Method and apparatus for adaptively interpolating a digital image

Digital image processing apparatus for interpolating a digital input image into an interpolated output image, comprising:

• an input buffer for accommodating pixel data of said input image;
• a coefficient buffer for storing pre-calculated interpolation weighting coefficients prior to real time image interpolation and comprising a first memory segment for containing a set of pre-calculated relatively sharp interpolating weighting coefficients obtained by using a relatively sharp interpolating algorithm;
• a second memory segment for containing a set of pre-calculated relatively soft weighting coefficients obtained by using a relatively soft interpolating algorithm;
• and a third memory segment for containing a set of pre-calculated weighting coefficients representative of a predetermined characteristic of said input image, and
• image convolution and interpolation means for calculating a relatively sharp interpolated output image pixel using said input pixel data and said set of relatively sharp weighting coefficients and for calculating a relatively soft interpolated output image pixel using said input pixel data and said set of relatively soft weighting coefficients and for calculating a value for said image characteristic using said input pixel data and said image characteristic weighting coefficients and for calculating a resultant output pixel by combining said relatively sharp and said relatively soft output image pixels as a function of said characteristic value.

Digital image processing apparatus for interpolating a digital input image into an interpolated output image, comprising:

• an input buffer for accommodating pixel data of said input image;
• means for locating a local position within said input image for interpolation;
• means for obtaining a first interpolated output image pixel at a predetermined pixel position in said interpolated output image;
• means for obtaining a second interpolated output image pixel at said predetermined pixel position,
• means for deriving an input image characteristic value from said input image pixel data;
• and
• means for calculating a resultant output image pixel at said predetermined position in said output image by combining said first and second interpolated output image pixels as a function of said input image characteristic value.