Non-parametric method of and system for estimating dimensions of objects of arbitrary shape
A non-parametric method of, and system for, dimensioning an object of arbitrary shape, captures a three-dimensional (3D) point cloud of data points over a field of view containing the object and a base surface on which the object is positioned, detects a base plane indicative of the base surface from the point cloud, extracts the data points of the object from the point cloud, processes the extracted data points of the object to obtain a convex hull, and fits a bounding box of minimum volume to enclose the convex hull. The bounding box has a pair of mutually orthogonal planar faces, and the fitting is performed by orienting one of the faces to be generally perpendicular to the base plane, and by simultaneously orienting the other of the faces to be generally parallel to the base plane.
1. A non-parametric method of dimensioning an object of arbitrary shape, the method comprising:
capturing, via an image sensor, a three-dimensional (3D) point cloud of data points over a field of view containing the object and a base surface on which the object is positioned;
detecting, using a logic circuit, a base plane indicative of the base surface from the point cloud;
extracting, using the logic circuit, first ones of the data points corresponding to the object from the point cloud;
obtaining, using the logic circuit, a convex hull for the extracted first ones of the data points;
generating, using the logic circuit and based on data defining the convex hull, a bounding box of minimum volume to enclose the convex hull, the bounding box having first and second mutually orthogonal planar faces;
constraining the first planar face to be generally perpendicular to the base plane; and
constraining the second planar face to be generally parallel to the base plane.
2. The method of claim 1 , wherein each of the data points includes a length coordinate, a width coordinate, and a depth coordinate.
3. The method of claim 1 , wherein the detecting of the base plane is performed by determining a plane having a largest area in the field of view.
4. The method of claim 1 , wherein the detecting of the base plane is performed by executing a random sampling consensus (RANSAC) algorithm.
5. The method of claim 1 , wherein the extracting of the first ones of the data points corresponding to the object is performed by removing second ones of the data points corresponding to the base plane.
6. The method of claim 1 , further comprising clustering data points indicative of multiple objects in the field of view, and wherein the extracting is performed by selecting one of the multiple objects as the object to be dimensioned.
7. The method of claim 6 , wherein the clustering of the data points is performed by Euclidean clustering.
8. The method of claim 1 , wherein the bounding box is a cuboid having three pairs of mutually orthogonal planar faces.
9. A non-parametric system for dimensioning an object of arbitrary shape, the non-parametric system comprising:
an imaging device to generate a three-dimensional point cloud of data points over a field of view containing the object and a base surface on which the object is positioned; and
a controller to:
detect a base plane indicative of the base surface from the point cloud;
extract first ones of the data points corresponding to the object from the point cloud;
obtain a convex hull for the extracted first ones of the data points;
generate, based on data defining the convex hull, a bounding box of minimum volume to enclose the convex hull, the bounding box having first and second mutually orthogonal planar faces;
constrain the first planar face to be generally perpendicular to the base plane; and
constrain the second planar face to be generally parallel to the base plane.
10. The system of claim 9 , wherein each of the data points includes data indicative of a length coordinate, a width coordinate, and a depth coordinate.
11. The system of claim 9 , wherein the controller is operative to determine the base plane by identifying a plane having a largest area in the field of view.
12. The system of claim 9 , wherein the controller is operative to determine the base plane by executing a random sampling consensus (RANSAC) algorithm.
13. The system of claim 9 , wherein the controller is operative to extract the first ones of the data points corresponding to the object by removing second ones of the data points corresponding to the base plane.
14. The system of claim 9 , wherein the controller is operative to:
cluster data points indicative of multiple objects in the field of view; and
respond to an input in which at least one of the objects to be dimensioned is selected.
15. The system of claim 14 , wherein the controller is operative to cluster the data points by Euclidean clustering.
16. The system of claim 9 , wherein the bounding box is a cuboid having three pairs of mutually orthogonal planar faces.