Method and system for detecting an object in relation to a surface
AssigneePoseidon
InventorsJerome Meniere, Christophe Migliorini
Life of this patent
Estimating the term…
Abstract
The invention relates to a method for detecting an object (1) in relation to a surface (2). A synthetic image (3c, 4c) is produced from at least two distinct viewpoints (3, 4), for each of said viewpoints, by means of a geometrical transformation comprising at least one perspective projection (3a, 4a) of the object on the surface (2), each synthetic image being associated with the viewpoint concerned. By discriminating between the synthetic images, it is possible to detect the presence of and determine the position of the object (1) in relation to the surface (2), by searching for contrast inversions. The invention can be used for surveying a swimming pool.
Claims28
A process for detecting an object (1) in relation to a surface (2), notably a virtual surface, said surface (2) not necessarily being flat;
- said process includes the stage:
- for realising a computer-generated image (3c, 4c) associated to the viewpoint being considered for each viewpoint by means of a geometric conversion at least including a projection in perspective (3a, 4a) of said object (1) on said surface (2) from at least two distinct viewpoints (3, 4),
- for distinguishing between the computer-generated images (3c, 4c) for detecting a presence and determining the position of said object (1) in relation to said surface (2);
- said process includes the following stages in order to distinguish between the different computer-generated images (3c, 4c):
- the stage for breaking said surface (2) down into a mosaic of fields (Zi) each including a centre (Ci),
- the stage for associating a specific direction (3Di, 4Di) to each centre (Ci) and each viewpoint (3, 4);
- in this way each computer-generated image (3c, 4c) is broken down in virtual fields (zi) each consisting of as many specific virtual directions (3di, 4di) as there are viewpoints.
- The said process also includes the following in order to distinguish between the computer-generated images (3c, 4c):
- the stage of selecting at least a couple of computer-generated images (3c, 4c) associated to at least a couple of viewpoints (3, 4),
- the stage of choosing zones (3fi, 3gi, 4fi, 4gi) situated according to each of the specific virtual directions (3di, 4di) of each virtual field (zi) on each computer-generated image (3c, 4c) of the couple selected;
- the stage of comparing (3hi, 4hi) said zones (3fi, 3gi, 4fi, 4gi) in searching for opposing differences from one computer-generated image (3c) to another (4c).
Process according to any of the claims from 1 to 8 more particularly intended to detect an object (1) in front of a bottom (200), said bottom (200) consisting of said surface (2).
A system for detecting an object (1) in relation to a surface (2), notably a virtual surface, said surface (2) not necessarily being flat;
- said system includes:
- means of detection (31, 41) and the first means of calculation of a central processing unit (73) in order to realise a computer-generated image (3c, 4c) associated to the viewpoint being considered by means of a geometric conversion including at least one projection in perspective (3a, 4a) of said object (1) on said surface (2) from at least two distinct viewpoints (3, 4) and for each of these,
- means of discrimination (73) in order to distinguish between the computer-generated images (3c, 4c) in order to detect presence and to determine the position of said object (1) in relation to said surface (2);
- said means of discrimination of said system include:
- means of partition for breaking said surface (2) down into a mosaic of fields (Zi) each including a centre (Ci) and as many specific directions (3Di, 4Di) as there are viewpoints (3, 4);
- so that each computer-generated image (3c, 4c) is broken down into virtual fields (zi) each including as many specific virtual directions (3di, 4di) as there are viewpoints;
- said means of discrimination also include the following in order to distinguish between the computer-generated images (3c, 4c):
- the means of selection for selecting at least a couple of computer-generated images (3c, 4c) associated to at least a couple of viewpoints (3, 4),
- the means for choosing the zones (3fi, 3gi, 4fi, 4gi) situated according to each of the specific virtual directions (3di, 4di) of each virtual field (zi) on each computer-generated image (3c, 4c) of the couple selected,
- the means of comparison for comparing (3hi, 4hi) said zones (3fi, 3gi, 4fi, 4gi) in searching for opposing differences from one computer-generated image (3c) to another (4c)
Dependencies are inferred from the claim text.