I'm currently working on VHDL code to stitch multiple tiles from the ADV212 encoder together in an FPGA. I understand the method of doing this to be the following-
1) Change the frame size of the first J2P header.
2) Remove all the remaining J2P headers and EOD markers (except the final EOD marker).
3) Change the tile index.
4) Change the tile size.
I have done this manually by taking a single J2P file, copying and pasting the data in the file 4 times, and adding/removing various header information. The end result is a tiled image consisting of 4 of the original J2P files arranged in a square.
My question is- in my FPGA application, how would I be able to calculate the tile size for each tile? From what I've seen on the forum, the tile size field from the ADV212 is all zeros. But, I need the exact tile size in bytes- the word size would include the padded zeros at the end of the file, which are variable. Also, I am putting the data into RAM and would prefer not to read-modify-write after the tile has been parsed and written.
Also, if possible, does anyone have a jp2 file encoded using the ADV212 includuing the ADV212 header?