Color-based foreign object detection system

Patent metadata

Field	Value
Publication number	US-9541507-B2
Application number	US-201414469125-A
Country	US
Kind code	B2
Filing date	Aug 26, 2014
Priority date	Aug 26, 2014
Publication date	Jan 10, 2017
Grant date	Jan 10, 2017

Abstract

Systems and methods are provided for detecting foreign objects in or on a subject of interest. A first set of images is captured of a sample subject known to be free of foreign objects. A plurality of colors from a defined color palette are classified according to a color content of the first set of images into at least first and second classes. A second set of images of the subject of interest are captured. It is determined that a foreign object is present in or on the subject of interest if a color from the first class is present in the second set of images.

First claim

Having described the invention, we claim:

1. A method for detecting foreign objects in or on a subject of interest, the method comprising:
capturing a first set of images of a sample subject known to be free of foreign objects;
classifying a plurality of colors from a defined color palette according to a color content of the first set of images into at least first and second classes;
capturing a second set of images of the subject of interest; and
determining if a foreign object is present in or on the subject of interest if a color from the first class is present in the second set of images.

2. The method of claim 1 , wherein classifying a plurality of colors from a defined color palette according to a color content of the first set of images comprises classifying the plurality of colors via a naïve Bayesian classification.

3. The method of claim 1 , wherein classifying a plurality of colors from a defined color palette according to a color content of the first set of images comprises assigning all colors of the plurality of colors appearing in the first set of images to the second class and all other colors of the plurality of colors to the first class.

4. The method of claim 1 , further comprising indicating a location of a foreign object to a user if the foreign object is determined to be present.

5. The method of claim 4 , further comprising:
receiving feedback from the user that the foreign object determined to be present is not a foreign object;
determining a set of colors associated with the foreign object; and
transferring the determined set of colors from the first class to the second class.

6. The method of claim 4 , wherein indicating a location of a foreign object to a user comprises displaying a given image of the second set of images to the user with the location of the foreign object displayed in color and a remainder of the image displayed in grayscale.

7. The method of claim 4 , wherein indicating a location of a foreign object to a user comprises displaying a given image of the second set of images to the user with the location of the foreign object highlighted in a preselected color.

8. The method of claim 4 , wherein indicating a location of a foreign object to a user comprises displaying a text message to a user.

9. A method for detecting foreign objects in or on a subject of interest, the method comprising:
capturing a plurality of intensity values for each pixel of a first set of images of a sample subject known to be free of foreign objects, each of the plurality of intensity values being associated with a specific wavelength of visible, infrared, or ultraviolet light a first set of images;
quantizing each of the plurality of intensity values to a desired resolution, such that a defined color palette according to the desired resolution and a number of intensity values in the plurality of intensity values for each pixel;
classifying a plurality of colors from the defined color palette according to a color content of the first set of images into at least first and second classes;
capturing a second set of images of the subject of interest; and
determining if a foreign object is present in or on the subject of interest if a color from the first class is present in the second set of images.

10. The method of claim 1 , wherein the subject of interest is the interior of an airplane, and the sample subject is a portion of the interior of the aircraft that has been previous inspected and found to be clear of foreign objects.

11. A system for detecting foreign objects in or on a subject of interest comprising:
an imaging system;
a system control configured to instruct the imaging system to capture a first set of images of a sample subject known to be free of foreign objects; and
a color selection component configured to select a plurality of colors from a defined color palette according to a color content of the first set of images;
wherein the system control is further configured to instruct the imaging system to capture a second set of images of the subject of interest and notify a user if a selected color is present in the second set of images.

12. The system of claim 11 , wherein the imaging system, the system control, the color selection component, and the display are collectively implemented as a wearable computer with an optical head-mounted display.

13. The system of claim 11 , wherein the color selection component is configured to select the plurality of colors from the defined color palette according to a color content of the first set of images by classifying the plurality of colors via a naïve Bayesian classification.

14. The system of claim 11 , wherein the color selection component is configured to select all colors appearing the first set of images.

15. The system of claim 11 , further comprising a user interface configured to receive one of a confirmation that the presence of the selected color represents a foreign object in or on the subject of interest and an indication that the presence of the selected color does not represent a foreign object.

16. The system of claim 15 , wherein the selected color present in the second set of images is removed from the selected plurality of colors in response to the indication that the presence of the selected color does not represent a foreign object.

17. The system of claim 11 , wherein a system control is configured to instruct the imaging system to capture a plurality of intensity values for each pixel of the first set of images, each of the plurality of intensity values being associated with a specific wavelength of visible, infrared, or ultraviolet light, and quantize each of the plurality of intensity values to a desired resolution, such that the defined color palette is defined according to the desired resolution and a number of intensity values in the plurality of intensity values for each pixel.

18. A non-transitory computer readable medium comprising:
a system control configured to instruct an imaging system to capture a first set of images of a sample subject known to be free of foreign objects, the system control comprising an imager interface configured to receive the captured first set of images and a user interface configured to receive an input from a user; and
a color selection component configured to classify a plurality of colors from a defined color palette according to a color content of the first set of images into at least first and second classes;
wherein the system control is further configured to instruct the imaging system to capture a second set of images of a subject of interest, notify a user when color from the first class is present in the second set of images, and receive an input from the user via the user interface indicating if the color represents a foreign object in the subject of interest.

19. The non-transitory computer readable medium of claim 18 , wherein the color selection component is configured to assign all colors of the plurality of colors appearing in the first set of images to the second class and all other colors of the plurality of colors to the first class.

20. The non-transitory computer readable medium of claim 19 , wherein, in response to an input for the user indicating that a color from the first class is not associated with a foreign object, the system control is configured to remove the color from the first class and add it to the second class.

Assignees

• Engel Stephen J
• Christ Robert J
• Madsen John
• Nardiello Jerrell A
• Cobb Skylar Blaiz
• Shaffar Scott W
• Northrop Grumman Systems Corp

Inventors

• Engel Stephen J
• Christ Robert J
• Madsen John
• Nardiello Jerrell A
• Cobb Skylar Blaiz
• Shaffar Scott W

Classifications

• G06V10/56 Primary

  relating to colour · CPC title

• G01N21/94 Primary

  Investigating contamination, e.g. dust (G01N21/85 takes precedence) · CPC title

• G01N21/8851 Primary

  Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges (G01N21/8806 and G01N21/93 - G01N21/95692 take precedence; optical measurement of dimensions G01B11/00; optical scanning G02B26/10; image transformation G06T3/00; computerised image enhancement G06T5/00; image processing per se for flaw detection G06T7/0002) · CPC title

• G06K9/4652

  Physics · mapped topic

• G06K9/00664

  Physics · mapped topic