HEVC encoding device and method for determining intra-prediction mode using the same

1. A high efficiency video coding (HEVC) encoding device for determining an intra-prediction mode of an image, the HEVC encoding device comprising:

a candidate group updater configured to select a plurality of representative modes as a candidate group from among intra-prediction modes and update the candidate group using a plurality of minimum modes selected from the candidate group, the plurality of representative modes representing a range where there is an optimal mode; and

an optimal mode selector configured to select any one mode as an optimal mode from among a plurality of minimum modes selected from the updated candidate group,

wherein, upon the penultimate update of the candidate group, the candidate group updater updates a candidate group before the penultimate update by using a DC mode and a planar mode in addition to a plurality of minimum modes selected from the candidate group before the penultimate update,

wherein the candidate group updater updates the candidate group by adding and subtracting variable mode values to/from each of the plurality of minimum modes, while proceeding from a second update of the candidate group to a penultimate update of the candidate group,

wherein the variable mode value is decreased by a predetermined ratio as a number of update repetition is increased.

2. The HEVC encoding device of claim 1 , wherein the candidate group updater sets the plurality of representative modes based on a vertical prediction mode, a horizontal prediction mode, and a diagonal prediction mode, which are directional prediction modes included in the intra-prediction modes, and selects the set plurality of representative modes as a first candidate group.

3. The HEVC encoding device of claim 2 , wherein the candidate group updater sets the plurality of representative modes at intervals of 45 degrees based on the directional prediction modes.

4. The HEVC encoding device of claim 1 , wherein the candidate group updater calculates a rate-distortion cost (RD-cost) for each of plural modes included in the candidate group and selects the plurality of minimum modes from the candidate group based on a result of calculating the RD-cost.

5. The HEVC encoding device of claim 4 , wherein the candidate group updater selects two modes having two smallest RD-costs as first and second minimum modes from among the plural modes included in the candidate group.

6. The HEVC encoding device of claim 1 , wherein the candidate group updater decreases the variable mode values as the number of repetitions increases in order to reduce an interval between modes in the candidate group, from the second update of the candidate group to the penultimate update of the candidate group.

7. The HEVC encoding device of claim 1 , wherein, upon the last update of the candidate group, the candidate group updater sets at least one segment block having a highest probability of similarity to a current block as a most probable mode (MPM) from among previously predicted segment blocks of the image and selects the set MPM as a final candidate group generated by the last update in addition to the plurality of minimum modes.

8. The HEVC encoding device of claim 7 , wherein the optimal mode selector selects any one mode as the optimal mode from among the MPM and the plurality of minimum modes selected as the final candidate group.

9. The HEVC encoding device of claim 1 , wherein the optimal mode selector selects the optimal mode from the final candidate group generated by the last update of the candidate group by using rate-distortion optimization (RDO).

10. A method of determining an intra-prediction mode using a high efficiency video coding (HEVC) encoding device, the method comprising:

selecting, by a candidate group updater, a plurality of representative modes as a candidate group from among intra-prediction modes, the plurality of representative modes representing a range where there is an optimal mode;

updating, by the candidate group updater, the candidate group using a plurality of minimum modes selected from the candidate group; and

selecting, by an optimal mode selector, any one mode as the optimal mode from among a plurality of minimum modes selected from the updated candidate group,

wherein the updating of the candidate group comprises, upon a penultimate update of the candidate group, updating, by the candidate group updater, a candidate group before the penultimate update by using a DC mode and a planar mode in addition to a plurality of minimum modes selected from the candidate group before the penultimate update,

wherein the updating of the candidate group comprises updating the candidate group by adding and subtracting variable mode values to/from each of the plurality of minimum modes, while proceeding from a second update of the candidate group to a penultimate update of the candidate group,

wherein the variable mode value is decreased by a predetermined ratio as a number of update repetition is increased.

11. The method of claim 10 , wherein the updating of the candidate group comprises:

setting the plurality of representative modes based on a vertical prediction mode, a horizontal prediction mode, and a diagonal prediction mode, which are directional prediction modes included in the intra-prediction modes; and

selecting the set plurality of representative modes as a first candidate group.

12. The method of claim 10 , wherein the updating of the candidate group comprises:

calculating a rate-distortion cost (RD-cost) for each of plural modes included in the candidate group; and

selecting the plurality of minimum modes from the candidate group based on a result of calculating the RD-cost.

13. The method of claim 10 , wherein the updating of the candidate group comprises:

upon the last update of the candidate group, setting at least one segment block having a highest probability of similarity to a current block as a most probable mode (MPM) from among previously predicted segment blocks of the image; and

selecting the set MPM as a final candidate group generated by the last update in addition to the plurality of minimum modes.

14. The method of claim 13 , wherein the selecting of any one mode as the optimal mode comprises selecting any one mode as the optimal mode from among the MPM and the plurality of minimum modes selected from the final candidate group.

15. The method of claim 10 , wherein the selecting of any one mode as the optimal mode comprises selecting the optimal mode from the final candidate group generated by the last update of the candidate group by using rate-distortion optimization (RDO).

16. The HEVC encoding device of claim 4 , wherein the RD-cost is calculated by:

J=SATD+λ*B mode,

wherein J is a result of calculating the RD-coat, SATD is a difference between an interpolated value and an original value, λ is a Lagrangian constant according to a quantization constant, and Bmode is a bit that represents a mode.

17. The HEVC encoding device of claim 1 , wherein the candidate group updater selects the plurality of minimum modes from the candidate group before the penultimate update based on a below equation:

CSi={x|x =FMM i− 1±24− i,x =SMM i− 1±24− i,x =FMM i− 1, x =SMM i− 1, i= 2,3,4}

wherein the CSi is i-th candidate group, the FMMi−1 is (i−1)-th the first minimum mode, the SMMi−1 is (i−1) the second minimum mode, and ±24−i is the variable mode value.

18. The method of claim 12 , wherein the RD-cost is calculated by: J=SATD+λ*Bmode,

wherein J is a result of calculating the RD-coat, SATD is a difference between an interpolated value and an original value, λ is a Lagrangian constant according to a quantization constant, and Bmode is a bit that represents a mode.

19. The method of claim 10 , wherein the candidate group updater selects the plurality of minimum modes from the candidate group before the penultimate update based on a below equation:

CSi={x|x =FMM i− 1±24− i,x =SMM i− 1±24− i,x =FMM i− 1, x =SMM i− 1, i= 2,3,4}

wherein the CSi is i-th candidate group, the FMMi−1 is (i−1)-th the first minimum mode, the SMMi−1 is (i−1) the second minimum mode, and ±24−i is the variable mode value.

20. A high efficiency video coding (HEVC) encoding device for determining an intra-prediction mode of an image, the HEVC encoding device comprising:

a candidate group updater configured to select a plurality of representative modes as a candidate group from among intra-prediction modes, select N modes on basis of a result of calculating a rate-distortion cost (RD-cost) for each mode included in the candidate group, the N mode having the N-th smallest RD-cost and update the candidate group using the selected N modes, the plurality of representative modes each representing a vertical line, a horizontal line or a diagonal line; and

an optimal mode selector configured to select any one mode from among the N modes selected from the updated candidate group,

wherein, upon the penultimate update of the candidate group, the candidate group updater updates a candidate group before the penultimate update by using a DC mode and a planar mode in addition to a plurality of minimum modes selected from the candidate group before the penultimate update,

wherein each of the plurality of representative modes shares one mode with their neighboring representative mode when the candidate group updater selects the plurality of representative modes,

when the candidate group updater updates the candidate group, the candidate group updater applies variable mode values to each of the N modes depending on a predetermined number of repetitions while proceeding from a second update of the candidate group to a penultimate update of the candidate group and decreases the variable mode values as the number of repetitions increases in order to reduce an interval between modes in the candidate group, from the second update of the candidate group to the penultimate update of the candidate group.