Inspection system and method for determining three dimensional model of an object
View Patent ↗Embodiments of a system and method are described that automate processes to reconstruct models from two-dimensional images. In one embodiment, the systems comprise an apparatus that can acquire a plurality of images of an object and automatedly form a boundary about the object in each of the images. The boundary highlights a portion of the image that will form the basis for the model. The apparatus can also select a primary image from the images, wherein the boundary in the primary image is narrowest relative to an axis that is common to each of the images. In one example, the apparatus can select a secondary image, which is perpendicular to the primary image, and reconstruct the primary image and the second image to generate the model.
1. An inspection system, comprising:
an apparatus comprising a processor, a memory coupled to the processor, and one more executable instructions stored on the memory and configured to be executed by the processor, the one or more executable instructions comprising instructions for:
acquiring a plurality of images of an object;
segmenting one or more of the images to obtain segmented images comprising predominantly pixels that define the object in the segmented images;
placing a boundary about the object in the segmented images;
selecting a primary image from the segmented images in which the boundary is narrowest relative to an axis;
selecting a secondary image from the segmented images that is offset from the primary image by an offset angle;
generating a model from the boundaries in the primary image and the secondary image; and
reconstructing a volume representing the object in the model from the plurality of images of the object.
2. The inspection system of claim 1 , further comprising instructions for distinguishing between pixels in the image based on grey scales, wherein the pixels that define the object have pixels with darker grey scales relative to pixels that define an area surrounding the object in the image.
3. The inspection system of claim 1 , further comprising instructions for distinguishing between pixels in the image based on grey scales, wherein the pixels that define the object have pixels with lighter grey scales relative to pixels that define an area surrounding the object in the image.
4. The inspection system of claim 1 , wherein the primary image and the secondary image are perpendicular to each other.
5. The inspection system of claim 1 , wherein the images show the object at angles from 0° to 360°.
6. The inspection system of claim 1 , wherein the boundary circumscribes an area that is the smallest area in which the object fits.
7. The inspection system of claim 1 , further comprising an external server coupled to the apparatus, wherein one or more of the executable instructions are stored on the external server.
8. The inspection system of claim 1 , further comprising a scan device coupled to the apparatus, wherein the scan device acquires the image of the object.
9. The inspection system of claim 8 , wherein the scan device comprises an x-ray scanner.
10. The inspection system of claim 8 , wherein the scan device comprises a computed-tomography scanner.
11. The inspection system of claim 1 , further comprising obtaining the images using forward projection of a previously generated volume.
12. A method for generating a three-dimensional model, said method comprising steps for:
acquiring a plurality of images of an object;
segmenting one or more of the images to obtain segmented images comprising predominantly pixels that define the object in the image;
placing a boundary about the object in the segmented images;
selecting a primary image from the segmented images in which the boundary is narrowest relative to an axis;
selecting a secondary image from the segmented image that is offset from the primary image by an offset angle; and
generating a model from the boundaries in the primary image and the secondary image; and
reconstructing a volume representing the object in the model from the plurality of images of the object.
13. The method of claim 12 , further comprising a step for distinguishing between pixels in the image based on grey scales.
14. The method of claim 12 , wherein the boundary circumscribes an area that is the smallest area in which the object fits.
15. The method of claim 12 , further comprising a step for acquiring the images using an x-ray scanner.
16. The method of claim 12 , further comprising a step for acquiring the images using a computed-tomography scanner.
17. The method of claim 12 , wherein the primary image is perpendicular to the secondary image.
18. A non-transitory computer-readable medium comprising executable instructions for generating a three-dimensional model, the executable instructions comprising instructions for:
acquiring a plurality of images of an object;
segmenting one or more of the images to obtain segmented images comprising predominantly pixels that define the object in the image;
placing a boundary about the object in the segmented images;
selecting a primary image from the segmented images in which the boundary is narrowest relative to an axis;
selecting a secondary image from the segmented image that is offset from the primary image by an offset angle;
generating a model from the boundaries in primary image and the secondary image;
reconstructing a volume representing the object in the model from the plurality of images of the object.
19. The non-transitory computer-readable medium of claim 18 , wherein the boundary circumscribes an area that is the smallest area in which the object fits.
20. The non-transitory computer-readable medium of claim 18 , further comprising a step for distinguishing between pixels in the image based on grey scales.