Method and apparatus for per-pixel motion adaptive de-interlacing of interlaced video fields

A method for converting a subject interlaced video field into a progressive video frame, wherein the interlaced video field comprises a plurality of available pixels and a plurality of missing pixels, comprising:

• (a) calculating motion data related to each of the missing pixels in the field, wherein the motion data represents motion between at least two available pixels and between at least two interlaced fields;
• (b) determining whether to merge the subject field with one of a previous field, subsequent field or a combination of the previous and subsequent fields based on the motion data between the at least two fields;
• (c) if it is determined not to merge fields, considering a first missing pixel in the field and selecting at least one of a plurality of per-pixel interpolation methods based on the motion data related to the missing pixel via at least one lookup table, wherein the lookup table correlates motion data to at least one of a plurality of de-interlacing methods during a training process;
• and
• (d) utilizing the selected at least one de-interlacing method to generate a value for the missing pixel.

A method for selecting a per-pixel interpolation process for determining a value for a missing target pixel in a subject interlaced video field comprising:

• (a) providing at least one lookup table that correlates motion data to at least one of a plurality of per-pixel interpolation processes;
• (b) calculating motion data related to the missing target pixel, wherein calculating the motion data comprises:
• (i) analyzing pixel data associated with a plurality of pixels temporally and spatially proximate to the missing target pixel, wherein the plurality of proximate pixels includes at least one pixel in the subject field, at least one pixel in at least one field preceding the subject field and at least one pixel in at least one field subsequent to the subject field;
• (ii) calculating per-pixel motion measurements between the plurality of pixels;
• (iii) utilizing the per-pixel motion measurements to calculate per-pixel motion data and per-field motion data;
• (c) utilizing the per-pixel motion data and a first portion of the per-field motion data to perform a lookup on the at least one lookup table, wherein the first portion of the per-field motion data indicates the presence of higher levels of per-field motion measurements;
• and
• (d) selecting the at least one per-pixel interpolation method correlated to the per-pixel motion data and the first portion of the per-field motion data.

A de-interlacing system comprising:

• (a) a motion detection unit for generating motion data related to a missing target pixel in a subject interlaced video field based on per-pixel motion measurements calculated from temporally and spatially proximate pixels in the subject field, at least one preceding field and at least one subsequent field;
• (b) a video processing unit for receiving the motion data, and for selecting and utilizing at least one of a plurality of de-interlacing methods to determine a value for the missing target pixel based on the motion data related to the missing target pixel;
• and
• (c) at least one lookup table accessible to the video processing unit for correlating motion data to at least one of the plurality of de-interlacing methods, whereby the video processing unit selects the at least one de-interlacing method by utilizing the motion data related to the missing target pixel to perform a lookup on the at least one lookup table.

A progressive scan display system comprising:

• (a) a signal receiving unit;
• (b) a tuner box for transforming the signal into an analog signal;
• (c) a video decoder for transforming the analog signal into a plurality of interlaced video fields;
• (d) a de-interlacing system for converting the interlaced video fields into a progressive video signal, the de-interlacing system comprising:
• (i) a motion detection unit for generating motion data related to a missing target pixel in a subject interlaced video field based on per-pixel motion measurements calculated from temporally and spatially proximate pixels in the subject field, in at least one preceding field and in at least one subsequent field;
• (ii) a video processing unit for receiving the motion data, and for selecting and utilizing at least one of a plurality of de-interlacing methods to determine a value for the missing target pixel based on the motion data related to the missing target pixel;
• and
• (iii) at least one lookup table accessible to the video processing unit for correlating motion data to at least one of the plurality of de-interlacing methods, whereby the video processing unit selects the at least one de-interlacing method by utilizing the motion data related to the missing target pixel to perform a lookup on the at least one lookup table;
• and
• (e) a display for displaying the progressive video signal.