Red-eye removal using multiple recognition channels

摘要

This disclosure pertains to apparatuses, methods, and computer readable media for red-eye removal techniques using multiple recognition channels. In the following examples, red, golden, and white recognition channels are used. A recognition channel is the monochrome extraction from a color photograph in a manner designed to make one kind of red-eye artifact glow with maximum contrast. Once the red-eye artifact has been characterized by, e.g., size and location, the techniques disclosed herein may then discern whether the red-eye artifact is, for example, a red-, golden-, or white-eye case by examining the configuration and characteristics of pro prominence bitmasks created for the various recognition channels. Once the type of red-eye case has been discerned, the techniques disclosed herein may then replace the artifact with a photographically reasonable result based on the type of red-eye case being repaired. Specular reflection may also be re-added to the photograph.

主权项

1. An image processing method, comprising:
receiving user input indicating a region in an image, the region including at least a part of an artifact and wherein the image is stored in a memory; extracting a plurality of recognition channels from the indicated region of the image, wherein the plurality of recognition channels comprises at least a white color recognition channel, wherein extracting the plurality of recognition channels comprises:
deriving a plurality of prominence bitmasks with a prominence bitmask derived for each of the plurality of recognition channels; andgrowing at least one prominence bitmask of the plurality of prominence bitmasks by seed filling the at least one prominence bitmask using successively lower thresholds until the at least one prominence bitmask transitions until a breakout occurs, wherein the breakout comprises a transition from an expected shape for a pupil to an unexpected shape for a pupil; characterizing the plurality of recognition channels; identifying an artifact-type for the artifact based on the characterized plurality of recognition channels; and repairing the artifact based on the identified artifact-type.