Method for red green blue (RGB) stereo sensor fusion

• US 6,597,807 B1
• Filed: 09/27/1999
• Issued: 07/22/2003
• Est. Priority Date: 09/27/1999
• Status: Expired due to Fees

• Current Assignee: United States Of America As Represented By The Secretary Of The Army (Government of the United States of America), United States Secretary of the Army
• Original Assignee: The United States Of America As Represented By The Secretary Of The Army
• Inventors: Jordan, Jay B., Watkins, Wendell R.
• Primary Examiner(s): Au, Amelia M.
• Assistant Examiner(s): LAROSE, COLIN M
• Application Number:
  US09/408,655
• Time in Patent Office: 1,394 Days
• Field of Search: 382/154, 382/164, 348/47, 348/48, 348/32, 348/33, 348/44, 348/490, 348/29, 348/60, 356/12
• US Class Current: 382/164
• CPC Class Codes: G02B 30/34 Stereoscopes providing a stereoscopic pair of separated images corresponding to parallactically displaced views of the same object, e.g. 3D slide viewers
• CPC Class Codes: G06T 2207/10012 Stereo images
• CPC Class Codes: G06T 5/50 using two or more images, e.g. averaging or subtraction
• CPC Class Codes: H04N 13/15 for colour aspects of image signals

Abstract

Red-Green-Blue color-coding of overlaid registered stereo pair imagery from different spectral regions (for example, near-, mid-, and far-infrared) can be used to derive depth perception information. The human preattentive vision process uses coarse spatial structure in the scene to fuse the images for the detection of depth and motion cues. Optimization is accomplished by assigning green to the lowest spatial frequency content image, then red to the next lowest, and finally blue. If the fusion of one of the colored imagery layers produces depth perception information in one portion of the scene whereas the other two colored layers do not, the human mind can retain the information from this color and discard the inputs from the other two colors as noise. The same is true for the other two colors. Human color discrimination and depth perception are combined to greatly enhance preattentive object recognition in the fused stereo pair imagery.

Claims

1. A method for processing signals using preattentive human vision from a plurality of sensors representing different spectral ranges of the same three dimensional scene, comprising:

obtaining first left and right stereo signals from a pair of left and right sensors sensitive in a first spectral range, said first spectral range having a spatial frequency content lower than any other spectral range;

assigning a first Red-Green-Blue (RGB) color code green, for which humans are most sensitive, to the first left and right stereo signals to enhance object recognition and perception;

obtaining second left and right stereo signals from a pair of left and right sensors sensitive in a second spectral range, said second spectral range having the next lower spatial frequency content;

assigning a second Red-Green-Blue (RGB) color code red for which humans are next most sensitive, to the second left and right stereo signals to enhance object recognition and perception.

2. A method according to claim 1 wherein the first left and right sensor pairs are sensitive in the infrared range.

3. A method according to claim 2 wherein the second left and right sensor pairs are sensitive in the visible range.

4. A method according to claim 1 further comprising:

obtaining signals from third left and right sensor pairs sensitive in a third spectral range different from the first and second spectral ranges, said third spectral range having the next lowest spatial frequency content, and assigning a third Red-Green-Blue (RGB) color code blue, for which humans are least sensitive, to the third left and right stereo signals to enhance object recognition and perception.

5. A method according to claim 4 wherein the first, second, and third left and right sensor pairs are sensitive in the near, mid, and far infrared ranges, respectively.

6. A method of displaying fused sensory data as color stereo images for viewing using preattentive human vision, the data obtained from stereo pair sensors, comprising:

obtaining first left and right stereo signals from a pair of left and right sensors sensitive in a first spectral range, said first spectral range having a spatial frequency content lower than any other spectral range;

assigning a first Red-Green-Blue (RGB) color code green, for which humans are most sensitive, to the first left and right stereo signals to enhance object recognition and perception;

obtaining second left and right stereo signals from a pair of left and right sensors sensitive in a second spectral range, said second spectral range having the next lower spatial frequency content;

assigning a second Red-Green-Blue (RGB) color code red for which humans are next most sensitive, to the second left and right stereo signals to enhance object recognition and perception; and

displaying the RGB coded data in a color stereoscopic image.

7. A method according to claim 6 further comprising:

overlaying the RGB coded signals on a display so that periodically the RGB sensor fused information ramps in overall intensity from zero to a maximum percentage.

8. A method according to claim 6 further comprising:

overlaying the RGB coded signals on a display so that the RGB sensor fused information is adjustable in overall intensity from zero to a maximum percentage.