AUTOMATIC EXTRACTION OF PLANETARY IMAGE FEATURES

• US 20110026832A1
• Filed: 05/19/2010
• Published: 02/03/2011
• Est. Priority Date: 05/20/2009
• Status: Active Grant

• Current Assignee: United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration
• Original Assignee: United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration
• Inventors: Benediktsson, Jon A., LeMoigne-Stewart, Jacqueline J., Moses, Gabriele, Troglio, Giulia, Serpico, Sebastiano B.
• Granted Patent: US 8,355,579 B2
• Time in Patent Office: Days
• US Class Current: 382/199
• CPC Class Codes: G06T 2207/10032 Satellite or aerial image; Remote sensing
• CPC Class Codes: G06T 2207/10048 Infrared image
• CPC Class Codes: G06T 2207/20061 Hough transform
• CPC Class Codes: G06T 2207/20152 Watershed segmentation
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/30 Determination of transform parameters for the alignment of images, i.e. image registration
• CPC Class Codes: G06V 10/48 by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation

Abstract

A method for the extraction of Lunar data and/or planetary features is provided. The feature extraction method can include one or more image processing techniques, including, but not limited to, a watershed segmentation and/or the generalized Hough Transform. According to some embodiments, the feature extraction method can include extracting features, such as, small rocks. According to some embodiments, small rocks can be extracted by applying a watershed segmentation algorithm to the Canny gradient. According to some embodiments, applying a watershed segmentation algorithm to the Canny gradient can allow regions that appear as close contours in the gradient to be segmented.

Claims

1. A feature extraction method, the method comprising:

computing an image gradient of a Lunar image by using an edge detector;

detecting features in the image gradient by applying a Hough Transform to the image gradient; and

at least one of mapping the features to generate a map, printing the features to generate a printout, and using the detected features to register the Lunar image and generate a registered Lunar image.

2. The method of claim 1, the method further comprising:

applying a watershed segmentation algorithm to the image gradient and registering the Lunar image against an image of an object to extract an object feature of the Lunar image.

3. The method of claim 2, wherein the features detected are ridges and the applying a Hough Transform comprises using a standard Hough accumulator.

4. The method of claim 1, wherein the edge detector comprises a Canny edge detector.

5. The method of claim 1, wherein the method comprises using the detected features to register the image and generate a registered image.

6. The method of claim 2, the method further comprising:

preprocessing the Lunar image and the image of an object to smooth noise in the pair of images prior to computing the image gradient.

7. The method of claim 1, wherein the registering comprises registering the image to a reference image.

8. A feature extraction method, the method comprising:

computing an image gradient of an image by using an edge detector;

applying a watershed segmentation algorithm to the image gradient to extract first objects;

applying a generalized Hough Transform to detect ellipses in the image gradient;

applying a watershed segmentation algorithm to the image gradient using the detected ellipses as seed points, to extract objects and recesses of elliptical shape;

detecting ridges in the image gradient by applying a standard Hough Transform to detect straight lines in the image gradient; and

distinguishing one or more objects in the image from a background of the image.

9. The method of claim 8, wherein the edge detector comprises a Canny edge detector.

10. The method of claim 8, wherein the image comprises a Lunar image.

11. The method of claim 8, the method further comprising:

preprocessing the image to smooth noise in the image prior to computing the image gradient.

12. The method of claim 8, the method further comprising:

registering the image with one or more reference images.

13. The method of claim 12, wherein the registering comprises extracting features from the reference image and computing a geometric transformation between the image and the reference image.

14. The method of claim 12, wherein the registering comprises:

extracting features from the reference image;

matching features between the image and the reference image;

estimating a transformation model;

transforming the image with respect to the reference image; and

generating a registered image.

15. The method of claim 14, wherein a processor is used to carry out one or more of the extracting, the matching, the estimating, the transforming, and the generating.

16. The method of claim 14, wherein a processor is used to carry out each of the extracting, the matching, the estimating, the transforming, and the generating.

17. The method of claim 14, the method further comprising:

printing out the registered image.

18. The method of claim 14, the method further comprising:

displaying the registered image.

19. An image feature extraction system, the system comprising:

a processor configured to

compute an image gradient of an image by using an edge detector, apply a watershed segmentation algorithm to the image gradient to extract first objects,

apply a generalized Hough Transform to detect ellipses in the image gradient,

apply a watershed segmentation algorithm to the image gradient using the detected ellipses as seed points, to extract objects and recesses of elliptical shape,

detect ridges in the image gradient by applying a Hough Transform to detect straight lines in the image gradient, and

distinguish one or more objects in the image from a background of the image.

20. The system of claim 19, the system further comprising:

a Canny edge detector, wherein the processor is configured to compute the image gradient of the image using the Canny edge detector.

21. The system of claim 19, the system further comprising:

a camera configured to acquire the image.

22. The system of claim 19, the system further comprising:

a preprocessor configured to preprocess the image to smooth noise in the image.