Optics: measuring and testing – By polarized light examination – With light attenuation
Reexamination Certificate
1999-02-04
2001-08-07
Rosenberger, Richard A. (Department: 2877)
Optics: measuring and testing
By polarized light examination
With light attenuation
Reexamination Certificate
active
06271918
ABSTRACT:
FIELD OF THE INVENTION
The invention relates to a 3-D range sensor for optical ranging of points on a target surface and in particular a 3-D range sensor employing redundancy to validate a range measurement and enhance reliability.
BACKGROUND OF THE INVENTION
Two popular techniques currently in use for optical ranging of a target surface are known, respectively, as the standard optical triangulation system and the Biris (bi-iris) system, the latter employing an apertured mask in a converging lens system of an imaging device having a position sensitive detector, e.g. a CCD camera.
Essentially, in order to determine a distance to a point within an image a triangle is formed with the point at a vertex thereof. Another corner of the triangle is formed by a detector, commonly a CCD imaging device. By knowing some of the dimensions of the triangle, a distance from the detector to the point within the image can be determined. This is referred to as triangulation.
A known method of triangulation requires two detectors for imaging a scene. Each image is analysed to extract a same feature. The two detectors and the feature form the three angles of a triangle and triangulation is performed. Known problems with such a system include computational complexity, feature extraction problems, and perspective related problems which can reduce accuracy.
Another common approach to triangulation uses only a single detector and a laser. The laser shines a target in the form of a dot of known colour onto a surface and the detector images the dot. The detector, laser, and dot form the angles of a triangle. Though such a system is useful in controlled environments, when one tries to use it in uncontrolled environments, noise and other issues prevent accurate measurement of distances.
These systems are described and compared in F. Blais et al. (88), i.e. “Practical Considerations for a Design of a High Precision 3-D Laser Scanner System”, published in Optomechanical and Electro-optical Design of Industrial Systems, SPIE Vol. 959, 1988, pp 225-246, and also in F. Blais et al. (91), i.e. “Optical Range Image Acquisition for the Navigation of a Mobile Robot”, published in the Proceedings of the 1991 IEEE International Conference on Robotics and Automation, Sacramento, Calif., Apr. 9-11, 1991. The Biris system had previously been reported by M. Rioux et al. (86) in “Compact Three-Dimensional Camera For Robotic Applications”, published in the Journal of the Optical Society of America A, Vol. 3, p 1518, September 1986, and in M. Rioux U.S. Pat. No. 4,645,347 issued Feb. 24, 1987. The herein referenced documents are hereby incorporated by reference.
The Biris system uses a laser to form a target as well as a dual iris detector for forming an image with two separately imaged views of the target. This permits verification of target position and increased accuracy. An advantage of the Biris system is its small size and the robustness of the range sensor. The Biris system is better than the above triangulation systems, because it uses the redundancy introduced by an apertured mask to validate the measurements and to compensate for small errors introduced into range measurements due to detector resolution. For example, such a system is disclosed in U.S. Pat. No. 5,270,795 issued in 1993 to Blais and is incorporated herein by reference. More recently, an anamorphic lens system has been used that increases the field of view without compromising accuracy as is disclosed in: F. Blais, J. A. Beraldin, “Calibration of an anamorphic laser based 3-D range sensor” Videometric V, SPIE Proc. 3174, San Diego, Jul. 30-31, 1997, pp 113-122.
Unfortunately, new generations of CCD and CMOS image detectors are smaller due to new technologies that reduce production costs. For example, at present most newly released CCDs measure ⅓″ to a side compared to ⅔″ of the former generation. To obtain the same field of view with a ⅓″ CCD image detector as with a ⅔″ CCD image detector a lens of half the focal length is needed. Because the lens size is reduced, the overall aperture sizes are also reduced. This results in two apertures having little spacing therebetween and allowing less light to pass therethrough. Furthermore, the contribution of the Biris, the aperture spacing, to the triangulation is negligible due to a very small separation of the two apertures. This makes the method prone to false measurements.
It is well known in the art to use a beam splitter for range measurements. For example, in “Modern Optical Engineering”, Second Edition, by Warren J. Smith published by McGraw-Hill, 1990 pp. 254-257 range measurement using a beam splitter is discussed. Two different images of a single feature are superimposed on a screen, for example. The superposition is performed using a beam splitter or semi-transparent mirror This allows viewing of the feature from two different “locations” a greater distance apart compared to the Biris system. The two images of the feature are then adjusted so that the feature overlaps itself. Commonly, the images are adjusted by adjusting the angle two mirrors, one in each optical path. Once the feature is coincident on the screen, a triangle is known between the mirrors and the feature. This system provides a greater base for triangulation increasing accuracy of the range measurement. Unfortunately, commonly a human operator is used to overlap the image. Even when automated, since feature extraction is a difficult process, the accuracy of such a system reduces to a the accuracy of a known stereoscopic triangulation system with the same drawbacks.
It is therefore an object of this invention, to provide a range sensor that is highly accurate and reliable. This is accomplished by using a beam splitter to provide a plurality of images of a target point on a surface, each image relating to a triangle having a different triangle base for use in triangulation of target point than the other images in order to reduce a number of false measurements and increase overall accuracy.
It is further an object of this invention, to provide a range sensor that is compact and inexpensive to manufacture by using commonly available low cost imaging components.
SUMMARY OF THE INVENTION
According to the present invention there is provided a range sensor for measuring a range from the range sensor to a target point on a surface comprising:
a light source for projecting a target at the target point on the surface;
a first detector for capturing a first image;
a second detector for capturing a second other image;
a first optical path for providing the first image of the target point to the first detector the first image provided from a first viewing angle;
a second optical path for providing a second image of the same target point to the second detector from a different viewing angle;
a processor for determining a distance measurement in dependence upon the two captured images, the distance measurement determined using at least two imaged target points,
imaged target point positions used to determine a range measurement, the processor for validating the determined range measurement.
In accordance with the invention there is also provided a range sensor for measuring a range from the range sensor to a target point on a surface comprising:
a light source for projecting a target at the target point on the surface;
a detector for capturing an image;
a lens for focussing the first image and the second other image onto the detector;
a first optical path having a first optical axis for providing the first image of the target point to the detector the first image provided from a first viewing angle;
a second optical path having a second different optical axis for providing a second image of the same target point to the detector from a different viewing angle;
a processor for determining a distance measurement in dependence upon a captured image, the distance measurement determined using at least detected target points.
Preferably, the range sensor as defined in claim
6
wherein the processor c
Freedman & Associates
National Research Council of Canada
Rosenberger Richard A.
LandOfFree
Virtual multiple aperture 3-D range sensor does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Virtual multiple aperture 3-D range sensor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Virtual multiple aperture 3-D range sensor will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2468619