Method and apparatus for dimensioning box object
A method to determine the volume of a box object from the captured image of the box object. The method includes identifying a geometric mark on the box object in the captured image to find the positions of two reference points of the geometric mark. The two reference points are separated by a predetermined distance. The method also includes processing a group of parameters and a predetermined mapping obtained from a calibration process. The group of parameters includes the positions of the two reference points and the predetermined distance separating the two reference points.
1. A method comprising:
detecting light from a box object through an imaging lens arrangement with an imaging sensor having two-dimensional photosensitive elements to capture at least one image of the box object, wherein the box object has the shape of a rectangular solid;
identifying a geometric mark with fixed dimensions on the box object in the at least one image to find positions of at least two reference points of the geometric mark in the at least one image, wherein the at least two reference points are separated by a predetermined distance;
identifying corners of the box object in the at least one image;
determining, using a processor and based on the corners identified in the at least one image, normal vectors corresponding to orientations of faces of the box object;
determining, using the processor and based on (i) the normal vectors and (ii) the positions of the at least two reference points, positions of at least four corner points of the box object in an object space;
determining, using the processor and based on the positions of the at least four corner points in the object space, dimensions of the box object, with the box dimensions including the width, the length, and the height of the box object.
2. The method of claim 1 , wherein the geometric mark includes one of a barcode, a logo, and a shipping label.
3. The method of claim 1 , wherein the geometric mark includes two isolated point marks.
4. The method of claim 1 , further comprising:
identifying edges of the box object based on the corners identified in the at least one image, wherein determining the normal vectors is further based on the identified edges.
5. The method of claim 1 , further comprising:
finding positions of at least four reference points of the geometric mark in the at least one image.
6. The method of claim 1 , further comprising:
finding positions of at least four reference points of the geometric mark in the at least one image; and
finding positions of at least four reference points of another geometric mark in the at least one image.
7. A method comprising:
detecting light from a box object through an imaging lens arrangement with an imaging sensor having two-dimensional photosensitive elements to capture at least one image of the box object, wherein the box object has the shape of a rectangular solid;
identifying at least two reference points on a surface of the box object to find positions of the at least two reference points in the at least one image, wherein the at least two reference points are two fixed points that are stationary in a coordinate fixed on the box object and the at least two reference points are separated by a predetermined distance on the surface of the box object, the predetermined distance being independent of geometric orientation of the box object;
identifying corners of the box object in the at least one image;
determining, using a processor and based on the corners of the box object, normal vectors indicative of face orientations corresponding to surfaces of the box object;
determining, using the processor and based on (i) the normal vectors, (ii) the positions of the at least two reference points, and (iii) the predetermined distance separating the at least two reference points, positions of at least four corner points of the box object in an object space;
determining, using the processor, dimensions of the box object based on the positions of the at least four corner points in the object space, with the box dimensions including the width, the length, and the height of the box object.
8. The method of claim 7 , further comprising:
determining a mapping in a calibration process.
9. The method of claim 8 , wherein the mapping is operative to establish a one-to-one corresponding relationship between a point in an image plane of the imaging sensor and a point in a reference plane in physical space.
10. The method of claim 8 , wherein the mapping is a matrix that models intrinsic parameters of an imaging system that includes the imaging lens arrangement and the imaging sensor.
11. The method of claim 7 , further comprising:
determining a volume of the box object based on the dimensions.
12. The method of claim 7 , further comprising:
displaying the width, the length, and the height of the box object.
13. The method of claim 7 , further comprising:
displaying at least one of the width, the length, and the height of the box object.
14. The method of claim 7 , further comprising:
identifying at least six corners of the box object to find the positions of the at least six corners in the at least one image.
15. The method of claim 7 , further comprising:
processing multiple images to identify the corners of the box object.
16. The method of claim 7 , further comprising:
identifying edges of the box object based on the corners identified in the at least one image, wherein determining the normal vectors is further based on the identified edges.
17. The method of claim 7 , further comprising:
processing the at least one image to identify at least one of the at least two reference points.
18. The method of claim 7 , further comprising:
processing the at least one image to identify all of the at least two reference points.
19. The method of claim 7 , further comprising:
displaying the at least one image of the box object on a display screen; and
identifying at least one corner of the box object with a user-selected position on the display screen.
20. The method of claim 7 , further comprising:
displaying the at least one image of the box object on a display screen; and
identifying at least one of the reference points with a user-selected position on the display screen.
21. An apparatus comprising:
an imaging lens arrangement;
an imaging sensor having two-dimensional photosensitive elements configured to detect light from a box object through the imaging lens arrangement and configured to capture at least one image of the box object that has the shape of a rectangular solid;
a controller operative to decode a barcode and operative to determine dimensions of the box object by:
determining positions of the two reference points on a surface of the box object in the at least one image, the two reference points being spaced apart by a known distance;
identifying corners of the box object in the at least one image;
determining normal vectors of the box object based on positions of the corners in the at least one image;
determining positions of corners of the box in an object space based on (i) the normal vectors, (ii) the positions of the two reference points in the at least one image with the two reference points being at fixed positions in a coordinate fixed on the box object, and (iii) a predetermined mapping operative to establish a one-to-one corresponding relationship between a point in an image plane of the imaging sensor and a point in physical space.
22. The apparatus of claim 21 , comprising:
a display screen operative to display the at least one image of the box object and to display a user-selected position specifying at least one corner of the box object in the at least one image.
23. The apparatus of claim 21 , comprising:
a display screen operative to display the at least one image of the box object and to display a user-selected position specifying at least one of the two reference points in the at least one image.