Digital camera and method

• US 7,903,870 B1
• Filed: 02/22/2007
• Issued: 03/08/2011
• Est. Priority Date: 02/24/2006
• Status: Active Grant

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: Budagavi, Madukar
• Primary Examiner(s): Le; Brian Q
• Assistant Examiner(s): BEKELE, MEKONEN T
• Application Number:
  US11/677,649
• Time in Patent Office: 1,475 Days
• Field of Search: 382/167, 382/165, 382/117, 382/274, 382/275, 382/103, 382/118, 382/119
• US Class Current: 382/162
• CPC Class Codes: G06F 18/2148 characterised by the process organisation or structure, e.g. boosting cascade
• CPC Class Codes: G06T 2207/30216 Redeye defect
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06V 10/446 using Haar-like filters, e.g. using integral image techniques
• CPC Class Codes: G06V 10/7747 Organisation of the process, e.g. bagging or boosting
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation
• CPC Class Codes: G06V 40/193 Preprocessing; Feature extraction

Abstract

Redeye removal methods detect redeyes in a two-step procedure. In the first step it detects faces in the input image and in the second step it searches for redeyes in the detected faces. The methods include the introduction of an efficient skin tone detector that reduces the average complexity of a Viola-Jones based face. We also introduce a summed area table (SAT) based optimization in our eye detector stage that reduces the complexity in the eye detector by a factor of 5× when compared to an optimized direct search method.

Claims

1. A redeye detection method of a processor, comprising the steps of:

(a) providing a red plane of a portion of a face image, wherein the face is detected by a two pass process via a processor;

(b) blob enhancement filtering of said red plane in the processor, said filtering using sum of area tables for said red plane;

(c) thresholding said blob-enhancement-filtered red plane;

wherein the thresholding blob-enhancement-filter utilizes the equation wherein f_B(x,y) is the output of the blob enhancement filter, and f_{B, MAX} and f_{B, MIN} are the maximum and the minimum values of f_B(x,y), respectively

2. The redeye detection method of claim 1, wherein the two pass process comprises a first pass utilizing at least one square window and a second pass utilizing at least one rectangular scanning window.

3. The redeye detection method of claim 2, where in the rectangular scanning window comprises a width to height ration of 1.15:

1.

4. The redeye detection method of claim 1, wherein the thresholding is performed in the first pass.

5. The redeye detection method of claim 1, wherein when the first pass does not detect red eyes, the thresholding utilizes the equation wherein f_B(x,y) is the output of the blob enhancement filter, and f_{B, MAX} and f_{B, MIN} are the maximum and the minimum values of f_B(x,y), respectively