Method for analyzing images and for correcting the values of video signals

• US 6,683,982 B1
• Filed: 11/28/2000
• Issued: 01/27/2004
• Est. Priority Date: 03/23/1998
• Status: Expired due to Fees

• Current Assignee: Heidelberger Druckmaschinen Ag
• Original Assignee: Heidelberger Druckmaschinen Ag
• Inventors: Kohn, Michael
• Primary Examiner(s): Tran, Phuoc
• Application Number:
  US09/646,818
• Time in Patent Office: 1,155 Days
• Field of Search: 382/162, 382/167, 382/168, 382/172, 382/274, 348/672, 358/518, 358/521, 358/522, 345/604
• US Class Current: 382/167
• CPC Class Codes: H04N 9/11 Scanning of colour motion picture films, e.g. for telecine

Abstract

In a method for image analysis and correction of video signal values in the processing of video films, whereby analysis values with respect to image range, gradation and color cast of the individual frames of the video film are identified by the image analysis of the video signal values, and the correction of the video signal values is implemented on the basis of the analysis values, the video signal values are initially transformed according to a three-dimensional conversion function into the video signal values of an independent and sensation-conforming, equidistant color space. For each frame of a video sequence composed of a plurality of frames, at least one analysis value is determined on the basis of the transformed video signal values. At least one analysis values representative of the video sequence is formed by forming the average of the corresponding analysis values of the frames of the video sequence.

Claims

1. A method for image analysis and correction of video signal values in processing of a video film composed of video sequences, each video sequence being composed of a plurality of frames, comprising the steps of:

transforming video signal values of a first color space according to a three-dimensional conversion function into video signal values of a second color space that is independent of the first color space and is equidistant and conforming to sensation;

forming at least one analysis value representative of a video sequence by forming an average of analysis values of analyzed frames of the video sequence;

respectively determining a three-dimensional correction function from the analysis values representative of the video sequence; and

correcting at least the video signal values of the corresponding video sequence according to the identified correction function.

2. The method according to claim 1 wherein

all frames of the video sequence are utilized for the image analysis of the video sequence; and

at least one analysis value representative of the video sequence is formed by forming an average of the analysis values of the frames of the video sequence.

3. The method according to claim 1 wherein

frames of the video sequence are selected for the image analysis;

and

at least one analysis value representative of the video sequence is formed by forming an average of the analysis values of the selected frames of the video sequence.

4. The method according to claim 1 wherein the correction function is an ICC color profile.

5. The method according to claim 1 wherein in the analysis of the image range of a frame,

the frequency distribution of the brightness component of the transformed video signal values is identified in the frame;

and

as analysis values, an image light value of the frame and an image dark value of the frame are determined from the frequency distribution by defining maximum and minimum brightness values dependent on the curve of the frequency distribution in an image light region and an image dark region.

6. The method according to claim 1 wherein in the analysis of the image gradation of a frame,

the frame to be analyzed is geometrically divided into a plurality of sub-images;

a frequency distribution of brightness components of the video signal values in the corresponding sub-image is defined as a sub-image histogram for each sub-image;

the sub-image histograms of the individual sub-images are interpreted and the sub-images that are image-important for the image gradation are identified on the basis of the evaluation;

an aggregate histogram is calculated from the sub-image histograms of the image-important sub-images, said aggregate histogram corresponding to the brightness distribution of the brightness component of the image values in the image-important sub-images; and

a correction curve for correction of an image gradation characteristic of the frame is determined as an analysis value from the aggregate histogram according to a method of histogram modification, which is determined for the purpose of contrast correction.

7. The method according to claim 1 wherein in the analysis of a color cast of a frame,

for acquiring a color cast in the frame, dividing a region of high brightness values and a region of low brightness values of the brightness component of the video signal values in the frame to be analyzed into at least one brightness range;

determining a value of a potentially existing color cast in the respective brightness range by averaging the color components of the video signal values in the brightness ranges; and

forming the color cast value for the evaluation of the color cast analysis as an analysis value by selecting or combining color cast values determined for the individual brightness ranges.

8. A method for image analysis and correction of video signal values in processing of a video film composed of video sequences, each video sequence being composed of a plurality of frames, comprising the steps of:

transforming video signal values of a first color space according to a three-dimensional conversion function into video signal values of a second color space different than the first color space and conforming to sensation;

forming at least one analysis value representative of a video sequence by forming a representative value of analysis values of analyzed frames of the video sequence;

respectively determining a three-dimensional correction function from the analysis values representative of the video sequence; and

correcting at least the video signal values of the corresponding video sequence according to the identified correction function.