Image analysis – Applications – 3-d or stereo imaging analysis
Patent
1996-02-08
1998-05-12
Boudreau, Leo
Image analysis
Applications
3-d or stereo imaging analysis
382288, 382291, 395358, G06T 760
Patent
active
057518432
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
The invention relates generally to a method for detecting the spatial position and rotational position of suitably marked objects in digital image sequences, in particular also for detecting and tracking the movements of a human hand on which, for example with the aid of a glove specially produced for this purpose, suitable marks are provided.
DESCRIPTION OF RELATED ART
Such methods are required for controlling data processing systems or devices associated therewith with the aid of hand signals (gestures) in various fields of technology in which the simplest possible man-machine communication is important. In the article by M. A. Clarkson, An easier interface, BYTE 16 (2), February 1991, a description is given of "3D-Rooms", which is an impressive example of a three-dimensional system, an extension of the known two-dimensional "Desktop" user interfaces for the presently customary computers with a graphic user interface, which example particularly illustrates the value of three-dimensional input devices. A method for detecting the spatial position and rotational position of suitably marked objects in digital image sequences thus represents an important building block for future more user-friendly user interfaces for data processing systems.
Articles M. W. Krueger, "Artificial Reality II", Addison-Wesley, 1991; V. J. Vincent, Delving in the depths of the mind, Proc. Interface to real & virtual worlds, Montpellier, 1991; P. Wellner, The Digital Desk Calculator: Tangible Manipulation on a Desk Top Display, Proceedings of the ACM Symposium on User Interface Software and Technology (UIST '91), November 1991, Hilton Head, USA; and J. Segen, Gest: A learning computer vision system that recognizes gestures, to appear in Machine Learning 4, which are known from the literature and based on image processing all have the disadvantage that they can recognize only two-dimensional gestures and that it is necessary to assume a uniform image background with prescribed characteristics.
The German Patent document DE-A-37 18 620 discloses a method for determining an angle of rotation by means of a so-called inter-two-centroid method. However, this method presupposes knowledge of the position of the object and its rotational position in the other two directions in space. It can therefore detect only the rotational position about an axis and cannot detect spatial positions.
SUMMARY OF THE INVENTION
It is the object of the invention to provide a method for detecting the spatial position and rotational position of suitably marked objects in digital image sequences, which is not subject to these limitations, but permits the recognition of three-dimensional gestures against an arbitrary image background.
This and other objects and advantages are achieved according to the invention by a method for detecting the spatial position and rotational position of suitably marked objects in digital image sequences wherein structures arranged eccentrically relative to one another are used to mark the objects.
The images of an image sequence are put into binary form by applying a threshold value operation, as shown at 10 in FIG. 9. As a result of putting the image into a binary form, a binary image sequence is formed. The moments of the individual segments of each individual binary image of the binary image sequence are determined up to a prescribed order as shown at 20 in FIG. 9. Parameters which describe the position and rotational position of the segments in space are determined from the moments as shown at 30 in FIG. 9; the markings and their position and rotational position in space are determined with the aid of these parameters and the moments as shown at 40 in FIG. 9. Finally, the position and rotational position of the marked objects in space are determined using the known arrangement of the markings on the marked objects as shown at 50 in FIG. 9.
Advantageous developments of the invention provide that the contours of the individual segments of each individual binary image of the binary image sequence
REFERENCES:
patent: 4748676 (1988-05-01), Miyagawa et al.
patent: 4776464 (1988-10-01), Miller et al.
patent: 5072294 (1991-12-01), Engle
patent: 5285273 (1994-02-01), James et al.
patent: 5452370 (1995-09-01), Nagata
M. Celenk et al., Hypercube Concurrent Processor Implementation of a Position Invariant Object Classifier, IEE Proceedings-E, vol. 138, No. 2, Mar. 1991, pp. 73-78.
VC. Maggioni et al., "IM Handumdrehen", 202 Funkschau, vol. 66 (1994) 21 Jan., No. 3, Germany, pp. 82-87.
K. Hemmi et al., "3-D Natural Interactive Interface--Using Marker Tracking from a Single View", Systems and Computers in Japan, vol. 23, No. 11, 1992, pp. 62-73.
Z. Zhou et al., "Morphological Skeleton Transforms for Determining Position and Orientation of Pre-Marked Objects", IEEE Pacific RIM Conf. on Communications, Computers and Signal Processing, 1989, pp. 301-305.
Theo Pavlidis, Algorithms for Graphics and Image Processing, Bell Laboratories, Springer-Verlag, 1982, pp. 316-401.
M. Krueger, "Artificial Reality II", Addison-Wesley Publ. Co., 1991, pp. 101-207.
M. Clarkson, "An Easier Interface", Byte, Feb. 1991, pp. 277-282.
P. Wellner, "The DigitalDesk Calculator: Tangible Manipulation on a Desk Top Display", UIST '91, 1991, pp. 27-33.
Maggioni Christoph
Wirtz Brigitte
Boudreau Leo
Siemens Aktiengesellschaft
Werner Brian P.
LandOfFree
Method for detecting the spatial position and rotational positio does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for detecting the spatial position and rotational positio, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for detecting the spatial position and rotational positio will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-991482