| 摘要 |
The present invention relates to an image distortion correction apparatus. The image distortion correction apparatus according to one embodiment of the present invention comprises an automatic tolerance correction logic wherein the automatic tolerance correction logic acquires an edge image of a remaining partial image, except for a vehicle mask, from an image synthesized from images inputted from one or more cameras, calculates one or more first direction linear equations from the acquired edge image, explores intersection points included in an image of each camera from the first linear equations, does a parking space modeling from the first linear equations and the explored intersection points, predicts a corresponding point for each pixel included in an image of each camera present on the first line equations by using a difference between second linear equations constituting the modeled parking space and the first linear equation, configures each homography matrix for an image of each camera based on the predicted corresponding points for each pixel, thereafter, updates the corresponding points for the pixels during a N frame, accumulates the homography matrix based on the updated corresponding points for the pixels during the N frame, and if a margin of error of more than N frame in an image of at least one camera among cameras is greater than a threshold value, updates around view monitor table values of the relevant camera by using the accumulated homography matrix. |
