Systems and methods for forming models of three-dimensional objects
A 3D imaging system is proposed in which an object is successively illuminated in at least three directions and at least three images of the object are captured by one or more energy sensors. A set of images is produced computationally showing the object from multiple viewpoints, and illuminated in the at least three directions simultaneously. This set of images is used stereoscopically to form an initial 3D model of the object. Variations in the brightness of the object provides features useful in the stereoscopy. The initial model is refined using photometric data obtained from images in which the object is illuminated in the at least three directions successively.
1. An apparatus for computing a three-dimensional (3D) model of an object, comprising:
at least one directional energy source arranged to directionally illuminate the object from at least three directions;
one or more energy sensors arranged to capture images of the object from multiple viewpoints;
a processor arranged to receive images captured by the energy sensors, and control the at least one energy source and the energy sensors;
the processor being arranged to control the at least one energy source and one or more energy sensors to capture for each energy sensor, and for each said direction, at least one corresponding single-direction image which shows the object illuminated in the said direction;
the processor further being arranged to combine corresponding ones of said single-direction images computationally to produce combined-direction images for each of the viewpoints, the combined-direction images each showing the object illuminated in a plurality of the directions, the processor being arranged to combine the single-direction images by:
(i) using the single-direction images from each of the viewpoints, to produce, for each of a plurality of combination algorithms, a respective set of candidate combined-direction images from each of the viewpoints;
(ii) obtaining a quality measure for each set of candidate combined-direction images;
(iii) identifying the set of candidate combined-direction images for which the quality measure is highest; and
(iv) using the identified set of candidate combined-direction image as the combined-direction images;
the processor further being arranged:
to analyze the combined-direction images stereoscopically to obtain an initial model of the object;
to obtain photometric data from the single-direction images; and
to refine the initial model using the photometric data.
2. An apparatus according to claim 1 in which the combined-direction images have an intensity, for each viewpoint and in each pixel, which is a function of the respective intensities of the respective corresponding pixels of the single-direction images taken from the same viewpoint.
3. An apparatus according to claim 2 in which the function is selected from the group comprising:
the mean of the intensities;
the maximum of the intensities; and
the median of the intensities.
4. An apparatus according to claim 1 comprising at least one directional energy source located proximate to one of the energy sensors.
5. An apparatus according to claim 1 in which each directional energy source includes one or more energy generation elements and one or more collimating elements for collimating energy generated by the energy generation elements.
6. A method for computing a three-dimensional (3D) model of an object using one or more energy sensors arranged to capture images of the object from multiple viewpoints, the method comprising:
directionally illuminating the object from at least three directions, capturing for each energy sensor at least one corresponding single-direction image which shows the object illuminated in the corresponding direction;
computationally combining corresponding ones of said single-direction images, to producing for each energy sensor combined-direction images, the combined-direction images each showing the object illuminated in the plurality of directions, said combining the single-direction images being performed by:
(i) using the single-direction images from each of the viewpoints, to produce, for each of a plurality of combination algorithms, a respective set of candidate combined-direction images from each of the viewpoints;
(ii) obtaining a quality measure for each set of candidate combined-direction images;
(iii) identifying the set of candidate combined-direction images for which the quality measure is highest; and
(iv) using the identified set of candidate combined-direction images as the combined-direction images;
analyzing the combined-direction images stereoscopically to obtain an initial model of the object;
obtaining photometric data from the single-direction images; and
refining the initial model using the photometric data.
7. A method according to claim 6 in which the combined-direction images have an intensity, for each viewpoint and in each pixel, which is a function of the respective intensities of the respective corresponding pixels of the single-direction images captured from the same viewpoint.
8. A method according to claim 7 in which the function is selected from the group comprising:
the mean of the intensities;
the maximum of the intensities; and
the median of the intensities.
9. A method according to claim 6 in which each directional energy source includes one or more energy generation elements and one or more collimating elements for collimating energy generated by the energy generation elements.
10. A method according to claim 6 comprising positioning at least one of the energy sources relative to the object with the energy generated by the at least one energy source incident on at least one point the object at an angle to a tangent of the object at that point which is no more than 20 degrees.