Team C:
R09922028 施承志
R09922071 劉俊緯
R09922091 張庭與
R09944021 鍾起鳴
JPEG codec divides image into 8x8 blocks for encoding. However, real-world images often constains large blocks with few details. If we can design a codec which takes image detail into consider, it could improve the compression rate of an real-life image.
- Modify JPEG so that it can divide image into coding blocks of different sizes based on local image detail.
- An additional table for block division records.
- Larger DC difference between larger blocks
- Image detail recognition
- DCT for different block sizes
- Different quantisation
- DCT for different block sizes
- Higher compression ratio (i.e. smaller image size) for general images
- Not much more computational costs
- Not much worse in special cases