Omnidirectional visual image detector and processor

Television – Panoramic

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C348S039000, C348S038000

Reexamination Certificate

active

06262760

ABSTRACT:

RELATED APPLICATIONS
This application is related to co-pending applications Ser. No. 08/364,160 entitled “Method and Apparatus for Automatically Tracking the Location of Vehicles”, filed on Dec. 27, 1994; Ser. No. 08/364,885 entitled “Discoidal Visual Image Detector”, filed on Dec. 27, 1994; and Ser. No. 08.364,880 entitled “Visual Incremental Turn Detector”, filed on Dec. 27, 1994. The teachings of the co-pending applications are incorporated herein by reference to the extent they not do conflict with the teachings herein.
FIELD OF THE INVENTION
The present invention is generally related to navigational systems, and more particularly to visual image detectors for use in such systems.
BACKGROUND OF THE INVENTION
The use of omnidirectional vision navigation systems for providing video or visual image information useful in robot navigation, or for the location of mobile systems, is known in the art. For example, Zhongfei Zhang, Richard Weiss, and Edward M. Riseman, presented a paper on Apr. 3, 1991, entitled “Segment-Based Matching for Visual Navigation”,
Computer Information Science Department.
University of Massachusetts, Amherst, Mass., “COINS PR91-35”. The paper teaches the use of a reflecting globe or spherical mirror, mounted on top of a mobile robot above a camera. The camera converts the picture received to a 360° video image of the surrounding environment. The video image from the camera is processed for obtaining a fixed set of target locations for permitting a robot to navigate between desired locations by carrying out a sequence of homing tasks relative to the target locations. The 360° view taken at a given time is condensed into a 1-dimensional location signature. Correlation techniques are used for providing matching between location signatures in navigating the robot. The location signature represented by three types of segments identified as increasing, decreasing, and constant, respectively. In the system, a “horizon circle” is superimposed upon the 360° image for taking a sample of the image every degree. The “horizon circle”, forms a circle band composed of 360 ticks. The circle is designated as being the actual horizon circle, with each tick being a sample thereof, as a function of the azimuth orientation. The resultant sequences of linear segments obtained are not all used for matching, whereby selective ones are obtained for providing “characteristic features” for matching between images, and navigating a robot to move from one image location to a next, in a successive manner. Each 360° video image is processed using a spherical coordinate system centered upon the origin of the image plane. The “characteristic features” chosen for matching are those which appear to be most distinctive and reliable for such use, for example parts of the processed waveform having a large slope for feature transition.
Another paper by Sung Jun Oh and Ernest L. Hall, entitled “A Study of the Characteristics of a Omnidirectional Vision Sensor”, published in SPIE, Volume 804 of
Advances and Image Processing
in 1987, on pages 259 through 267. The detector of the system includes a fish eye lens mounted over a video camera, for projecting a 360° image of its surroundings. The video camera consists of a CCD or charge couple device camera for providing a video signal representative of the 360° image to an image processing system.
Another known system for providing route recognition in robot navigation includes the use of a rotating slit in combination with a video camera for providing a video image band representative of a panoramic view of the environment a robot is to navigate. As the rotating slit camera apparatus is moved linearly the panoramic view changes. Successive 2D panoramic image strips are connected together for providing a continuous panoramic view relative to a robot moving through the chosen environment. The images are processed through use of circular dynamic programming to obtain vertical line segments from the images for use in matching techniques for guiding the robot's movement. The robot's signal processor compares its present panoramic view with recorded panoramic views in order to determine the robot's heading, and correct the same if it is off course. See J. Y. Zheng and S. Tsuji, “Panoramic Representation for Route Recognition by a Mobile Robot”,
International Journal of Computer Vision,
Volume 9:1, pages 55-76 (1992), Kluwer Academic Publishers, The Netherlands.
Present research for obtaining video information useful in route recognition for robot navigation, and for vehicle locator systems, is centered upon simplifying such systems while retaining reliable pattern matching. It is important that the amount of video image data necessary for reliable pattern matching be reduced to a minimum in order to reduce the size of the computer memories required by the associated systems.
SUMMARY OF THE INVENTION
An object to this invention is to provide an improved method and apparatus for obtaining strips of video or visual image information.
Another object of the invention is to provide an improved method and apparatus for compressing image data from a visual image strip extractor, while retaining sufficient data for reliable operation.
Another object of the invention is to provide an improved method and apparatus for obtaining compressed image data representative of an omnidirectional view of scenery about the sensor, for applying the data for use with navigational algorithms.
With the problems of the prior art in mind, an omnidirectional visual image detector is provided by mounting at least one quarter of a rotund (i.e.—spherical or cone shaped, for example) reflector means within one end of a transparent cylindrical housing means, with video or visual image camera means being located at the other end of the cylinder for receiving light reflected from the reflector means and focused through an intervening lens therebetween, for providing a video signal of a detected panoramic image of an associated horizon. In the extreme, if at least one half of the rotund reflector means is utilized, a 360° panoramic view can be obtained. A circle or semicircular portion of information appearing on the image plane is extracted for providing visual image data representative of a horizontal view of the surroundings at a given elevation. Digital signal processing means maps the image data to specific pixel locations, and discards all other image information not within a specified ring or ring segment area. The digital strip of data extracted from the pixel locations contains sufficient information for use with navigational algorithms or vehicle locator algorithms, to reliably navigate or locate a vehicle via the compressed image data of the extracted strip.


REFERENCES:
patent: 4100571 (1978-07-01), Dykes et al.
patent: 4908705 (1990-03-01), Wight
patent: 5159361 (1992-10-01), Cambier et al.
patent: 5166878 (1992-11-01), Poelstra
patent: 5177685 (1993-01-01), Davis et al.
patent: 5212643 (1993-05-01), Yoshida
patent: 5262775 (1993-11-01), Tamai et al.
patent: 5283575 (1994-02-01), Kao et al.
patent: 5291412 (1994-03-01), Tamai et al.
patent: 5291413 (1994-03-01), Tamai et al.
patent: 5291414 (1994-03-01), Tamai et al.
patent: 5303159 (1994-04-01), Tamai et al.
patent: 5311434 (1994-05-01), Tamai
patent: 5444478 (1995-08-01), Lelong et al.
patent: 5497188 (1996-03-01), Kaye
Zhang et al., “Segment-Based Matching For Visual Navigation”, Computer and Information Science, University of Massachusetts at Amherst, COINS TR91-35, pp. 1-37 (Apr. 1991).
Oh et al., “A study of the characteristics of an omnidirectional vision sensor”, Advances in Image Processing, SPIE vol. 804, pp. 259-268(1987).
Zheng et al., “Panoramic Representation for Route Recognition by a Mobile Robot”, International Journal of Computer Vision, 9:1, pp. 55-76 (1992).
Hong et al., “Image-Based Navigation Using 360◯ Views”, Proc. of Image Understanding Workshop, pp. 782-791 (1990).
Oh et al., “Guidance of a Mobile Robot Using an Omnidirectional Vision Navigation System”, Mobile Robots II, SPIE

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Omnidirectional visual image detector and processor does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Omnidirectional visual image detector and processor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Omnidirectional visual image detector and processor will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2556170

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.