Method for the automatic detection of red-eye defects in photographic image data

• US 20030044178A1
• Filed: 07/09/2002
• Published: 03/06/2003
• Est. Priority Date: 09/03/2001
• Status: Abandoned Application

• Current Assignee: Agfaphoto GmbH (Agfaphoto Holding GmbH)
• Original Assignee: Agfaphoto GmbH (Agfaphoto Holding GmbH)
• Inventors: Schindler, Hans-Georg, Taresch, Gudrun, Oberhardt, Knut, Damm, Tobias, Jacob, Friedrich
• Application Number:
  US10/192,712
• Publication Number:
  US 20030044178A1
• Time in Patent Office: Days
• US Class Current: 396/158
• CPC Class Codes: G06T 2207/30216 Redeye defect
• CPC Class Codes: G06T 5/77 Retouching; Inpainting; Scratch removal
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 40/193 Preprocessing; Feature extraction

Abstract

A method for the automatic detection of red-eye defects in photographic image data includes the step of determining a value that provides information about the presence of a flash when recording the image data, as a criterion for the presence of such defects.

Claims

1. In a method for the automatic detection of red-eye defects in photographic image data, the improvement comprising the step of determining a value that represents the presence of a flash when taking a picture and recording the image data, as a criterion for the presence of such defects.

2. Method as set forth in claim 1, wherein the determining step comprises the step of determining the presence of a flash marker in recorded auxiliary film data as an indication of whether a flash has been used when taking a picture and recording the image data.

3. Method as set forth in claim 1, wherein the determining step comprises the step of determining the presence of an artificial light photograph as an indication that a flash was not the dominant light source when taking the picture.

4. Method as set forth in claim 1, wherein the determining step comprises the step of determining the presence of a picture low in contrast as an indication that no flash has been used when taking the picture.

5. Method as set forth in claim 1, wherein the determining step comprises the step of determining the presence of a result of an image analysis as an indication of whether a flash has been used when taking the picture.

6. Method as set forth in claim 5, wherein hard shadows in the image are an indication of whether a flash has been used when taking the picture.

7. Method as set forth in claim 1, wherein a plurality of processes to determine the use of a flash are carried out independently of one another.

8. Method as set forth in claim 7, wherein said plurality of processes are carried out simultaneously.

9. Method as set forth in claim 1, wherein a decision as to the presence of red-eye defects is made, based on an overall evaluation of said value, together with other values determined as criteria for the presence of such defects.

10. Method as set forth in claim 9, wherein the values are probabilities.

11. Method as set forth in claim 9, wherein the overall evaluation is made using a neural network.

12. Method as set forth in claim 1, wherein the process for automatic detection of red-eye defects is automatically terminated if it is determined that no flash has been used.