Deinterlacing apparatus of digital image data

• US 6,396,543 B1
• Filed: 12/30/1999
• Issued: 05/28/2002
• Est. Priority Date: 12/31/1998
• Status: Expired due to Fees

• Current Assignee: LG Electronics, Inc. (LG Corporation)
• Original Assignee: LG Electronics, Inc. (LG Corporation)
• Inventors: Shin, Chang Yong, Han, Dong Il
• Primary Examiner(s): Kostak, Victor R.
• Application Number:
  US09/475,257
• Time in Patent Office: 880 Days
• Field of Search: 348/452,448,458,607,625,618
• US Class Current: 348/452
• CPC Class Codes: H04N 7/012 Conversion between an interlaced and a progressive signal for graphics images G09G2310/0229
• CPC Class Codes: H04N 7/0137 dependent on presence/absence of motion, e.g. of motion zones H04N7/014 takes precedence; movement detection in television signals H04N5/144

Abstract

A deinterlacing method and apparatus for digital image data is disclosed. In the present invention, one of various interpolations appropriately performed according to a detected extent and direction of motion in a pixel to be interpolated. Thus, a screen quality can significantly be improved.

Claims

1. A deinterlacing apparatus comprising:

a field data storage unit which stores field data;

a BD &

BPPD detecting unit which calculates a brightness profile pattern difference (BPPD) value and a brightness difference (BD) value between a current interlaced field n and a previous interlaced field (n−

2);

a combiner which respectively limits the BD and BPPD values by a threshold value, maps the limited BD and BPPD values to preset levels, compares the magnitudes and decides a presence and extent of motion to output detected motion;

a motion extender which extends an effect of the detected motion to pixels adjacent to a pixel to be interpolated;

an edge detector which detects an edge direction around a pixel to be interpolated; and

a interpolation unit which interpolates a pixel according to the output of the edge detector and the output of the motion extender, and outputs a non-interlaced field.

2. An apparatus of claim 1, wherein said edge detector detects an edge direction using the following equations:

3. An apparatus of claim 2, wherein the value of Med₄₅, Med₀, and Med₋

45 are each compared with either an identical threshold value or respectively compared with three different threshold values, and an edge direction is determined to be present in at least one direction corresponding to the values Med₄₅, Med₀, and Med<sub>−

45</sub> depending upon the comparison.

4. An apparatus of claim 2, wherein the values of Med₄₅, Med₀, and Med₋

45 are each compared to a threshold value E_th as follows;

If Med₄₅<

E_th, edge45°

=0 else edge45°

=Med₄₅

If Med₀<

E_th, edge0°

=0 else edge0°

=Med₀

If Med<sub>−

45</sub><

E_th, edge−

45°

=0 else edge−

45°

=Med<sub>−

45</sub>

and a edge direction corresponding to a minimum value among edge45°

, edge0°

, edge−

45°

is determined to be present.

5. An apparatus of claim 4, wherein the edge direction is determined according to the following table where α

<

β and

l*m*n,

6. An apparatus of claim 4, wherein a pixel is interpolated according to the following table,

7. An apparatus of claim 1, wherein said edge detector also obtains a local correlation around a pixel to be interpolated.

8. An apparatus of claim 7, wherein the local correlation is obtained using one of three methods below, where V_th indicates a vertical direction threshold, H_th represents a horizontal direction threshold, and V_cor=0 indicates a local correlation in a vertical direction:

(1) V_cor=1 if

Med{|P(i, j−

1)−

P(i+1, j−

1)|, |P(i, j)−

P(i+1, j)|, |P(i, j+1)−

P(i+1, j+1)}<

V_th;

(2) V_cor=1 if

|Med{P(i, j−

1), P(i, j), P(i, j+1)}−

Med{P(i+1, j−

1), P(i+1, j), P(i+1, j+1)}|<

V_th;

or

(3)

i) Hedge₁=0 if

|Med{P(i, j−

2), P(i, j−

1), P(i, j)}−

Med{P(i, j), P(i, j+1), P(i, j+2)}|<

H_th,

ii) Hedge₂=0 if

Med{P(i+1, j−

2), P(i+1, j−

1), P(i+1, j)}−

Med{P(i+1, j), P(i+1, j+1), P(i+1, j+2)}|<

H_th,

iii) Vedge=0 if

|Med{P(i, j−

1), P(i, j), P(i, j+1)}−

Med{P(i+1, j−

1), P(i+1, j), P(i+1, j+1)}|<

V_th,

where V_cor=NOT(Hedge₁ XOR Hedge₂ OR Vedge).

9. An apparatus of claim 1, wherein the interpolation unit comprises:

a soft switch which mixes the output value of the edge detector and the output of the motion extender, and outputs the mixed value to appropriately interpolate a pixel value; and

a vertical line converter which converts a number of vertical lines to conform with a required display format, by using data of the current interlaced field and interpolated pixels from the soft switch, and outputs a non-interlaced field.

10. An apparatus of claim 1, further comprising a median filter which eliminates a noise component in the motion detected by the combiner, groups portions with motion and outputs the result to the motion extender.

11. A method for deinterlacing digital image comprising:

(a) calculating a BPPD value and a BD value between a current interlaced field and a previous interlaced field;

(b) limiting the BD and BPPD values respectively by a threshold value, mapping the limited BD and BPPD values to preset levels, comparing the magnitudes and deciding a presence and extent of motion to output detected motion;

(c) extending an effect of the detected motion to pixels adjacent to a pixel to be interpolated;

(d) detecting an edge direction around the pixel to be interpolated; and

(e) interpolating the pixel according to an edge direction and a value from (c), and outputting a non-interlaced field.

12. A method of claim 11, wherein in (d), detecting an edge direction using the following equations:

13. A method of claim 12, wherein the value of Med₄₅, Med₀, and Med₋

45 are each compared with either an identical threshold value or respectively compared with three different threshold values, and an edge direction is determined to be present in at least one direction corresponding to the values Med₄₅, Med₀, and Med<sub>−

45</sub> depending upon the comparison.

14. A method of claim 12, wherein the values of Med₄₅, Med₀, and Med₋

45 are each compared to a threshold value E_th as follows;

If Med₄₅<

E_th, edge45°

=0 else edge45°

=Med₄₅

If Med₀<

E_th, edge0°

=0 else edge0°

=Med₀

If Med<sub>−

45</sub><

E_th, edge−

45°

=0 else edge−

45°

=Med<sub>−

45</sub>

and a edge direction corresponding to a minimum value among edge45°

, edge0°

, edge−

45°

is determined to be present.

15. A method of claim 14, wherein the edge direction is determined according to the following table where α

<

β and

l*m*n,

16. A method of claim 14, wherein a pixel is interpolated according to the following table,

17. A method of claim 11, wherein in (d) further obtaining a local correlation around a pixel to be interpolated and wherein in (e) interpolating according to the edge direction, a local correlation and the value from (c).

18. A method of claim 17, wherein the local correlation is obtained using one of three methods below, where V_th indicates a vertical direction threshold, H_th represent a horizontal direction threshold, and V_cor=0 indicates a local correlation in a vertical direction:

(1) V_cor=1 if

Med{|P(i, j−

1)−

P(i+1, j−

1)|, |P(i, j)−

P(i+1, j)|, |P(i, j+1)−

P(i+1, j+1)|}<

V_th;

(2) V_cor=1 if

|Med{P(i, j−

1), P(i, j), P(i, j+1)}−

Med{P(i+1, j−

1), P(i+1, j), P(i+1, j+1)}|<

V_th;

or

(3)

i) Hedge₁=0 if

|Med{P(i, j−

2), P(i, j−

1), P(i, j)}−

Med{P(i, j), P(i, j+1), P(i, j+2)}|<

H_th,

ii) Hedge₂=0 if

Med{P(i+1, j−

2), P(i+1, j−

1), P(i+1, j)}−

Med{P(i+1, j), P(i+1, j+1), P(i+1, j+2)}|<

H_th,

iii) Vedge=0 if

|Med{P(i, j−

1), P(i, j), P(i, j+1)}−

Med{P(i+1, j−

1), P(i+1, j), P(i+1, j+1)}|<

V_th,

where V_cor=NOT(Hedge₁ XOR Hedge₂ OR Vedge).

19. A method of claim 11, wherein (e) comprises:

mixing the edge direction and the value from (c), and outputting the mixed value to appropriately interpolate a pixel value; and

converting a number of vertical lines to conform with a required display format, by using data of the current interlaced field and interpolated pixels from the soft switch, and outputting a non-interlaced field.

20. A method of claim 11, further comprising eliminating a noise component in the detected motion in (b) and grouping portions with motion.