Image processing apparatus and image processing method

1. An image processing apparatus comprising:

a geometric position obtaining unit receiving a geometric transformation parameter, a block size, and a tile size, obtaining a plurality of base-point coordinates of the geometric transformation parameter according to the block size and the tile size, and building a base-point coordinate table according to the base-point coordinates, wherein, each of a plurality of base-points is one point of each of the blocks in a tile;

an image retrieving unit coupled to the geometric position obtaining unit, the image retrieving unit reading the base-point coordinate table, scanning the base-point coordinate table according a fixed block size to generate a plurality of reference image ranges, and respectively retrieving a plurality of partial image data of a to-be-processed image data according to each of the reference image ranges; and

an image rectifying unit coupled to the image retrieving unit and the geometric position obtaining unit, the image rectifying unit rectifying each of the partial image data according to the geometric transformation parameter.

2. The image processing apparatus as claimed in claim 1 , wherein the image retrieving unit generates each of the reference image ranges based on a difference between a maximum coordinate and a minimum coordinate of the base-point coordinates in any one of the fixed block sizes in the base-point coordinate table.

3. The image processing apparatus as claimed in claim 1 , wherein the image retrieving unit comprises:

a direct memory access for retrieving each of the partial image data of the to-be-processed image data based on each of the reference image ranges; and

a buffering memory coupled to the direct memory access for registering each of the partial image data in the image retrieving unit.

4. The image processing apparatus as claimed in claim 1 , wherein the geometric position obtaining unit comprises a digital signal processor.

5. The image processing apparatus as claimed in claim 1 , wherein the tile size is greater than the block size.

6. An image processing method comprising:

receiving a geometric transformation parameter, a block size, and a tile size;

obtaining a plurality of base-point coordinates of the geometric transformation parameter based on the block size and the tile size and building a base-point coordinate table based on the base-point coordinates, wherein, each of a plurality of base-points is one point of each of the blocks in a tile;

reading the base-point coordinate table and scanning the base-point coordinate table according to a fixed block size to generate a plurality of reference image ranges;

retrieving a plurality of partial image data of a to-be-processed image data based on each of the reference image ranges; and

rectifying each of the partial image data based on the geometric transformation parameter.

7. The image processing method as claimed in claim 6 , the step of scanning the base-point coordinate table according to the fixed block size to generate the reference image ranges comprising:

generating each of the reference image ranges based on a difference between a maximum coordinate and a minimum coordinate of the base-point coordinates in any one of the fixed block sizes in the base-point coordinate table.

8. The image processing method as claimed in claim 6 , further comprising:

registering each of the partial image data in a buffering memory.

9. The image processing method as claimed in claim 6 , wherein the tile size is greater than the block size.