METHOD AND SYSTEM FOR DETERMINING A QUALITY MEASURE FOR AN IMAGE USING MULTI-LEVEL DECOMPOSITION OF IMAGES

• US 20110038548A1
• Filed: 10/17/2010
• Published: 02/17/2011
• Est. Priority Date: 02/11/2009
• Status: Active Grant

• Current Assignee: Ecole de technologie superieure
• Original Assignee: Ecole de technologie superieure
• Inventors: COULOMBE, Stephane, REZAZADEH, Soroosh
• Granted Patent: US 8,515,182 B2
• Time in Patent Office: Days
• US Class Current: 382/199
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06T 2207/20064 Wavelet transform [DWT]
• CPC Class Codes: G06T 2207/30168 Image quality inspection
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection

Abstract

Method and system for determining a measure of quality for images by using multi-level decomposition are presented. Multi-level decomposition of images is performed in the wavelet domain producing subbands at each level of decomposition. Aggregation of subbands is performed across multiple levels to produce an accurate measure of image quality. By aggregating only selected subbands the computational complexity of the method is greatly reduced.

Claims

1. A method for determining a measure of quality for a distorted image Y, characterizing a similarity between the image Y and an undistorted reference image X, having the same number of rows and columns of pixels as the image Y, the method comprising:

(a1) applying a N level multiresolution decomposition, comprising levels 1, 2, . . . i, i+1, . . . N, to the image X, to produce;

for each level i, with i ranging from 1 to N−

1, intermediate subbands of image X for processing at level 41; and

for the level N, an approximation subband containing main content of the image X and detail subbands containing edges of the image X;

(b1) applying said N level multiresolution decomposition, comprising levels 1, 2, . . . i, i+1, . . . N, to the image Y, to produce;

for each level i, with i ranging from 1 to N−

1, intermediate subbands of image Y for processing at level 41; and

for the level, N, an approximation subband containing main content of the image Y and detail subbands containing edges of the image Y;

(c1) applying an image quality metric (IQM) to the approximation subband of the image X and the approximation subband of the image Y to produce an approximation quality measure characterizing similarity between the main content of the image X and the main content of the image Y;

(d1) aggregating the intermediate subbands at the level i for the image X, with i ranging from levels 1 to N−

1, and the detail subbands of the image X to produce an edge map of the image X characterizing the edges of the image X;

(e1) aggregating the intermediate subbands at the level i for the image Y, with i ranging from 1 to N−

1, and the detail subbands of the image Y to produce an edge map of the image Y characterizing the edges of image Y;

(f1) applying the IQM between the edge map of the image X and the edge map of the image Y to produce an edge quality measure characterizing similarity between the edges of the image X and the edges of the image Y; and

(g1) processing the approximation quality measure and the edge quality measure to determine the measure of quality.

2. The method of claim 1, wherein:

(a2) the step (d1) further comprises;

(a2i) selecting the intermediate subbands at each level i, with i ranging from 1 to N-1, and the detail subbands for the image X based on an accuracy to be achieved in determining the measure of quality; and

(a2ii) aggregating only selected intermediate subbands and selected detail subbands for the image X; and

(b2) the step (e1) further comprises;

(b2i) selecting the intermediate subbands at said each level i and the detail subbands for the image Y based on said accuracy to be achieved in determining the measure of quality; and

(b2ii) aggregating only selected intermediate subbands and selected detail subbands for the image Y.

3. The method of claim 2, wherein the step (a2i) and the step (b2i) further comprise selecting the intermediate subbands at said each level i and the detail subbands for the image X and the image Y based on a number of the intermediate and the detailed subbands required for achieving the accuracy.

4. The method of claim 3, further comprising:

(a4) at the step (a2ii);

(a4i) for said each level i, aggregating the selected intermediate subbands for the image X producing an edge map at the level i for the image X;

(a4ii) aggregating the selected detail subbands for the image X producing an edge map at the level N for the image X; and

(a4iii) aggregating the edge map at said each level i for the image X with the edge map at the level N for the image X; and

(b4) at the step (b2ii);

(b4i) for said each level i, aggregating the selected intermediate subbands for the image Y producing an edge map at the level i for the image Y;

(b4ii) aggregating the selected detail subbands for the image Y producing an edge map at the level N for the image Y; and

(b4iii) aggregating the edge map at said each level i for the image Y with the edge map at the level N for the image Y.

5. The method of claim 4, wherein:

(a5) the step (a4iii) further comprises;

(a5i) multiplying the edge map at said each level i for the image X with a predetermined weight for the level i;

(a5ii) multiplying the edge map at the level N for the image X with a predetermined weight at the level N; and

(a5iii) summing products resulting from the multiplying in the step (a5i) and a product resulting from the multiplying in the step (a5ii); and

(b5) the step (b4iii) further comprises;

(b5i) multiplying the edge map at said each level i for the image Y with the predetermined weight for the level i;

(b5ii) multiplying the edge map at the level N for the image Y with the predetermined weight at the level N; and

(b5iii) summing products resulting from the multiplying in the step (b5i) and a product resulting from the multiplying in the step (b5ii).

6. The method of claim 5, wherein:

(a6) the step (a4i) comprises;

(a6i) for said each level i, squaring each selected intermediate subband for the image X;

(a6ii) multiplying a result of the squaring performed in the step (a6i) with a predetermined weight for said each selected intermediate subband for the image X;

(a6iii) summing products resulting from the multiplying performed in the step (a6ii);

(a6iv) applying a square root function to a result of the summing performed in the step (a6iii) producing the edge map at the level i for the image X;

(b6) the step (a4ii) comprises;

(b6i) squaring each said selected detail subband for the image X;

(b6ii) multiplying a result of the squaring performed in the step (b6i) with a predetermined weight for said each selected detail subband for the image X;

(b6iii) summing products resulting from the multiplying performed in the step (a6ii);

(b6iv) applying a square root function to a result of the summing performed in the step (a6iii) producing the edge map at the level N for the image X;

(c6) the step (b4i) comprises;

(c6i) for said each level i, squaring each selected intermediate subband for the image Y;

(c6ii) multiplying a result of the squaring performed in the step (c6i) with a predetermined weight for said each selected intermediate subband for the image Y;

(c6iii) summing products resulting from the multiplying performed in the step (c6ii);

(c6iv) applying a square root function to a result of the summing performed in the step (c6iii) producing the edge map at the level i for the image Y; and

(d6) the step (b4ii) comprises;

(d6i) squaring each said selected detail subband for the image Y;

(d6ii) multiplying a result of the squaring performed in the step (d6i) with a predetermined weight for said each selected detail subband for the image Y;

(d6iii) summing products resulting from the multiplying performed in the step (d6ii);

(d6iv) applying a square root function to a result of the summing performed in the step (d6iii) producing the edge map at the level N for the image Y.

7. The method of claim 3, wherein the intermediate subbands at said each level i for the image X and the image Y include for each respective image:

level i-detail subbands and level i-wavelet packet (WP) subbands;

the level i-WP subbands comprising level i-WP approximation subbands and level i-WP detail subbands; and

the level i-WP approximation subbands further comprising a level i-horizontal WP approximation subband, a level i-vertical WP approximation subband and a level i-diagonal WP approximation subband.

8. The method of claim 7, wherein:

(a8) the step (a2i) further comprises selecting one or more of the level i-WP approximation subbands and one or more of the detail subbands for the image X; and

(b8) the step (b2i) further comprises selecting one or more of the level i-WP approximation subbands and one or more of the detail subbands for the image Y.

9. The method of claim 8, wherein:

(a9) the step (a8) further comprises including one or more of the level i-detail subbands for the image X in the selecting; and

(b9) the step (b8) further comprises including one or more of the level i-detail subbands for the image Y in the selecting.

10. The method of claim 9, wherein:

(a10) the step (a2ii) further comprises;

(a10i) aggregating the level i-horizontal WP approximation subband for the image X at said each level i and the horizontal subband for the image X producing a horizontal edge map for the image X;

(a10ii) aggregating the level i-vertical WP approximation subband for the image X at said each level i and the vertical subband for the image X producing a vertical edge map for the image X;

(a10iii) aggregating the level i-diagonal WP approximation subband for the image X at said each level i and the diagonal subband for the image X producing a diagonal edge map for the image X; and

(a10iv) aggregating the horizontal edge map, the vertical edge map and the diagonal edge map for the image X; and

(b10) the step (b2ii) further comprises;

(b10i) aggregating the level i-horizontal WP approximation subband for the image Y at said each level i and the horizontal subband for the image Y producing a horizontal edge map for the image Y;

(b10ii) aggregating the level i-vertical WP approximation subband for the image Y at said each level i and the vertical subband for the image Y producing a vertical edge map for the image Y;

(b10iii) aggregating the level i-diagonal WP approximation subband for the image Y at said each level i and the diagonal subband for the image Y producing a diagonal edge map for the image Y; and

(b10iv) aggregating the horizontal edge map, the vertical edge map and the diagonal edge map for the image Y.

11. The method of claim 10, wherein the level i-WP detail subbands for the image X and the image Y further comprise for each respective image, a level i-WP horizontal subband, a level i-WP vertical subband and a level i-WP diagonal subband and:

(a11) the step (a10i) further comprises including the level i-WP horizontal subband for the image X in the aggregating;

(b11) the step (a10ii) further comprises including the level i-WP vertical subband for the image X in the aggregating;

(c11) the step (a10iii) further comprises including the level i-WP diagonal subband for the image X in the aggregating;

(d11) the step (b10i) further comprises including the level i-WP horizontal subband for the image Y in the aggregating;

(e11) the step (b10ii) further comprises including the level i-WP vertical subband for the image Y in the aggregating; and

(f11) the step (b10iii) further comprises including the level i-WP diagonal subband for the image Y in the aggregating.

12. The method of claim 10, wherein:

(a12) the step (a10iv) comprises;

(a12) applying a square root function to the horizontal edge map, the vertical edge map and the diagonal edge map for the image X; and

(b12) summing results of applying said square root function; and

(b12) the step (b10iv) comprises;

(a12) applying a square root function to the horizontal edge map, the vertical edge map and the diagonal edge map for the image Y; and

(b12) summing results of applying said square root function.

13. The method of claim 11, wherein:

(a13) the step (a10iv) comprises;

(a13i) summing the horizontal edge map, the vertical edge map and the diagonal edge map for the image X; and

(a13ii) applying a square root function to result of the summing performed in the step (a13i); and

(b13) the step (b10iv) comprises;

(b13i) summing the horizontal edge map, the vertical edge map and the diagonal edge map for the image Y; and

(b13ii) applying a square root function to result of the summing performed in the step (b13i).

14. The method of claim 1, wherein the steps (a1) and (b1) comprise applying a N level discrete wavelet transform (DWT).

15. The method of claim 14, wherein the discrete wavelet transform is a Haar transform.

16. The method of claim 14, wherein the discrete wavelet transform is one of a Newland transform, or wavelet transform using a Daubechies filter.

17. The method of claim 3, comprising one or more of the following:

(a17) at the step (a1);

(a17i) for the level 1, applying the multiresolution decomposition to the image X producing intermediate subbands at the level 1 for processing at level 2; and

(a17ii) for the level i, with i ranging from 2 to N, applying the multiresolution decomposition to one or more of the intermediate subbands produced by the multiresolution decomposition performed at the level i−

1; and

(b17) at the step (b1);

(b17i) for the level 1, applying the multiresolution decomposition to the Image Y producing intermediate subbands at the level 1 for processing at level 2; and

(b17ii) for the level i, with i ranging from 2 to N, applying the multiresolution decomposition to one or more of the intermediate subbands produced by the multiresolution decomposition performed at the level i−

1.

18. The method of claim 2, wherein the approximation quality measure is an approximation quality map, and the edge quality measure is an edge quality map, and the step (g1) further comprises:

(a18) generating a contrast map, including assigning corresponding values to the pixels of the approximation subband and the edge map of the image X and the image Y according to their respective importance to human visual system;

(b18) performing weighted pooling of the approximation quality map using the contrast map to produce an approximation quality score;

(c18) performing weighted pooling of the edge quality map using the contrast map to produce an edge quality score; and

(d18) combining the approximation similarity score from the step (b18) with the edge similarity score from the step (c18) to determine the measure of quality.

19. The method of claim 1, further comprising the step of determining N as a function of a minimum size of the approximation subband, S, which produces a substantially peak response for human visual system.

20. The method of claim 1, wherein:

(a20) the step (c1) comprises applying a structural similarity (SSIM) IQM to the approximation subband of the image X and the approximation subband of the image Y to produce an approximation SSIM map, SSIM_A;

(b20) the step (f1) comprises applying the SSIM IQM between the edge map of the image X and the edge map of the image Y to produce an edge SSIM map, SSIM_E; and

(c20) the step (g1) comprises processing the SSIM_A and the SSIM_E to determine a SSIM_DWT score as the measure of quality.

21. The method of claim 1, wherein:

(a21) the step (c1) comprises applying an Absolute Difference (AD) IQM to the approximation subband of the image X and the approximation subband of the image Y to produce an approximation AD map, AD_A;

(b21) the step (f1) comprises applying the AD IQM between the edge map of the image X and the edge map of the image Y to produce an edge AD map, AD_E; and

(c21) the step (g1) comprises processing the AD_A and the AD_E to determine an AD_DWT score as the measure of quality.

22. The method of claim 1, wherein:

(a22) the step (c1) comprises applying a peak-signal-to-noise ratio (PSNR) IQM to the approximation subband of the image X and the approximation subband of the image Y to produce a PSNR approximation quality score, PSNR_A;

(b22) the step (f1) comprises applying the PSNR IQM between the edge map of the image X and the edge map of the image Y to produce a PSNR edge quality score, PSNR_E; and

(c22) the step (g1) comprises processing the PSNR_A and the PSNR_E to determine a PSNR_DWT score as the measure of quality.

23. The method of claim 1, wherein:

(a23) the step (c1) comprises applying a Visual Information Fidelity (VIF) IQM to the approximation subband of the image X and the approximation subband of the image Y to produce a VIF approximation quality score, VIF_A;

(b23) the step (f1) comprises applying the VIF IQM between the edge map of the image X and the edge map of the image Y to produce a VIF edge quality score, VIF_E; and

(c23) the step (g1) comprises processing the VIF_A and the VIF_E to determine a VIF_DWT score as the measure of quality.

24. The method of claim 2, wherein:

(a24) the step (a2) further comprises choosing the selected intermediate subbands at said each level i and the detail subbands for the image X, having substantially same resolution, for aggregating; and

(b24) the step (b2) further comprises choosing the selected intermediate subbands at said each level i and the detail subbands for the image Y, having substantially same resolution, for aggregating.

25. A system for determining a measure of quality for a distorted image Y, characterizing a similarity between the image Y and an undistorted reference image X, having the same number of rows and columns of pixels as the image Y, the system comprising:

a processor, and a computer readable storage medium having computer readable instructions stored thereon, which, when executed by the processor, form the following;

(a25) a first decomposition unit applying a N level multiresolution decomposition, comprising levels 1, 2, . . . i, i+1, . . . N, to the image X, to produce;

for each level i, with i ranging from 1 to N−

1, intermediate subbands of image X for processing at level i+1; and

for the level N, an approximation subband containing main content of the image X and detail subbands containing edges of the image X;

(b25) a second decomposition unit applying said N level multiresolution decomposition, comprising levels 1, 2, . . . i, i+1, . . . N, to the image Y, to produce;

for each level i, with i ranging from 1 to N−

1, intermediate subbands for image Y; and

for the level, N, an approximation subband containing main content of the image Y and detail subbands containing edges of the image Y;

(c25) an approximation quality measure unit applying an image quality metric (IQM) to the approximation subband of the image X and the approximation subband of the image Y to produce an approximation quality measure characterizing similarity between the main content of the image X and the main content of the image Y;

(d25) a first aggregation unit aggregating the intermediate subbands at the level i for the image X, with i ranging from levels 1 to N−

1 and the detail subbands of the image X to produce an edge map of the image X characterizing the edges of the image X;

(e25) a second aggregation unit aggregating the intermediate subbands at the level i for the image Y, with i ranging from 1 to N−

1 and the detail subbands of the image Y to produce an edge map of the image Y characterizing the edges of image Y;

(f25) an edge quality measure unit applying the IQM between the edge map of the image X and the edge map of the image Y to produce an edge quality measure characterizing similarity between the edges of the image X and the edges of the image Y; and

(g25) a quality measure unit processing the approximation quality measure and the edge quality measure to determine the measure of quality.

26. The system of claim 25, wherein:

(a26) the first aggregation unit (d25) further comprises;

(a26i) a first selection unit selecting the intermediate subbands at each level i, with i ranging from 1 to N−

1, and the detail subbands for the image X based on an accuracy to be achieved in determining the measure of quality; and

(a26ii) a first selective aggregation unit aggregating only selected intermediate subbands and selected detail subbands for the image X; and

(b26) the second aggregation unit (e1) further comprises;

(b26i) a second selection unit selecting the intermediate subbands at said each level i and the detail subbands for the image Y based on said accuracy to be achieved in determining the measure of quality; and

(b26ii) a second selective aggregation unit aggregating only selected intermediate subbands and selected detail subbands for the image Y.

27. The system of claim 26, wherein:

(a27) the first selection unit (a26i) further comprises a first selection sub-unit selecting the intermediate subbands at said each level i and the detail subbands for the image X based on a number of the intermediate and the detailed subbands required for achieving the accuracy; and

(b27) the second selection unit (b26i) further comprises a second selection sub-unit selecting the intermediate subbands at said each level i and the detail subbands for the image Y based on a number of the intermediate and the detailed subbands required for achieving the accuracy.

28. The system of claim 27, further comprising:

(a28) at the first selective aggregation unit (a26ii);

(a28i) a first level i-intermediate subbands aggregation unit aggregating, for said each level i, the selected intermediate subbands for the image X producing an edge map at the level i for the image X;

(a28ii) a first detail subbands aggregation unit aggregating the selected detail subbands for the image X producing an edge map at the level N for the image X; and

(a28iii) a first edge map determination unit aggregating the edge map at said each level i for the image X with the edge map at the level N for the image X; and

(b28) at the second selective aggregation unit (b26ii);

(b28i) a second level i-intermediate subbands aggregation unit aggregating, for said each level i, the selected intermediate subbands for the image Y producing an edge map at the level i for the image Y;

(b28ii) a second detail subbands aggregation unit aggregating the selected detail subbands for the image Y producing an edge map at the level N for the image Y; and

(b28iii) a second edge map determination unit aggregating the edge map at said each level i for the image Y with the edge map at the level N for the image Y.

29. The system of claim 28, wherein:

(a29) the first edge map determination unit (a28iii) further comprises a first computation unit for;

multiplying the edge map at said each level i for the image X with a predetermined weight for the level i;

multiplying the edge map at the level N for the image X with a predetermined weight for the level N; and

summing products resulting from the multiplying the edge map at said each level i for the image X and the multiplying the edge map at the level N for the image X.

(b29) the second edge map determination unit (b28iii) further comprises a second computation unit for;

multiplying the edge map at said each level i for the image Y with the predetermined weight for the level i;

multiplying the edge map at the level N for the image Y with the predetermined weight for the level N; and

summing products resulting from the multiplying the edge map at said each level i for the image Y and the multiplying the edge map at the level N for the image Y.

30. The system of claim 29, wherein:

(a30) the first level i-intermediate subbands aggregation unit (a28i) comprises a first level i-computation unit for;

squaring each selected intermediate subband for the image X for said each level i;

multiplying a result of the squaring with a predetermined weight for said each selected intermediate subband for the image X;

summing products resulting from the multiplying; and

applying a square root function to a result of the summing producing the edge map at the level i for the image X;

(b30) the first detail subbands aggregation unit (a28ii) comprises a first detail subbands computation unit for;

squaring each said selected detail subband for the image X;

multiplying a result of the squaring with a predetermined weight for said each selected detail subband for the image X;

summing products resulting from the multiplying; and

applying a square root function to a result of the summing producing the edge map at the level N for the image X;

(c30) the second level i-intermediate subbands aggregation unit (b28i) comprises a second level i-computation unit for;

squaring each said selected intermediate subband for the image Y for said each level i;

multiplying a result of the squaring performed with a predetermined weight for said each selected intermediate subband for the image Y;

summing products resulting from the multiplying;

applying a square root function to a result of the summing producing the edge map at the level i for the image Y; and

(d30) the second detail subbands aggregation unit (b28ii) comprises a second detail subbands computation unit for;

squaring each said selected detail subband for the image Y;

multiplying a result of the squaring with a predetermined weight for said each selected detail subband for the image Y;

summing products resulting from the multiplying; and

applying a square root function to a result of the summing producing the edge map at the level N for the image Y.

31. The system of claim 27, wherein the intermediate subbands at said each level i for the image X and the image Y include for each respective image:

level i-detail subbands and level i-wavelet packet (WP) subbands;

the level i-WP subbands comprising level i-WP approximation subbands and level i-WP detail subbands; and

the level-WP approximation subbands further comprising a level i-horizontal WP approximation subband, a level i-vertical WP approximation subband and a level i-diagonal WP approximation subband.

32. The system of claim 31, wherein:

(a32) the first selection unit (a26i) further comprises a first WP-detail selection sub-unit selecting one or more of the level i-WP approximation subbands and one or more of the detail subbands for the image X; and

(b32) the second selection unit step (b26i) further comprises a second WP-detail selection sub-unit selecting one or more of the level i-WP approximation subbands and one or more of the detail subbands for the image Y.

33. The system of claim 32, wherein:

(a33) the first WP-detail selection sub-unit (a32) further comprises a first detail-selection module including one or more of the level i-detail subbands for the image X in the selecting; and

(b33) the second WP-detail selection sub-unit (b32) further comprises a second detail-selection module including one or more of the level i-detail subbands for the image Y in the selecting.

34. The system of claim 33, wherein:

(a34) the first selective aggregation unit (a26ii) further comprises;

(a34i) a first horizontal edge map unit aggregating the level i-horizontal WP approximation subband for the image X at said each level i and the horizontal subband for the image X producing a horizontal edge map for the image X;

(a34ii) a first vertical edge map unit aggregating the level i-vertical WP approximation subband for the image X at said each level i and the vertical subband for the image X producing a vertical edge map for the image X;

(a34iii) a first diagonal edge map unit aggregating the level i-diagonal WP approximation subband for the image X at said each level i and the diagonal subband for the image X producing a diagonal edge map for the image X; and

(a34iv) a first combination unit aggregating the horizontal edge map, the vertical edge map and the diagonal edge map for the image X; and

(b34) the second selective aggregation unit (b26ii) further comprises;

(b34i) a second horizontal edge map unit aggregating the level i-horizontal WP approximation subband for the image Y at said each level i and the horizontal subband for the image Y producing a horizontal edge map for the image Y;

(b34ii) a second vertical edge map unit aggregating the level i-vertical WP approximation subband for the image Y at said each level i and the vertical subband for the image Y producing a vertical edge map for the image Y;

(b34iii) a second diagonal edge map unit aggregating the level i-diagonal WP approximation subband for the image Y at said each level i and the diagonal subband for the image Y producing a diagonal edge map for the image Y; and

(b34iv) a second combination unit aggregating the horizontal edge map, the vertical edge map and the diagonal edge map for the image Y.

35. The system of claim 34, wherein the level i-WP detail subbands for the image X and the image Y further comprise for each respective image, a level i-WP horizontal subband, a level i-WP vertical subband and a level i-WP diagonal subband;

and

(a35) the first horizontal edge map unit (a34i) further comprises a first horizontal sub-unit including the level i-WP horizontal subband for the image X in the aggregating;

(b35) the first vertical edge map unit (a34ii) further comprises a first vertical sub-unit including the level i-WP vertical subband for the image X in the aggregating;

(c35) the first diagonal edge map unit (a34iii) further comprises a first vertical sub-unit including the level i-WP diagonal subband for the image X in the aggregating;

(d35) the second horizontal edge map unit (b34i) further comprises a second horizontal sub-unit including the level i-WP horizontal subband for the image Y in the aggregating;

(e35) the second vertical edge map unit (b34ii) further comprises a second vertical sub-unit including the level i-WP vertical subband for the image Y in the aggregating; and

(f35) the second diagonal edge map unit (b34iii) further comprises a second diagonal sub-unit including the level i-WP diagonal subband for the image Y in the aggregating.

36. The system of claim 34, wherein:

(a36) the first combination unit (a34iv) comprises a first combination sub-unit for;

applying a square root function to the horizontal edge map, the vertical edge map and the diagonal edge map for the image X; and

summing results of applying said square root function; and

(b36) the second combination unit (b34iv) comprises a second combination sub-unit for;

applying a square root function to the horizontal edge map, the vertical edge map and the diagonal edge map for the image Y; and

summing results of applying said square root function.

37. The system of claim 35, wherein:

(a37) the first combination unit (a34iv) comprises a third combination sub-unit for;

summing the horizontal edge map, the vertical edge map and the diagonal edge map for the image X; and

applying a square root function to result of the summing; and

(b37) the second combination unit (b34iv) comprises a fourth combination sub-unit for;

summing the horizontal edge map, the vertical edge map and the diagonal edge map for the image Y; and

applying a square root function to result of the summing.

38. The system of claim 25, wherein the first decomposition unit (a1) and the second decomposition unit (b1) comprise computational means for applying a N level discrete wavelet transform (DWT).

39. The system of claim 38, wherein the discrete wavelet transform is a Haar transform.

40. The system of claim 38, wherein the discrete wavelet transform is one of a Newland transform, or wavelet transform using a Daubechies filter.

41. The system of claim 27, comprising one or more of the following:

(a41) at the first decomposition unit (a25);

(a411) a first level 1 decomposition unit applying the multiresolution decomposition to the image X producing intermediate subbands at the level 1 for processing at level 2;

(a41ii) a first level i decomposition unit applying the multiresolution decomposition to one or more of the intermediate subbands produced by the multiresolution decomposition performed at the level i−

1, with i ranging from 2 to N;

(b41) at the second decomposition unit (b25);

(b41i) a second level 1 decomposition unit applying the multiresolution decomposition to the Image Y producing intermediate subbands at the level 1 for processing at level 2;

(b41ii) a second level i decomposition unit applying the multiresolution decomposition to one or more of the intermediate subbands produced by the multiresolution decomposition performed at the level i−

1, with i ranging from 2 to N.

42. The system of claim 26, wherein the approximation quality measure is an approximation quality map, and the edge quality measure is an edge quality map, and the quality measure unit (g25) further comprises:

(a42) a contrast map unit generating a contrast map, including assigning corresponding values to the pixels of the approximation subband and the edge map of the image X and the image Y according to their respective importance to human visual system;

(b42) a first pooling unit performing weighted pooling of the approximation quality map using the contrast map to produce an approximation quality score;

(c42) a second pooling unit performing weighted pooling of the edge quality map using the contrast map to produce an edge quality score; and

(d42) a score combination unit combining the approximation similarity score from the step (b42) with the edge similarity score from the step (c42) to determine the measure of quality.

43. The system of claim 25, further comprising computational means for determining N as a function of a minimum size of the approximation subband, S, which produces a substantially peak response for human visual system.

44. The system of claim 25, wherein:

(a44) the approximation quality measure unit (c25) comprises an approximation SSIM map unit applying a SSIM IQM to the approximation subband of the image X and the approximation subband of the image Y to produce an approximation structural similarity (SSIM) map, SSIM_A;

(b44) the edge quality measure unit (f25) comprises an edge SSIM map unit applying the SSIM IQM between the edge map of the image X and the edge map of the image Y to produce an edge SSIM map, SSIM_E; and

(c44) the quality measure unit (g25) comprises a SSIM_DWT score unit processing the SSIM_A and the SSIM_E to determine a SSIM_DWT score as the measure of quality.

45. The system of claim 25, wherein:

(a45) the approximation quality measure unit (c25) comprises an approximation AD map unit applying an AD IQM to the approximation subband of the image X and the approximation subband of the image Y to produce an approximation AD map, AD_A;

(b45) the edge quality measure unit (f25) comprises an edge AD map unit applying the AD IQM between the edge map of the image X and the edge map of the image Y to produce an edge AD map, AD_E; and

(c45) the quality measure unit (g25) comprises an AD_DWT score unit processing the AD_A and the AD_E to determine an AD_DWT score as the measure of quality.

46. The system of claim 25, wherein:

(a46) the approximation quality measure unit (c25) comprises a PSNR approximation quality score unit applying a PSNR IQM to the approximation subband of the image X and the approximation subband of the image Y to produce a PSNR approximation quality score, PSNR_A;

(b46) the edge quality measure unit (f25) comprises a PSNR edge quality score determination unit applying the PSNR IQM between the edge map of the image X and the edge map of the image Y to produce a PSNR edge quality score, PSNR_E; and

(c46) the quality measure unit (g25) comprises a PSNR_DWT score unit processing the PSNR_A and the PSNR_E to determine a PSNR_DWT score as the measure of quality.

47. The system of claim 25, wherein:

(a47) the approximation quality measure unit (c25) comprises a VIF approximation quality score unit applying a VIF IQM to the approximation subband of the image X and the approximation subband of the image Y to produce a VIF approximation quality score, VIF_A;

(b47) the edge quality measure unit (f25) comprises a VIF edge quality score determination unit applying the VIF IQM between the edge map of the image X and the edge map of the image Y to produce a VIF edge quality score, VIF_E; and

(c47) the quality measure unit (g25) comprises a VIF_DWT score unit processing the VIF_A and the VIF_E to determine a VIF_DWT score as the measure of quality.

48. The system of claim 26, wherein:

(a48) the first aggregation unit (a26) further comprises computational means for choosing the selected intermediate subbands at said each level i and the detail subbands for the image X, having substantially same resolution, for aggregating; and

(b48) the second aggregation unit (b26) further comprises computational means for choosing the selected intermediate subbands at said each level i and the detail subbands for the image Y, having substantially same resolution, for aggregating.

49. A computer readable storage medium, having computer readable program code instructions stored thereon, which, when executed by a processor, perform the following:

(a49) applying a N level multiresolution decomposition, comprising levels 1, 2, . . . i, i+1, . . . N, to the image X, to produce;

for each level i, with i ranging from 1 to N−

1, intermediate subbands of image X for processing at level i+1; and

for the level N, an approximation subband containing main content of the image X and detail subbands containing edges of the image X;

(b49) applying said N level multiresolution decomposition, comprising levels 1, 2, . . . i, i+1, . . . N, to the image Y, to produce;

for each level i, with i ranging from 1 to N−

1, intermediate subbands of image Y for processing at level i+1; and

for the level, N, an approximation subband containing main content of the image Y and detail subbands containing edges of the image Y;

(c49) applying an image quality metric (IQM) to the approximation subband of the image X and the approximation subband of the image Y to produce an approximation quality measure characterizing similarity between the main content of the image X and the main content of the image Y;

(d49) aggregating the intermediate subbands at the level i for the image X, with i ranging from levels 1 to N−

1, and the detail subbands of the image X to produce an edge map of the image X characterizing the edges of the image X;

(e49) aggregating the intermediate subbands at the level i for the image Y, with i ranging from 1 to N−

1, and the detail subbands of the image Y to produce an edge map of the image Y characterizing the edges of image Y;

(f49) applying the IQM between the edge map of the image X and the edge map of the image Y to produce an edge quality measure characterizing similarity between the edges of the image X and the edges of the image Y; and

(g49) processing the approximation quality measure and the edge quality measure to determine the measure of quality.