Three-dimensional measurement apparatus, processing method, and non-transitory computer-readable storage medium

Abstract

A three-dimensional measurement apparatus selects points corresponding to geometric features of a three-dimensional shape model of a target object, projects a plurality of selected points corresponding to the geometric features onto a range image based on approximate values indicating the position and orientation of the target object and imaging parameters at the time of imaging of the range image, searches regions of predetermined ranges respectively from the plurality of projected points for geometric features on the range image which correspond to the geometric features of the three-dimensional shape model, and associates these geometric features with each other. The apparatus then calculates the position and orientation of the target object using differences of distances on a three-dimensional space between the geometric features of the three-dimensional shape model and those on the range image, which are associated with each other.

8 . A three-dimensional measurement apparatus comprising:
 a processor; and   a memory having stored therein a program which, when executed by the processor, causes the apparatus to:   (1) acquire (a) a range image in which pixels represent distances from a target object and (b) a two-dimensional image representing a shape of the target object;   (2) acquire approximate values of a position and an orientation of the target object;   (3) (a) associate a plurality of geometric features on the range image with a plurality of first geometric features of a three-dimensional shape model and (b) associate a plurality of geometric features on the two-dimensional image with a plurality of second geometric features of the three-dimensional shape model, based on the approximate values of the position and the orientation of the target object; and   (4) derive a position and an orientation of the target object based on results of (a) the association between the plurality of geometric features on the range image and the plurality of first geometric features of the three-dimensional shape model and (b) the association between the plurality of geometric features on the two-dimensional image and the plurality of second geometric features of the three-dimensional shape model.   
     
     
         9 . The apparatus according to  claim 8 , wherein a normal direction to a geometric feature of the three-dimensional shape model is compared with a normal direction to a geometric feature detected from a predetermined range of a projected point on the range image, and when the two normal directions are different beyond a predetermined value, association processing of the plurality of geometric features is skipped. 
     
     
         10 . The apparatus according to  claim 8 , wherein the two-dimensional image is captured from the same viewpoint as the range image, and
 wherein edges of the target object are detected from the two-dimensional image as the geometric features of the two-dimensional image, and association processing between an edge which is not included in the geometric features on the range image and a corresponding geometric feature of the three-dimensional shape model upon execution of the association processing between the detected edges and the geometric features of the three-dimensional shape model is skipped.   
     
     
         11 . The apparatus according to  claim 8 , wherein the geometric features of the three-dimensional shape model are expressed by points or line segments. 
     
     
         12 . A processing method for a three-dimensional measurement apparatus for calculating a position and an orientation of a target object, the method comprising:
 a step of acquiring (a) a range image in which pixels represent distances from a target object and (b) a two-dimensional image representing a shape of the target object;   a step of acquiring approximate values of a position and an orientation of the target object;   (a) associating a plurality of geometric features on the range image with a plurality of first geometric features of the three-dimensional shape model (b) associating a plurality of geometric features on the two-dimensional image with a plurality of second geometric features of the three-dimensional shape model, based on the approximate values of the position and the orientation of the target object; and   a step of deriving a position and an orientation of the target object based on results of (a) the association between the plurality of geometric features on the range image and the plurality of first geometric features of the three-dimensional shape model and (b) the association between the plurality of geometric features on the two-dimensional image and the plurality of second geometric features of the three-dimensional shape model.   
     
     
         13 . A non-transitory computer-readable storage medium storing a computer program for controlling a computer incorporated in a three-dimensional measurement apparatus for calculating a position and an orientation of a target object, to cause the computer to:
 acquire (a) a range image in which pixels represent distances from a target object and (b) a two-dimensional image representing a shape of the target object;   acquire approximate values of a position and an orientation of the target object;   (a) associate a plurality of geometric features on the range image with a plurality of first geometric features of the three-dimensional shape model and (b) associate a plurality of geometric features on the two-dimensional image with a plurality of second geometric features of the three-dimensional shape model, based on the approximate values of the position and the orientation of the target object; and   derive a position and an orientation of the target object based on results of (a) the association between the plurality of geometric features on the range image and the plurality of first geometric features of the three-dimensional shape model and (b) the association between the plurality of geometric features on the two-dimensional image and the plurality of second geometric features of the three-dimensional shape model.