Video image transforming method and apparatus

• US 5,070,465 A
• Filed: 02/22/1988
• Issued: 12/03/1991
• Est. Priority Date: 02/25/1987
• Status: Expired due to Term

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Nakamura, Yoshiyuki, Kato, Tetsuro
• Primary Examiner(s): Herndon, Heather R.
• Assistant Examiner(s): Nguyen, Phu K.
• Application Number:
  US07/158,922
• Time in Patent Office: 1,380 Days
• Field of Search: 364/518, 364/521, 364/522, 340/723, 340/728, 340/747, 340/750, 340/707, 340/724, 382/25, 382/41, 382/44
• US Class Current: 345/441
• CPC Class Codes: G06F 17/175 of multidimensional data
• CPC Class Codes: G06T 3/4084 in the transform domain, e.g. fast Fourier transform [FFT] domain scaling
• CPC Class Codes: H04N 19/54 using feature points or meshes
• CPC Class Codes: H04N 5/2628 Alteration of picture size, shape, position or orientation, e.g. zooming, rotation, rolling, perspective, translation

Abstract

An image transforming method and apparatus divides an input image into a plurality of original minute rectangular segments. Calculations are performed to determine the positions of the four apexes of a new minute rectangular segment to which each of the original minute rectangular segments is to be transformed. The new minute rectangular segment is divided by a diagonal line into two triangular sections. The coordinates on the input image corresponding to the respective pixels in each of the triangular sections by interpolation using two of the sides of the triangular section except for the side made by the diagonal line used in dividing the new minute rectangular segment. The input image data are mapped on the corresponding output image based on the obtained coordinates.

Claims

1. A method of transforming a video image in the form of minute quadrangular segments, each having four apexes as defined by position coordinates, comprising the steps of:

obtaining first position data for the coordinates of each of the four vertices of respective first minute quadrangular regions of the input image;

obtaining second position data for the coordinates of each of four vertices (corners) of respective second minute quadrangular regions into which the first minute quadrangular regions are to be correspondingly transformed;

calculating parameters corresponding to each relationship between the respective first minute quadrangular regions and the respective second minute quadrangular regions using respective pairs of vectors which are congruent with the respective pair of adjacent sides included in each triangular region divided from the second minute quadrangular regions so as to contain the longer diagonal thereof, the respective pairs of adjacent sides constituting a pair of triangles together with the longer diagonal of each second minute quadrangular; and

transforming video information at each point within the respective first minute quadrangular regions into corresponding video information at each point within the respective second minute quadrangular regions using the respective parameters so as to constitute the output video image.

2. Apparatus for transforming input video image data comprising:

first memory means for storing input image data at an address specified by a first write address signal and from which the input image data is read out according to a first read address signal;

means for generating the first write address signal;

means for obtaining respective first position data with respect to each of the coordinates of four vertices (corners) of respective first minute quadrangular regions to be divided from the input video image;

means for obtaining respective second position data with respect to each of the coordinates of four vertices (corners) of respective second minute quadrangular regions into which the first minute quadrangular regions are to be correspondingly transformed by calculating using one or more respective transforming functions;

means for calculating respective parameters corresponding to the relationships between each first minute quadrangular region and a corresponding second minute quadrangular region using two respective pairs of vectors each of which are congruent with respective different pairs of adjacent sides of each second minute quadrangular region, the respective pairs of adjacent sides constituting a pair of triangles together with the longer of the two diagonals of the second minute quadrangular region;

second memory means for storing output image data at an address specified by a second write address signal and from which the output image data is readout under the control of a second read address signal so as to produce output video image data;

means for generating the second write address signals for each sample within each second minute quadrangular region and for generating first read address signals for each sample within each first minute quadrangular region using the respective parameters and the second write address signals; and

means for generating the second read address signal.

3. Apparatus for interpolating data at respective interpolation points from data sampled at respective sampling points comprising:

means for receiving data corresponding to the respective distances from the respective sampling points to the respective interpolation points and for producing compensating distance data according to these respective distances;

calculating means provided with the respective sampling data and the compensating distance data and for producing the respective interpolation data using the respective sampling data and the compensating distance data;

filter means provided with the respective interpolation data for producing respective output data having a predetermined frequency component from the respective interpolation data;

first adding means provided with the respective output data of the filter means and respective weighting data corresponding to the respective distances from the respective sampling points to the respective interpolation points and for producing first respective output data by adding together the respective weights to the respective output data of the filter means;

second adding means provided with the respective interpolation data and the respective output data of the first adding means and for producing second respective output data by adding together the first respective output data of the first adding means and the respective interpolation data.

4. Apparatus for interpolating data at respective points from data sampled at respective sampling points comprising:

first filtering means, having first filter characteristics and provided with the respective sampling data, for filtering the respective sampling data and for producing first output data;

second filtering means, having second filter characteristics and provided with the respective sampling data, for filtering the respective sampling data and for producing second output data;

first interpolation means provided with the first output data of the first filtering means for producing first interpolation data using the first output data of the first filtering means and a linear interpolation method;

second interpolation means provided with the second output data of the second filtering means for producing second interpolation data using the second output data of the second filtering means and a sinusoidal interpolation method;

adding means for adding together the first interpolation data from the first interpolation means and the second interpolation data from the second interpolation means.

5. Apparatus for filtering an input video image, whereby the input video image is transformed into an output video image, comprising;

a first filtering system being provided with the input video image data for producing first output data filtered in a vertical direction and a horizontal direction, the first filtering system having first filtering means for filtering input data thereof in the vertical direction and second filtering means for filtering input data thereof in the horizontal direction;

a second filtering system being provided with the input video image data for producing second data which is filtered in a first slant direction of left hand side down and right hand side up and a second slant direction of left hand side up and right hand side down, the second filtering system having third filtering means for filtering the input data thereof in the first slant direction and a fourth filtering means for filtering the input data thereof in the second slant direction;

whereby the first to fourth filtering means are controlled in the filter characteristics thereof according to the magnitudes of the transformations;

first weighting means for weighting the first output data of the first filtering system and for producing a first weighted output signal;

second weighting means for weighting the second output data of the second filtering system and for producing a second weighted output signal;

whereby the first and second weighting means are controlled in the weighting thereof according to the directions of the transformation; and

adding means for adding together the first weighted output data and the second weighted output data and for producing output video image data therefrom.