Combining two-dimensional images with depth data to detect junctions or edges
A method, system, apparatus, article of manufacture, and computer program product provide the ability to detect junctions. 3D pixel image data is obtained/acquired based on 2D image data and depth data. Within a given window over the 3D pixel image data, for each of the pixels within the window, an equation for a plane passing through the pixel is determined/computed. For all of the determined planes within the given window, an intersection of all of the planes is computed. A spectrum of the intersection/matrix is analyzed. Based on the spectrum, a determination is made if the pixel at the intersection is of 3 or more surfaces, 2 surfaces, or is 1 surface.
1. A computer-implemented method for detecting a feature in three-dimensional (3D) space, comprising:
obtaining three-dimensional (3D) pixel image data based on two-dimensional (2D) image data and depth data for the 2D image data;
within a given window over the 3D pixel image data, for each of one or more pixels within the given window, determining an equation for a plane passing through the pixel;
computing, for all of the determined planes within the given window, an intersection of all of the planes;
analyzing a spectrum of the intersection;
based on the spectrum, determining eigenvalues that determine a number of surfaces that intersect at the pixel wherein:
one and only one zero eigenvalue corresponds to a junction of three (3) or more surfaces;
two (2) zero eigenvalues correspond to a crease at the intersection of two (2) surfaces; and
three (3) zero eigenvalues correspond to one (1) planar surface; and
creating and displaying a computer-aided design (CAD) drawing that depicts the number of surfaces that intersect at the pixel.
2. The computer-implemented method of claim 1 , wherein the image data comprises color or intensity data.
3. The computer-implemented method of claim 1 , wherein the image data comprises point cloud data.
4. The computer-implemented method of claim 1 , wherein the 3D pixel image data is obtained by:
multiplying a 3×3 matrix representing 2D normal equations for the 2D image data by a 3×4 matrix representing a projection relation between the 2D image data and the depth data.
5. The computer-implemented method of claim 1 , wherein the one or more pixels within the given window comprise those pixels for which a gradient value along an image gradient is maximal.
6. The computer-implemented method of claim 1 , wherein the intersection comprises a least squares intersection.
7. The computer-implemented method of claim 1 , wherein the intersection is computed using a 4×4 matrix.
8. A computer-implemented system for detecting a feature in a computer system comprising:
(a) a computer having a processor;
(b) an image acquiring device, communicatively coupled to the computer, that acquires three-dimensional (3D) pixel image data based on two-dimensional (2D) image data and depth data, from an depth sensing device, for the 2D image data;
(c) an application, executing on the computer, wherein the application is configured to:
(1) within a given window over the 3D pixel image data, for each of one or more pixels within the given window, determine an equation for a plane passing through the pixel;
(2) compute, for all of the determined planes within the given window, an intersection of all of the planes;
(3) analyzing a spectrum of the intersection;
(4) based on the spectrum, determining eigenvalues that determine a number of surfaces that intersect at the pixel wherein:
(i) one and only one zero eigenvalue corresponds to a junction of three (3) or more surfaces;
(ii) two (2) zero eigenvalues correspond to a crease at the intersection of two (2) surfaces; and
(iii) three (3) zero eigenvalues correspond to one (1) planar surface; and
(5) creating and displaying a computer-aided design (CAD) drawing that depicts the number of surfaces that intersect at the pixel.
9. The computer-implemented system of claim 8 , wherein the image data comprises color or intensity data.
10. The computer-implemented system of claim 8 , wherein the image data comprises point cloud data.
11. The computer-implemented system of claim 8 , wherein the image acquiring device is configured to acquire the 3D pixel image data by:
multiplying a 3×3 matrix representing 2D normal equations for the 2D image data by a 3×4 matrix representing a projection relation between the 2D image data and the depth data.
12. The computer-implemented system of claim 8 , wherein the one or more pixels within the given window comprise those pixels for which a gradient value along an image gradient is maximal.
13. The computer-implemented system of claim 8 , wherein the intersection comprises a least squares intersection.
14. The computer-implemented system of claim 8 , wherein the intersection is computed using a 4×4 matrix.