| 摘要 |
An image compressing method for encoding images based on motion compensation is provided to reduce complexity of calculation required for image data compression and facilitate random access of a memory. An image compressing method includes a step of determining an image bundle type, and a step of giving a code word to the determined image bundle type through appropriate bit allocation. In the step of determining the image bundle type, an image is divided into a mild region, an edge region and a random pattern region based on the human visual characteristic. The characteristic of the image is finally judged from a difference and variation pattern between neighboring pels for each of the regions to give an appropriate code word corresponding to the variation pattern to the image bundle type.
|
