Method for processing a macro block of image data
View Patent ↗The image data are coded with field DCT or frame DCT depending on the characteristics of the image data. However different coding types will result in different boundary marks of boundaries between adjacent blocks or adjacent macro blocks. Therefore the de-blocking of a boundary between two adjacent blocks or adjacent macro blocks should be performed according to the format of image data and the coding type of the adjacent blocks or adjacent macro blocks.
1. A method for processing a macro block of image data comprising:
coding the macro block with frame discrete cosine transform (frame DCT) or field discrete cosine transform (field DCT); and
de-blocking a horizontal boundary of adjacent blocks of the macro block with four odd pixels per column closest to the horizontal boundary and four even pixels per column closest to the horizontal boundary when the format of the image data is of interlaced frame pictures and when the coding type of the macro block is frame DCT.
2. The method of claim 1 wherein de-blocking the horizontal boundary of adjacent blocks of the macro block according to the format of the image data and the coding type of the macro block is when the format of the image data is of progressive frame pictures and when the coding type of the macro block is frame DCT, de-blocking the horizontal boundary of adjacent blocks of the macro block with eight pixels per column closest to the horizontal boundary.
3. The method of claim 1 wherein de-blocking the horizontal boundary of adjacent blocks of the macro block according to the format of the image data and the coding type of the macro block is when the format of the image data is of interlaced field pictures, de-blocking the horizontal boundary of adjacent blocks of the macro block with eight odd pixels per column closest to the horizontal boundary and eight even pixels per column closest to the horizontal boundary.
4. A method for processing adjacent macro blocks of image data comprising:
coding the adjacent macro blocks with frame discrete cosine transform (frame DCT) or field discrete cosine transform (field DCT); and
de-blocking a horizontal boundary of the adjacent macro blocks with four odd pixels per column closest to the horizontal boundary and four even pixels per column closest to the horizontal boundary when the format of the image data is of interlaced frame pictures.
5. The method of claim 4 wherein de-blocking the horizontal boundary of the adjacent macro blocks according to the format of the image data and the coding type of the adjacent macro blocks is when the format of the image data is of progressive frame pictures and when the coding type of the adjacent macro blocks is frame DCT, de-blocking the horizontal boundary of adjacent macro blocks with eight pixels per column closest to the horizontal boundary.
6. The method of claim 4 wherein de-blocking the horizontal boundary of the adjacent macro blocks according to the format of the image data and the coding type of the adjacent macro blocks is when the format of the image data is of progressive frame pictures and when the coding type of the adjacent macro blocks is field DCT, de-blocking the horizontal boundary of adjacent macro blocks with four odd pixels per column closest to the horizontal boundary and four even pixels per column closest to the horizontal boundary.
7. The method of claim 4 wherein de-blocking the horizontal boundary of the adjacent macro blocks according to the format of the image data and the coding type of the adjacent macro blocks is when the format of the image data is of interlaced field pictures, de-blocking the horizontal boundary of adjacent macro blocks with eight odd pixels per column closest to the horizontal boundary and eight even pixels per column closest to the horizontal boundary.