Gemstone planning

1. A method of determining an optimal target gemstone to be obtained from a rough gemstone, the method comprising:

obtaining a first series of two-dimensional (2D) silhouette images of the rough gemstone;

providing a three-dimensional (3D) model of a target gemstone to be obtained from the rough gemstone;

generating a second series of virtual 2D silhouette images of the target gemstone from the 3D model of the target gemstone; and

comparing the first series of 2D silhouette images and the second series of virtual 2D silhouette images to determine an optimal transformation to be applied to the 3D model of the target gemstone.

2. The method of claim 1 , wherein the first series of 2D silhouette images and the second series of virtual 2D silhouette images are obtained from the same positions.

3. The method of claim 1 , further comprising correlating the first series of 2D silhouette images and the second series of virtual 2D silhouette images before comparing the first series of 2D silhouette images and the second series of virtual 2D silhouette images.

4. The method of claim 3 , wherein correlating the first series of 2D silhouette images and the second series of virtual 2D silhouette images comprises setting a geometric centre of the 3D model of the target gemstone to be coincident with an image centre of each of the first series of 2D silhouette images.

5. The method of claim 1 , wherein obtaining the first series of 2D silhouette images of the rough gemstone comprises illuminating the rough gemstone with collimated light and capturing a 2D silhouette image of the rough gemstone at each of a plurality of discrete rotational increments.

6. The method of claim 5 , further comprising rotating the rough gemstone as the first series of 2D silhouette images is obtained.

7. The method of claim 5 , further comprising obtaining a third series of 2D images of the rough gemstone under diffuse lighting, each of the third series of 2D images being captured at each of the plurality of discrete rotational increments.

8. The method of claim 1 , wherein obtaining the first series of 2D silhouette images of the rough gemstone comprises generating 2D silhouette images from a 3D model of the rough gemstone.

9. The method of claim 1 , wherein providing the 3D model of the target gemstone comprises selecting a 3D model from a plurality of 3D models of cut and faceted gemstones.

10. The method of claim 1 , wherein generating the second series of virtual 2D silhouette images of the target gemstone from the 3D model of the target gemstone comprises virtually positioning a plurality of virtual cameras around the 3D model of the target gemstone, each virtual camera configured to capture one or more 2D silhouette images of the target gemstone from a respective position of the virtual camera.

11. The method of claim 10 , further comprising positioning the plurality of virtual cameras in a geodesic arrangement around the 3D model of the target gemstone.

12. The method of claim 10 , further comprising virtually illuminating the 3D model of the target gemstone with collimated light.

13. The method of claim 1 , wherein comparing the first series of 2D silhouette images and the second series of virtual 2D silhouette images to determine an optimal transformation to be applied to the 3D model of the target gemstone comprises:

projecting each 2D silhouette image of the second series of virtual 2D silhouette images into a corresponding 2D silhouette image of the first series of 2D silhouette images, wherein the corresponding 2D silhouette images of the first series of 2D silhouette images and the second series of virtual 2D silhouette images are captured from the same positions;

determining a first scaling factor, which when applied to each of the second series of virtual 2D silhouette images produces a first scaled second series of virtual 2D silhouette images, each of the first scaled second series of virtual 2D silhouette images is too large to fit inside the corresponding 2D silhouette image of the first series of 2D silhouette images; and

determining a second scaling factor which, when applied to the each of the second series of virtual 2D silhouette images, produces a second scaled series of 2D silhouette images, each of the second scaled second series of virtual 2D silhouette images fits inside the corresponding 2D silhouette image of the first series of 2D silhouette images.

14. The method of claim 13 , wherein comparing the first series of 2D silhouette images and the second series of virtual 2D silhouette images to determine an optimal transformation to be applied to the 3D model of the target gemstone further comprises:

determining an average of the first scaling factor and the second scaling factor to produce a third scaling factor;

applying the third scaling factor to the 3D model of the target gemstone to generate a scaled 3D model;

determining combinations of translational and rotational transformations of the 3D model of the target gemstone;

for each combination of translational and rotational transformations, generating a series of transformed and scaled 2D silhouette images of the target gemstone from the scaled 3D model;

for each combination of translational and rotational transformations, projecting each 2D silhouette image of the series of transformed and scaled 2D silhouette images of the target gemstone into the corresponding 2D silhouette image of the first series of 2D silhouette images; and

iteratively increasing or decreasing the third scaling factor to identify a combination of translational and rotational transformations, wherein each of the 2D silhouette images of the series of transformed and scaled 2D silhouette images of the target gemstone fits within the corresponding 2D silhouette image of the first series of 2D silhouette images, and wherein a difference between the first scaling factor and the second scaling factor is smaller than a predetermined threshold.

15. The method of claim 14 , further comprising applying the identified combination of translational, rotational and scaling transformations to the 3D model of the target gemstone.

16. The method of claim 1 , wherein the optimal target gemstone corresponds to a largest target gemstone that is obtainable from the rough gemstone.

17. The method of claim 1 , wherein the rough gemstone is a diamond.

18. The method of claim 1 , further comprising generating a 3D model of the rough gemstone that comprises the optimal target gemstone to be obtained from the rough gemstone.

19. A method of determining an optimal target gemstone to be obtained from a rough gemstone, the method comprising:

obtaining a first series of two-dimensional (2D) silhouette images of the rough gemstone;

providing a three-dimensional (3D) model of a target gemstone to be obtained from the rough gemstone;

generating a second series of virtual 2D silhouette images of the target gemstone from the 3D model of the target gemstone;

comparing the first series of 2D silhouette images and the second series of virtual 2D silhouette images to determine an optimal transformation to be applied to the 3D model of the target gemstone; and

fitting each image of the second series of virtual 2D silhouette images into a corresponding image of the first series of 2D silhouette images to determine if the target gemstone fits into the rough gemstone.

20. The method of claim 19 , wherein corresponding images of the first series of 2D images and the second series of virtual 2D images comprise images captured from the same positions.

21. An apparatus for determining an optimal target gemstone to be obtained from a rough gemstone, the apparatus comprising:

one or more image capture devices configured to obtain a first series of two-dimensional (2D) silhouette images of the rough gemstone; and

a processor configured to:

generate a second series of virtual 2D silhouette images from a three-dimensional (3D) model of a target gemstone to be obtained from the rough gemstone; and

compare the first series of 2D silhouette images and the second series of virtual 2D silhouette images to determine an optimal transformation to be applied to the 3D model of the target gemstone.