Systems and methods for 3D image distification
摘要
說明書
Because the 2D matrix points of the 2D image matrix 402 do not have a depth value (e.g., z-value), it is desirable, in certain embodiments, to determine a depth coordinate from the point cloud 406 of the 3D image 404 and associate that depth coordinate with one or more 2D matrix points. For example, 2D matrix point (X17, Y3) is directly mapped (414) to a point in 3D image 404. However, 3D point 464 resides within a 3D space defined by the four 2D matrix points (X17, Y3), (X18, Y3), (X17, Y4), and (X18, Y4), and, therefore is not directly mapped to 2D matrix point (X17, Y3). In one embodiment, a Distification method, as part of its normalization process, can determine a nearest 2D matrix point by analyzing the horizontal and vertical coordinates of 3D point 464 (i.e., a 3D coordinate pair) and then finding the finding the 2D matrix point on the 2D image matrix 402 that has horizontal and vertical coordinates (i.e., a 2D coordinate pair) with the least distance (nearest distance) to the 3D coordinate pair when measured in the 2D plane of the 2D image matrix 402. For example, if it is determined that 3D point 464 has a 3D coordinate pair that is nearest to the 2D coordinate pair of the 2D matrix point (X17, Y3), then the depth coordinate (z-value) of 3D point 464 could be associated with 2D matrix point (X17, Y3). As describe herein, in certain embodiments, a distance value (470), such as a Euclidean distance value, may be also generated for the distance or space between the 2D matrix point (X17, Y3) and the 3D point 464.
