Image processing circuit and method thereof

1. An image processing circuit, adapted to calculate a first minimum weighted sum of absolute difference (WSAD) of a first search window in a reference image and a corresponding first target window in a target image, and a second minimum WSAD of a second search window in the reference image and a corresponding second target window in the target image, wherein the first search window and the second search window have a common matching window, and the first target window and the second target window have a common target block, the circuit comprising:

a calculation module implemented with hardware including logic gates, configured to generate a plurality of first WSADs according to the first search window, the first target window and a weight coefficient set, and to generate a plurality of second WSADs according to the second search window, the second target window and the weight coefficient set, wherein the calculation module comprises:

a plurality of absolute difference calculation units, configured to respectively calculate an absolute difference between a pixel in the common matching window and a corresponding pixel in the common target block;

a first multiplication unit, configured to multiply the absolute differences by a first weight coefficient group of the weight coefficient set to generate a plurality of first weighted absolute differences;

a first addition unit, configured to add the first weighted absolute differences to obtain one of the first WSADs;

a second multiplication unit, configured to multiply the absolute differences by a second weight coefficient group of the weight coefficient set to generate a plurality of second weighted absolute differences; and

a second addition unit, configured to add the second weighted absolute differences to obtain one of the second WSADs; and

a comparison unit, configured to compare the first WSADs and the second WSADs to generate the first minimum WSAD and the second minimum WSAD, respectively;

wherein the calculation module generates one of the first WSADs and one of the second WSADs according to the common matching window, the common target block and the weight coefficient set.

2. The image processing circuit according to claim 1 , wherein the first weight coefficient group is determined according to the first search window, and the second weight coefficient group is determined according to the second search window.

3. The image processing circuit according to claim 2 , wherein the first weight coefficient group is determined according to relative positions of the common matching window and the first search window, and the second weight coefficient group is determined according to relative positions of the common matching window and the second search window.

4. The image processing circuit according to claim 1 , wherein the weight coefficient set comprises a first weight coefficient group corresponding to the first search window and a second weight coefficient group corresponding to the second search window, the first weight coefficient group is equal to the second weight coefficient group, and the calculation module comprises:

a plurality of absolute difference calculation units, configured to respectively calculate an absolute difference between a pixel in the common matching window and a corresponding pixel in the common target block;

a multiplication unit, configured to multiply the absolute differences by the first weight coefficient group to generate a plurality of weighted absolute differences; and

an addition unit, configured to add the weighted absolute values to obtain an accumulated value, which simultaneously serves as one of the first WSADs and one of the second WSADs.

5. The image processing circuit according to claim 1 , the second WSADs being generated at different time points, the image processing circuit further comprising:

a delay unit, configured to delay one of the second WSADs;

wherein, the comparison unit generates the second minimum WSAD according to the one of the second WSADs delayed by the delay unit.

6. The image processing circuit according to claim 5 , wherein the delay unit comprises a flip flop.

7. An image processing method, for calculating a first weighted sum of absolute difference (WSAD) of a first search window in a reference image and a corresponding first target window in a target image, and a second WSAD of a second search window in the reference image and a corresponding second target window in the target image; the first search window and the second search window having a common matching window, the first target window and the second target window having a common target block; the method comprising:

a) calculating a plurality of absolute differences between a plurality of pixels in the common matching window and corresponding pixels in the common target block;

b) selecting a first weight coefficient group from a weight coefficient set;

c) selecting a second weight coefficient group from the weight coefficient set; and

d) sharing the absolute differences generated in step (a), summing up products from respectively multiplying the absolute differences by the first weight coefficient group to generate the first WSAD by utilizing a first multiplication unit and a first addition unit, and summing up products from respectively multiplying the absolute differences by the second weight coefficient group to generate the second WSAD by utilizing a second multiplication unit and a second addition unit.

8. The image processing method according to claim 7 , wherein the first weight coefficient group is determined according to relative positions of the common matching window and the first search window.

9. The image processing method according to claim 7 , wherein the first weight coefficient group is equal to the second weight coefficient group, and step (d) utilizes a same multiplication unit and a same addition unit to generate the first WSAD and the second WSAD.

10. The image processing method according to claim 7 , further comprising:

e) delaying the second WSAD; and

f) generating a minimum WSAD of the second search window according to the delayed second WSAD.

11. An image processing circuit, adapted to calculate a first weighted sum of absolute difference (WSAD) of a first search window in a reference image and a corresponding first target window in a target image, and a second WSAD of a second search window in the reference image and a corresponding second target window in the target image; the first search window and the second search window having a common matching window, the first target window and the second target window having a common target block, the common target block comprising a plurality of pixels; the circuit comprising:

a plurality of absolute difference calculation units implemented in hardware, configured to respectively calculate an absolute difference between a pixel in the common matching window and a corresponding pixel in the common target window;

a first multiplication unit implemented in hardware including logic gates, configured to multiply the absolute differences by a first weight coefficient group of a weight coefficient set to generate a plurality of first weighted absolute differences;

a first addition unit implemented in hardware including logic gates, configured to add the first weighted absolute differences to obtain the first WSAD;

a second multiplication unit implemented in hardware including logic gates, configured to multiply the absolute differences by a second weight coefficient group of the weight coefficient set to generate a plurality of second weighted absolute differences; and

a second addition unit implemented in hardware including logic gates, configured to add the second weighted absolute differences to obtain the second WSAD.

12. The image processing circuit according to claim 11 , wherein the first weight coefficient group is determined according to the first search window, and the second weight coefficient group is determined according to the second search window.

13. The image processing circuit according to claim 12 , wherein the first weight coefficient group is determined according to relative positions of the common matching window and the first search window, and the second weight coefficient group is determined according to relative positions of the common matching window and the second search window.

14. The image processing circuit according to claim 11 , wherein when the first weight coefficient group is equal to the second weight coefficient group, the first multiplication unit and the second multiplication unit are a same multiplication unit.