Systems and methods for analyzing image data using adaptive neighborhooding
First Claim
Patent Images
1. A method for segmenting an image comprising:
- storing, in a computer database, a set of image data representing a plurality of pixel locations within an image;
associating a respective neighborhood with each of the plurality of pixel locations in the stored image data;
identifying, by a computer, a plurality of segments within the image by iteratively performing the processes of;
individually adjusting an orientation and at least one of a shape and a size of a plurality of the neighborhoods;
for each pixel location and associated neighborhood, obtaining a smoothed data value for the pixel location based on data values of the other pixel locations within the associated neighborhood; and
segmenting the image using the image data, the obtained smoothed data values, and the plurality of adjusted neighborhoods; and
outputting, by the computer, the plurality of segments within the image identified by the iterative process.
1 Assignment
0 Petitions
Accused Products
Abstract
The invention relates to methods and systems for extracting information about a scene from a set of image data by interdependently smoothing the data, segmenting the imaged scene into coherent elements by determining edges, and estimating attributes present within the scene. These methods and systems include attribute estimation, adaptive neighborhood adjustment and preferential use of different images or imaging modalities for information extraction.
97 Citations
36 Claims
-
1. A method for segmenting an image comprising:
-
storing, in a computer database, a set of image data representing a plurality of pixel locations within an image; associating a respective neighborhood with each of the plurality of pixel locations in the stored image data; identifying, by a computer, a plurality of segments within the image by iteratively performing the processes of; individually adjusting an orientation and at least one of a shape and a size of a plurality of the neighborhoods; for each pixel location and associated neighborhood, obtaining a smoothed data value for the pixel location based on data values of the other pixel locations within the associated neighborhood; and segmenting the image using the image data, the obtained smoothed data values, and the plurality of adjusted neighborhoods; and outputting, by the computer, the plurality of segments within the image identified by the iterative process. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18)
-
-
19. A system for segmenting an image, the system comprising:
-
a memory; computerized image data representing a plurality of pixel locations within an image stored on the memory; a processor configured to; retrieve from memory a subset of the computerized image data; associate a respective neighborhood with each of the plurality of pixel locations in the stored image data; identify a plurality of segments within the image by iteratively performing the processes of; individually adjusting an orientation and at least one of a shape and a size of a plurality of the neighborhoods; for each pixel location and associated neighborhood, obtaining a smoothed data value for the pixel location based on data values of the other pixel locations within the associated neighborhood; and segmenting the image using the image data, the obtained smoothed data values, and the plurality of adjusted neighborhoods; and output the plurality of segments within the imaged scene identified by the iterative process. - View Dependent Claims (20, 21, 22, 23, 24, 25, 26, 27)
-
-
28. A non-transitory computer readable medium storing computer executable instructions, which, when executed by a processor, cause the processor to carry out a method for segmenting an image comprising:
-
storing by the processor, in a computer database, a set of image data representing a plurality of pixel locations within an image; associating a respective neighborhood with each of the plurality of pixel locations in the stored image data; identifying by the processor a plurality of segments within the image by iteratively performing the processes of; individually adjusting an orientation and at least one of a shape and a size of a plurality of the neighborhoods; for each pixel location and associated neighborhood, obtaining a smoothed data value for the pixel location based on data values of the other pixel locations within the associated neighborhood; and segmenting the image using the image data, the obtained smoothed data values, and the plurality of adjusted neighborhoods; and outputting by the processor the plurality of segments within the image identified by the iterative process. - View Dependent Claims (29, 30, 31, 32, 33, 34, 35, 36)
-
Specification