Television – Camera – system and detail – Optics
Reexamination Certificate
1996-03-28
2002-12-10
Ho, Tuan (Department: 2612)
Television
Camera, system and detail
Optics
C348S207100
Reexamination Certificate
active
06493031
ABSTRACT:
BACKGROUND OF THE INVENTION
This invention relates to a visual information processing method and apparatus, particularly a visual information processing method and apparatus capable of extracting feature quantities from a two-dimensional image signal and of encoding the signal.
A conventional image input unit uses an input system in which a coordinate transformation from a coordinate system in real space to coordinates on a sensor plane is linear. For example, there is the usual lens system employed in a video camera, namely a lens system which realizes an identity mapping, and a line sensor used in a scanner. If the spacing between the sensors of an array sensor is constant in such an input system, sampling data having the same resolution are obtained over the entire area covered by the array sensor. Accordingly, if the sampling frequency is high enough, an advantage is that the input data will contain enough information for processing to be executed continuously.
A system is available in which use is made of a wide-angle lens in order to a view a wide area at one time. This system implements a monotonous scale conversion of the type in which the image becomes smaller in size as distance from the optical axis of the lenses increases. The advantage of this system is that it is possible to view all of a hemisphere at one time.
An ordinary lens system equipped with an optic-axis control device and a wide-angle lens for viewing the entirety of a hemisphere at one time are used in combination as the visual-information input unit of a monitoring apparatus, self-propelled vehicle or self-controller robot, which require that visual information over a wide range be obtained at one time. The direction of the optic axis of the ordinary lens system is decided based upon coarse information of the surroundings obtained by the wide-angle lens. With a system of this kind, high-resolution visual information over a narrow range and low-resolution visual information of a wide range may be obtained simultaneously.
A panoramic input unit realized by rotating the array sensor also has been proposed in order to obtain high-resolution visual information covering a wide range. The system provides full-peripheral high-resolution data if a suitable amount of time is expended.
However, the examples of the prior art described above have the following drawbacks:
1. Absolutely no visual information regarding areas other than an area capable of being entered at one is obtained from an input system that relies upon fixed, uniform sampling.
2. The resolution of the periphery declines dramatically when an input is made through the intermediary of the wide-angle lens.
3. In a panoramic input unit, a certain fixed period of time is necessary to obtain high-resolution visual information of a wide area.
SUMMARY OF THE INVENTION
Accordingly, an object of the present invention is to provide a visual information processing method and apparatus for obtaining an appropriate amount of visual information at high speed.
Another object of the present invention is to provide a visual information processing method and apparatus whereby a multi-resolution approximation or multi-resolution space, which is effective as an expedient for expressing an image, is obtained through a simple arrangement.
In order to solve the problems of the prior art, the present invention provides a visual information processing method in a visual information processing apparatus for acquiring and processing visual information, characterized by successively changing visual information with regard to a desired spatial position, reading image data at each image-forming position and storing the image data in correspondence with the spatial position.
The visual information is successively changed by changing the image-forming position corresponding to the spatial position by moving of the optic axis of a lens. The lens is a wide-angle lens. Further, the visual information is successively altered by zooming. Further, the visual information is successively changed by filtering which relies upon a two-dimensional filter. Further, there are at least two spatial positions, image data at each image-forming position are read out in correspondence with each spatial position, and the image data are stored in correspondence with each spatial position. The method is further characterized by a step of detecting at least two above-mentioned spatial positions. The method is further characterized by a step of changing the read image data to image data of multi-resolution space corresponding to the spatial positions, or to a power spectrum.
The invention further provides a visual information processing apparatus for acquiring and processing visual information, characterized by comprising visual information changing means for changing obtained visual information with respect to a desired spatial position, and visual information memory means for reading image data at each image-forming position and storing the image data in correspondence with the spatial position.
The visual information changing means has optic-axis moving means for moving the optic axis of a lens. The lens is a wide-angle lens. Further, the visual information changing means has means for controlling zooming. Further, the visual information changing means has means for performing predetermined two-dimensional filtering. Further, there are two or more spatial positions, and the visual information changing means reads the changing image data in correspondence with each spatial position and stores the image data in correspondence with each spatial position. The method is further characterized by a step of detecting the plurality of spatial positions. The apparatus is further characterized by having position detecting means for detecting the plurality of spatial positions. The apparatus is further characterized by having visual information processing means for changing the read image data to image data of multi-resolution space corresponding to the spatial positions, or to a power spectrum.
Further, a visual information processing apparatus according to the present invention is characterized by comprising image input means which includes a wide-angle lens, autofocus means, rangefinding means and an array sensor, input-parameter control means for controlling an input parameter of the image input means, external memory means for storing image information that enters from the image input means, and calculating means which, on the basis of distance to an object obtained from the rangefinding means and a coordinate transformation equation of the wide-angle lens, calculates an input-parameter control quantity, which controls the input parameter, and timing at which the image information is recorded in external recording means.
The apparatus is further characterized by comprising two-dimensional filter means for subjecting an output image from the array sensor to more than one stage of filtering. The apparatus is further characterized by comprising extremum detecting means for detecting an extremum of a two-dimensional signal outputted by the array sensor or two-dimensional filter means, and extremum memory means for storing coordinates of the detected extremum.
Other features and advantages of the present invention will be apparent from the following description taken in conjunction with the accompanying drawings, in which like reference characters designate the same or similar parts throughout the figures thereof.
REFERENCES:
patent: 4227212 (1980-10-01), Woolfsun et al.
patent: 4346405 (1982-08-01), Yoda et al.
patent: 4849813 (1989-07-01), Hirao et al.
patent: 5031049 (1991-07-01), Tagana et al.
patent: 5073868 (1991-12-01), Tada
patent: 5103305 (1992-04-01), Watanabe
patent: 5267034 (1993-11-01), Miyatake et al.
patent: 5508734 (1996-04-01), Baker et al.
patent: 5539491 (1996-07-01), Toshikatsu et al.
patent: 5581309 (1996-12-01), Okina et al.
patent: 5682197 (1997-10-01), Moghadam et al.
patent: 5686957 (1997-11-01), Baker
patent: 5812193 (1998-09-01), Tomitaka et al.
patent: 5990941 (1999-11-01), Jackson
Ho Tuan
Morgan & Finnegan L.L.P.
LandOfFree
Visual information processing method and apparatus for... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Visual information processing method and apparatus for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Visual information processing method and apparatus for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2983841