishing line in the image plane based on a locus of these vanishing points and determines the alignment of the camera based on the position of a central vanishing point (corresponding to the zero degree angle) and the vanishing line.

Growing community of inventors

Toronto, Canada

Daan He