Method and device for optical detection of the position of an object
First Claim
1. A method of determining the position of an article in a zone of a substrate comprising a plurality of zone portions, the method comprising:
- placing a plurality of photodetectors each presenting a reception lobe in or around the zone, the set of reception lobes covering all of said zone;
structuring a neural network by supervised training;
applying to the input of the structured neural network data that results from converting signals delivered by said plurality of photodetectors that are sensitive to illumination of said zone; and
determining the position of the article in the zone as a function of at least one data item output by the structured neural network, by determining which of said zone portions has been touched or approached by the article forming an obstacle to the illumination of said zone;
wherein the supervised training is performed by presenting the input of the neural network with several hundreds or thousands of sample vectors in succession, each sample vector comprising a plurality of sample data items resulting from converting signals delivered by said plurality of photodetectors and corresponding to a plurality of different illumination conditions of said zone, and in which the zone is illuminated by a plurality of light emitters each comprising an emission lobe, the set of emission lobes illuminating all of said zone.
2 Assignments
0 Petitions
Accused Products
Abstract
The present invention relates to a method and to a device enabling data to be input, and also to an optical system for detecting the position of an article or a member and suitable for being used (or incorporated) in such a method (or device). The technical field of the invention is that of making keyboards and similar devices enabling manual input of data for processing by a digital computer. The method of the invention for determining the position of an article in a zone uses the steps of structuring a neural network by training, applying data to the input of the structured neural network, the data being the result of converting signals delivered by a plurality of detectors sensitive to illumination of said zone, and determining the position of the article in the zone as a function of at least one data item output by the structured neural network.
48 Citations
13 Claims
-
1. A method of determining the position of an article in a zone of a substrate comprising a plurality of zone portions, the method comprising:
-
placing a plurality of photodetectors each presenting a reception lobe in or around the zone, the set of reception lobes covering all of said zone; structuring a neural network by supervised training; applying to the input of the structured neural network data that results from converting signals delivered by said plurality of photodetectors that are sensitive to illumination of said zone; and determining the position of the article in the zone as a function of at least one data item output by the structured neural network, by determining which of said zone portions has been touched or approached by the article forming an obstacle to the illumination of said zone; wherein the supervised training is performed by presenting the input of the neural network with several hundreds or thousands of sample vectors in succession, each sample vector comprising a plurality of sample data items resulting from converting signals delivered by said plurality of photodetectors and corresponding to a plurality of different illumination conditions of said zone, and in which the zone is illuminated by a plurality of light emitters each comprising an emission lobe, the set of emission lobes illuminating all of said zone. - View Dependent Claims (2, 3)
-
-
4. A device for determining the position of an article in an input or detection zone of a substrate comprising a plurality of zone portions, the device comprising:
-
a plurality of light sources for illuminating the input or detection zone, each of the plurality of light sources comprising a set of emission lobes for illuminating all of said zone; a plurality of sensors sensitive to illumination of the zone and delivering electrical signals as a function of said illumination, each of the plurality of sensors comprising a set of reception lobes, the set of reception lobes of each of the plurality of sensors covering all of said zone; converter means connected to the plurality of sensors and adapted to convert said signals into data; and classifier means connected to the converter means and adapted to respond to the data that results from the conversion to deliver at least one output data item characteristic of the position of the article in the zone, by determining which portion of the zone the article is touching or is approaching, said classifier means comprising a neural network structured with supervised training obtained by presenting an input of the neural network with several hundreds or thousands of sample vectors in succession, each sample vector comprising a plurality of sample data items resulting from converting signals delivered by said plurality of detectors and corresponding to a plurality of different illumination conditions of said zone by said plurality of light sources. - View Dependent Claims (5, 6, 7, 8, 9, 10, 11, 12, 13)
-
Specification