Granted Patent B2

US 12,340,531 · App. 17/910,271 · Granted Jun 24, 2025

Inventors: Ilias Kalisperakis (Athens, GR); Lazaros Grammatikopoulos (Athens, GR)

Assignee: OTOLIFT TRAPLIFTEN B.V.

G06T7/60G06T7/13B66B9/08G06T2207/10028G06T2207/20021G06T2207/30242

Abstract

Claims (55)

1. A method for the extraction of information about a staircase, wherein use is made of:

a 3D scanning device which is arranged to obtain point clouds of surfaces of objects; and

a computer device comprising:

a central processing unit loaded with a computer program;

a digital memory connected to the central processing unit;

input means connected to the central processing unit; and

output means connected to the central processing unit;

wherein the method comprises:

scanning the staircase with the 3D scanning device to obtain a point cloud of the surface of the staircase, wherein the point cloud includes points corresponding to substantially horizontal surfaces of respective steps of the staircase;

calculating the estimated normal vector of at least a part of the points of the point cloud, including at least a part of the points that correspond to the substantially horizontal surfaces of the steps of the staircase;

filtering said at least part of the points of the point cloud based on their normal vectors by ignoring points which do not have a normal vector with a substantially vertically upward direction;

ordering the remaining points into sets of points based on their vertical elevation, so that each set of points represents a respective step of the staircase;

determining a straight edge line of each set, which edge line represents the respective step; and

storing a numerical representation of each determined edge line in the digital memory.

2. The method of claim 1 , further comprising the following steps for ordering the remaining points into sets of points based on their vertical elevation:

determining one or more expected number of steps based on the difference in elevation of the bottom points and the upper points of the filtered point cloud and a predetermined step height value;

applying to the filtered point cloud a robust plane fitting algorithm which detects horizontal planes, such as RANSAC, by using the expected number of steps; and

determining the vertical elevation of said horizontal planes.

3. The method of claim 1 , further comprising the following step:

before determining the straight edge line of each set of points, dividing the points of each set into clusters which have no mutual connection, calculate the largest cluster, and ignoring the points of the smaller cluster(s).

4. The method of claim 1 , further comprising the following step:

before determining the straight edge line of each set of points, calculating the center of mass of the set of points and determine the number of points inside a predetermined radius around the center of mass, and dismissing sets of points which have a number of points below a predetermined value.

5. The method of claim 1 , further comprising the following step:

before determining the straight edge line of each set of points, filtering the points of the sets by applying a principal component analysis (PCA), in order to exclude thin objects detected as planes.

6. The method of claim 1 , wherein determining the straight edge line of each set of points comprises:

determining boundary points of each set of points that belong to the boundary of said set of points;

determining a fixed number of most dominant straight lines in the boundary, the most dominant lines being the lines that comprise the most points in the boundary; and

determine said edge line among said determined most dominant lines, by selecting from said most dominant lines the line which shows a difference in the elevation of points of the point cloud on either side of said line.

7. The method of claim 1 , wherein determining a fixed number of most dominant straight lines in the boundary is done with an iterative robust line-fitting algorithm.

8. The method of claim 1 , further comprising the following step:

assigning an elevation to each determined edge line by calculating the median value of the elevations of the points of the corresponding set of points.

9. The method of claim 1 , wherein the numerical representation of each determined edge line comprises three-dimensional coordinates of the two respective endpoints of said edge line.

10. A method for the extraction of information about a staircase, wherein the method is used for designing and making a stair lift comprising a guide which is installed at the staircase and along which a frame on which a person can be seated can move, wherein use is made of:

a 3D scanning device which is arranged to obtain point clouds of surfaces of objects; and

a computer device comprising:

a central processing unit loaded with a computer program;

a digital memory connected to the central processing unit;

input means connected to the central processing unit;

output means connected to the central processing unit;

wherein the method comprises:

scanning the staircase with the 3D scanning device to obtain a point cloud of the surface of the staircase;

calculating the estimated normal vector of at least a part of the points of the point cloud;

filtering said at least part of the points of the point cloud based on their normal vectors by ignoring points which do not have a normal vector with a substantially vertically upward direction;

ordering the remaining points into sets of points based on their vertical elevation, so that each set of points represents a step;

determining a straight edge line of each set, which edge line represents the respective step; and

storing a numerical representation of each determined edge line in the digital memory.

11. The method according to claim 1 , wherein the 3D scanning device and the computer device are comprised in one housing.

12. A computer device, comprising:

a central processing unit loaded with a computer program;

a digital memory connected to the central processing unit; and

output means connected to the central processing unit;

wherein the computer program comprises instructions for carrying out the method in accordance with claim 1 .

13. A non-transitory computer readable carrier comprising a computer program comprising instructions for carrying out the method in accordance with claim 1 .

14. The method according to claim 7 , wherein the iterative robust line-fitting algorithm comprises RANSAC.

15. The method according to claim 10 , wherein the guide comprises a rail.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (16)