Everyone is not a trained photographer but you can still click great pics. Google patents a cool technique for automatic rectification of distortions in images

PatentYogi_9,064,309_Automatic-rectification-of-distortions-in-images

The popularity and convenience of digital cameras as well as the widespread of use of Internet communications have caused digital images to become ubiquitous. For example, user-produced digital photographs are posted to various Internet sites, such as web pages, social networking services, etc. for users and others to view. However, many images, such as photographs taken by a camera, include various forms of image distortion. In some cases, such distortion can be caused involuntary rotations of a camera capturing the images. In one example, an image depicts a scene having parallel linear features but these features are shown in the image as distorted non-parallel features.

Google’s patent discloses a method that extracts 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 then 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. Finally, the found vanishing points are used to transform the image to reduce image distortion.

Patent Information:

Patent Number : 9,064,309
Patent Title : Automatic rectification of distortions in images
Inventors: Chaudhury; Krishnendu (Saratoga, CA), DiVerdi; Stephen J. (Oakland, CA)
Assignee: Google Inc. (Mountain View, CA)
Family ID: 1000001172816
Appl. No.: 14/066,633
Filed: October 29, 2013
Abstract
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.