Method and system to segment depth images and to detect shapes in three-dimensionally acquired data
First Claim
1. A method for use in tracking movement of at least one object in a series of images, comprising:
- (a) receiving multiple images including image depth data from an array of pixels acquired by three-dimensional imaging system, each image including z-depth data that comprises a z-depth value of each pixel from at least a fraction of optical energy emitted by the imaging system reflected by at least one object, said image depth data generated by comparing emitted optical energy and the fraction of emitted optical energy detected by said system, and for each of the multiple images;
(a1) examining at least some of said image depth data captured during a single data acquisition;
(a2) using a depth value histogram that records the z-depth value of each pixel of one or more regions, forming segments by grouping together those of said pixels whose acquired said data captured at step (a1) have at least one common characteristic selected from a group consisting of depth distance, depth distance change, object size, and object shape, and assigning pixels in the group a single image cluster center having a depth group value, where the single image cluster center is the geometric center of said group of pixels;
(a3) assigning on a per-segment basis a label to pixels comprising each of said segments formed in step (a2), the pixels so labeled enabling recognition of said at least one object with image depth data; and
(b) tracking motion of the at least one object relative to the system.
1 Assignment
0 Petitions
Accused Products
Abstract
A method and system analyzes data acquired by image systems to more rapidly identify objects of interest in the data. In one embodiment, z-depth data are segmented such that neighboring image pixels having similar z-depths are given a common label. Blobs, or groups of pixels with a same label, may be defined to correspond to different objects. Blobs preferably are modeled as primitives to more rapidly identify objects in the acquired image. In some embodiments, a modified connected component analysis is carried out where image pixels are pre-grouped into regions of different depth values preferably using a depth value histogram. The histogram is divided into regions and image cluster centers are determined. A depth group value image containing blobs is obtained, with each pixel being assigned to one of the depth groups.
90 Citations
20 Claims
-
1. A method for use in tracking movement of at least one object in a series of images, comprising:
-
(a) receiving multiple images including image depth data from an array of pixels acquired by three-dimensional imaging system, each image including z-depth data that comprises a z-depth value of each pixel from at least a fraction of optical energy emitted by the imaging system reflected by at least one object, said image depth data generated by comparing emitted optical energy and the fraction of emitted optical energy detected by said system, and for each of the multiple images; (a1) examining at least some of said image depth data captured during a single data acquisition; (a2) using a depth value histogram that records the z-depth value of each pixel of one or more regions, forming segments by grouping together those of said pixels whose acquired said data captured at step (a1) have at least one common characteristic selected from a group consisting of depth distance, depth distance change, object size, and object shape, and assigning pixels in the group a single image cluster center having a depth group value, where the single image cluster center is the geometric center of said group of pixels; (a3) assigning on a per-segment basis a label to pixels comprising each of said segments formed in step (a2), the pixels so labeled enabling recognition of said at least one object with image depth data; and (b) tracking motion of the at least one object relative to the system. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
-
8. An image-analyzing system for tracking objects based on image data, the system including:
-
a three-dimensional imaging system having an emitter of optical energy, an array of pixels receiving at least a fraction of the emitted optical energy reflected by at least one object; a depth image data acquisition system operable to generate said image depth data by comparing emitted optical energy and detected optical energy by said array of pixels; and a processor and memory, the memory including code operable to instruct the processor to track movement of the at least one object by analyzing multiple images, the analyzing including, for each image; (a) examining at least some of said image depth data captured during a single data acquisition; (b) forming segments by grouping pixels in each image having a depth distance within a threshold distance from each other and assigning pixels in the group a single image cluster center having a depth group value, where the single image cluster center is the geometric center of said group of pixels, and wherein each pixel in the group is assigned the depth group value of the image cluster center; (c) assigning on a per-segment basis a label to pixels comprising each of said segments formed to enable recognition of said at least one object with a single acquisition of image depth data; and (d) for the multiple images, tracking movement of said at least one object by noting movement of commonly labeled pixels. - View Dependent Claims (9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. A computer readable hardware storage device including instructions causing a microprocessor to track movement of at least one object in a series of images, the instructions comprising:
-
(a) receiving multiple images including image depth data acquired by three-dimensional imaging system, image depth data comprising a z-depth value of each of a plurality of pixels, and for each of the multiple images; (a1) examining at least some of said image depth data; (a2) using a depth value histogram of one or more regions, wherein the depth value histogram records the z-depth values of pixels in the one or more regions, forming segments by grouping together those of pixels in the image data whose acquired said data captured at step (a) have at least one common characteristic selected from a group consisting of depth distance, object size, and object shape, and assigning pixels in the group a single image cluster center having a depth group value, where the single image cluster center is the geometric center of said group of pixels, and wherein each pixel in the group is assigned the depth group value of the image cluster center; (a3) assigning on a per-segment basis a label to pixels comprising each of said segments formed in step (a2), the pixels so labeled enabling recognition of said at least one object with an acquisition of image depth data; and (b) tracking motion of the at least one object relative to the system over the multiple images. - View Dependent Claims (18, 19, 20)
-
Specification