Perceptual segmentation of images

摘要

Method and system embodiments of the present invention are directed to automated identification of regions within digitally-encoded images that correspond to objects and features of scenes captured in the digitally-encoded images, a process referred to as ‘perceptual segmentation’ of an image. Regions or segments within an image are first identified by any of various region-identifying or segmentation methods. For each region or segment, features of pixels within the region or segment are employed to compute one or more segment features. The segment features are used, in turn, to identify the region or segment as belonging to a particular type of region or segment, and the region is then accordingly labeled or tagged as a region or segment of the determined type.

主权项

1. A perceptual-segmentation system, comprising:
a computer; and one or more perceptual-segmentation programs, stored within the computer and executed by the computer, that
identify regions within an image;for each identified region,
compute or access each of one or more pixel-feature values for pixels within region,using the one or more pixel-feature values computed or accessed for pixels within the region, compute one or more region-feature values, andusing the one or more region-feature values, determine a region type for the region and associating the region type with the region; andstore the associations of region types with regions for use in subsequent image-processing tasks; wherein the one or more perceptual-segmentation programs, stored within the computer and executed by the computer, identify regions within the image by:
for each of a number of specific object-identifying methods,
applying the object-identifying method to locate occurrences of objects of a specific type in the image; anddisambiguating any regions of overlap of two or more objects identified by the object-identifying methods.