AUTOMATIC RECTIFICATION OF DISTORTIONS IN IMAGES

摘要

Implementations relate to relate to rectification of distortion in an image. In some implementations, a method includes extracting edgelets from an image, each edgelet defined by a location of a pixel having an edge depicted in the image and defined by a direction of the edge. The method finds at least one vanishing point in the image based on intersections of pairs of the edges of the edgelets. Each vanishing point is selected based on a plurality of the edgelets in the image having edges approximately aligned with the vanishing point. Found vanishing points are used to transform the image to reduce image distortion.

主权项

1. A method comprising:
extracting a plurality of edgelets from an image, each edgelet defined by a location of a pixel having an edge depicted in the image and each edgelet defined by a direction of the edge; finding at least one vanishing point in the image based on intersections of pairs of the edges of the edgelets, wherein the at least one vanishing point is selected based on a plurality of the edgelets in the image having edges approximately aligned with the at least one vanishing point; and using the at least one vanishing point to transform the image to reduce distortion in the image.