Method and apparatus for automatic identification of linear objects in an image

• US 7,672,484 B2
• Filed: 03/10/2008
• Issued: 03/02/2010
• Est. Priority Date: 07/18/2003
• Status: Expired due to Fees

• Current Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Original Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Inventors: Skoblick, Richard, Phurrough, Larry, Wiedemann, Melissa, Gratsch, Colleen Flynn, Lee, Harry C.
• Primary Examiner(s): Mehta; Bhavesh M
• Assistant Examiner(s): Rashid; David P
• Application Number:
  US12/073,777
• Publication Number:
  US 20080219555A1
• Time in Patent Office: 722 Days
• Field of Search: 382/103
• US Class Current: 382/103
• CPC Class Codes: G06V 20/13 Satellite images
• CPC Class Codes: G06V 30/2504 Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches

Abstract

A method and system for processing image data to identify objects in an image. The method and system operate using various resolutions of the image to identify the objects. Information obtained while processing the image at one resolution is employed when processing the image at another resolution.

Claims

1. A method of identifying linear objects in an image comprising:

receiving a first image with a first resolution, the first image containing data representing physical linear objects;

generating a filtered image from the first image, the filtered image identifying potential objects which have a smaller radius than the size of a filter and a different brightness than pixels surrounding the potential objects;

receiving a second image identifying regions in the first image with the first resolution which are not to be processed;

generating a third image by removing regions in the filtered image which are identified in the second image as regions in the first image which are not to be processed;

identifying lines in the third image;

generating a fourth image by removing lines identified in the third image which do not meet predetermined criteria; and

identifying linear objects in the first image using the remaining lines in the fourth image.

2. The method of claim 1, wherein the step of generating a filtered image comprises:

performing a series of dilations and erosions of the first image to produce a spatially filtered image; and

subtracting said spatially filtered image from the first image to produce the filtered image.

3. The method of claim 1 wherein the identified regions in the second image are regions of the first image with the first resolution which contain either clouds or bodies of water.

4. The method of claim 1, wherein the step of identifying lines in the third image comprises:

performing a Hough transform on the third image to identify a number of pixels which lie on a line with a particular angle and location.

5. The method of claim 1, wherein the step of generating a fourth image comprises:

eliminating identified lines containing less than a predetermined number of pixels.

6. The method of claim 5, wherein the step of generating a fourth image further comprises:

determining a mean and standard deviation of intensities for portions of the first image on either side of the remaining lines; and

eliminating a line from the remaining lines if the mean and standard deviation of intensities of the portion of the first image on one side of the line is more than a predetermined amount different from the mean and standard deviation of intensities on the other side of the line.

15. A computer-readable recording medium having a computer program recorded thereon that causes a computer to carry out the method of claim 1.

7. A method of identifying linear objects in an image comprising:

receiving an input image with a first resolution, the input image containing data representing physical linear objects;

processing the input image to produce a first image at a second resolution;

generating a filtered image from the first image at the second resolution;

receiving a second image identifying portions of the input image with the first resolution which are not to be processed;

generating a third image by removing portions of the filtered image which are identified in the second image as portions of the input image which are not to be processed;

identifying lines in the third image;

generating a fourth image by removing lines identified in the third image which do not meet predetermined criteria; and

identifying linear objects in the input image using the remaining lines in the fourth image.

16. A computer-readable recording medium having a computer program recorded thereon that causes a computer to carry out the method of claim 7.

8. A method for identifying linear objects in an image comprising:

receiving an input image;

receiving a first image identifying linear objects representing first physical structures in the input image and a second image identifying linear objects representing second physical structures in the input image, the first image having a first resolution and the second image having a second resolution different from the first resolution;

processing the first image and the second image to produce a third image, the processing comprising;

combining the linear objects of the first image with the linear objects of the second image,

connecting at least two linear objects among the combined linear objects that are not connected to one another, based on a predetermined positional relationship between the at least two linear objects, to combine respective lengths of the at least two linear objects to obtain at least one extended linear object having the combined lengths of the at least two linear objects, and

producing the third image to include the combined linear objects and the at least one extended linear object; and

identifying the linear objects in the input image using the combined linear objects and the at least one extended linear object included in the third image.

9. The method of claim 8, further comprising:

receiving another image identifying linear objects representing third physical structures in the input image, the another image having a third resolution different from the first and second resolutions, wherein the processing step processes the another image when producing the third image by combining the linear objects of the first image, the linear objects of the second image, and the linear objects of the another image to produce the third image.

10. The method of claim 8, wherein the processing of the first image and the second image to produce the third image comprises:

determining a pixel-to-area ratio of each extended linear object;

determining if any of the at least one extended linear objects represents clutter based on the determined pixel-to-area ratio of the any of the at least one extended linear objects, respectively, and

wherein the producing of the third image comprises producing the third image by excluding the determined any one of the at least one extended linear objects that represent clutter.

11. The method of claim 8, wherein the processing of the first image and the second image to produce the third image comprises:

determining a respective best path of each extended linear object, each best path being the longest and most direct path between any two end points in the extended linear object, respectively; and

replacing each extended linear object in the third image with each respective determined best path so that the linear objects in the input image are identified using at least one best path included in the third image.

12. The method of claim 11, wherein the processing of the first image and the second image to produce the third image comprises:

determining a respective number of angles for each best path greater than a predetermined angle;

calculating, for each best path, respectively, a ratio of the number of angles of the best path to a length of the best path; and

determining whether each best path includes a zig-zag line segment based on the calculated respective ratio of the number of angles to the length of each best path, and

wherein the producing of the third image comprises producing the third image by excluding each best path that includes the zig-zag line segment.

13. The method of claim 8, wherein the connecting of the at least two linear objects comprises connecting at least one linear object of the first image with at least one linear object of the second image having the predetermined positional relationship with the at least one linear object of the first image.

14. The method of claim 8, wherein the predetermined positional relationship is determined based on at least one of an angle between the at least two linear objects, a distance between the at least two linear objects, and an intersection of the at least two linear objects.

17. A computer-readable recording medium having a computer program recorded thereon that causes a computer to carry out the method of claim 8.