1. A method of transcoding a first image file to a second image file, wherein the first image file comprises a single codestream and a table comprising references associated with different parsed forms of the single codestream for referencing segments of codestream header data and segments of the single codestream, the method comprising:
selecting a parsed form of the single image codestream;
locating, by using the table, those codestream segments associated with the selected parsed form;
locating, by using the table, those codestream header data associated with the selected parsed form; and
writing the located codestream header data and codestream segments as a single codestream of the selected parsed form to the second image file.
2. A method as claimed in claim 1, wherein the references associated with a selected parsed form describe a sequence of byte ranges that are to be spliced together from the segments of codestream header data and the segments of the single codestream and written as the single codestream of the selected parsed form to the second image file.
3. A method as claimed in claim 1, wherein the first image file comprises a label that is descriptive of the parsed form it contains.
4. A method as claimed in claim 1, wherein the second image file is JP2 compatible.
5. A method as claimed in claim 1, wherein the first image file is JPX compatible.
6. A method as claimed in claim 1, wherein each parsed form corresponds to a reduced resolution level of an original image.
7. A method as claimed in claim 1, wherein the parsed forms comprise one or more distinct composition layers of an original image.
8. A method as claimed in claim 7, wherein the number of composition layers in each parsed form is equal to the number of resolutions of the original image.
9. A method as claimed in claim 8, wherein the nth composition layer for the nth resolution yields a fixed effective bit rate for all n.
10. A method according to claim 1, wherein the first image file comprises a first coded representation of an original image and the second image file comprises a second coded representation of a reduced resolution of the original image, the image files each being a corresponding parsed form of the single image codestream, wherein the original image was previously encoded using an S-level DWT and blocks of DWT coefficients were previously coded in sub-passes to provide the first coded representation, the method comprising the steps of:
determining a desired reduced resolution level J of the first coded representation, wherein J is less than S;
determining using the table a number of sub passes to extract from each coded block of the first coded representation based on a predetermined image quality at the desired resolution level J; and
extracting those determined sub-passes to form the second coded representation of the original image, wherein a reconstructed image of the second coded representation of the original image is not significantly degraded visually compared with a reconstructed image of the first coded representation at the desired resolution level J.
11. A method as claimed in claim 10, wherein said step of determining a number of sub-passes to extract is based on a predetermined target bit rate threshold at the resolution J, and the predetermined target bit rate is a function of the resolution level J.
12. The method as claimed in claim 10, further comprising the step of producing a coded representation.
13. The method as claimed in claim 1, further comprising the step of producing an image file.
14. A method of generating a first image file from an original image, the method comprising the steps of:
generating a single codestream representative of the original image, wherein said single codestream comprises a plurality of different parsed forms of the image; and
generating a first image file, wherein the first image file comprises segments of the single codestream and segments of codestream header data associated with different parsed forms of the original image, and a table comprising references associated with the different parsed forms of the original image for referencing segments of codestream header data and segments of the single codestream.
15. The method as claimed in claim 14, further comprising the step of producing an image file.
16. A method as claimed in claim 14, wherein the references associated with a parsed form describe a sequence of byte ranges that are to be spliced together from the segments of codestream header data and the segments of the single codestream.
17. A method as claimed in claim 14, wherein the first image file comprises a label that is descriptive of the parsed form it contains.
18. A method as claimed in claim 14, wherein the first image file is JPX compatible.
19. A method as claimed in claim 14, wherein each parsed form corresponds to a reduced resolution level of the original image.
20. A method as claimed in claim 14, wherein the parsed forms comprise one or more distinct composition layers.
21. A method as claimed in claim 20, wherein the number of composition layers in each parsed form is equal to the number of resolutions of the original image.
22. A method as claimed in claim 21, wherein the nth composition layer for the nth resolution yields a fixed effective bit rate for all n.
23. A method as claimed in claim 14, wherein the original image is encoded as a single tile and the codestream thereof is divided into tile parts with packet length and packet header markers, when present, provided per tile part and so that the markers for each tile part comprise information for that tile part.
24. A method as claimed in claim 23, wherein tile part boundaries correspond to fragment boundaries.
25. A method according to claim 14, wherein said step of generating the single codestream comprises the sub-steps of:
transforming the original image with an S-level DWT transform to produce a plurality of subbands of DWT coefficients;
quantizing the DWT coefficients of each subband according to a quantisation table;
dividing each subband into one or more codeblocks of quantized DWT coefficients; entropy encoding the codeblocks to form a number of coded sub passes; and
arranging the coded sub-passes into a number of block layers, wherein the codeblocks in resolution levels greater than resolution level l make no contributions to block layers 0,1 . . . ,l, so as to enable a reconstruction of a resolution of the original image of just visually acceptable quality.
26. A method as claimed in claim 25, wherein the number of coded sub-passes of any codeblock arranged in any block layer is predetermined.
27. A method as claimed in claim 25, wherein the generating of the single codestream comprises the additional sub-step of:
arranging the block layers into the codestream according to a layer\u2014resolution level progression order.
28. A method as claimed in claim 25, wherein the quantization table comprises:
Sub-
band
level
1
2
3
4
5
6
7
LL
\u20023.5241
1.5655
1.2589
1.2589
1.2589
1.2589
1.2589
HL
\u20029.7500
2.5877
1.6696
1.5932
1.5932
1.5932
\u20091.5932,
LH
HH
24.5784
4.3179
1.8491
1.3709
1.3709
1.3709
\u20091.3709.
29. A method as claimed in claim 25, wherein said transforming sub-step separately transforms image components of the image using different level DWTs.
30. The method as claimed in claim 25, further comprising the step of producing a coded representation.
31. A method of decoding a first image file to provide an image, wherein the first image file comprises a single codestream and a table comprising references associated with different parsed forms of the single codestream for referencing segments of codestream header data and segments of the single code stream, the method comprising:
selecting a parsed form of the single image codestream;
locating, by using the table, those codestream segments associated with the selected parsed form;
locating, by using the table, those codestream header data associated with the selected parsed form;
constructing a single codestream of the selected parsed form from the located codestream header data and codestream segments; and
decoding the constructed single codestream to produce an image of the selected parsed form.
32. A method as claimed in claim 31, wherein the references associated with a selected parsed form describe a sequence of byte ranges of the segments of codestream header data and the segments of the single codestream.
33. A method as claimed in claim 31, wherein the first image file comprises a label that is descriptive of the parsed form it contains.
34. A method as claimed in claim 31, wherein the first image file is JPX compatible.
35. A method as claimed in claim 31, wherein each parsed form corresponds to a reduced resolution level of an original image.
36. A method as claimed in claim 31, wherein the parsed forms comprise one or more distinct composition layers of an original image.
37. A method as claimed in claim 36, wherein the number of composition layers in each parsed form is equal to the number of resolutions of the original image.
38. A method as claimed in claim 37, wherein the nth composition layer for the nth resolution yields a fixed effective bit rate for all n.
39. A method according to claim 31, wherein the first image file is a coded representation of an original image, wherein the original image was previously encoded using an S-level DWT transform and blocks of DWT coefficients were previously coded in sub-passes to provide the coded representation, being one of the parsed forms, the method comprising;
determining a desired reduced resolution level J to decode, wherein J is less than S;
determining, using the table, a number of sub-passes to discard for each coded block based on a predetermined image quality at the resolution level J;
discarding the determined sub-passes from the code blocks; and
decoding those sub-passes not discarded to reconstruct the image at the desired reduced resolution level J, wherein the reconstructed image is not significantly degraded visually compared with an image reconstructed from the coded representation at the desired resolution level J using all sub-passes.
40. A method as claimed in claim 39, wherein said step of determining a number of sub-passes to discard is based on a predetermined target bit rate threshold at the resolution J, and the predetermined target bit rate is a function of the resolution level J.
41. A method according to claim 31, the image being encoded to provide at least one parsed form using an S-level DWT, wherein S is an integer and wherein blocks of DWT coefficients are coded in sub-passes, the method comprising:
determining a desired resolution J to decode, wherein J is less than S;
determining, using the table, a number of sub-passes to extract for each block code based on the resolution;
extracting determined sub-passes from the block codes; and
reconstructing the image from the extracted sub-passes at the desired reduced resolution J, wherein the reconstructed image is not significantly degraded visually compared with an image reconstructed from the coded representation at the desired resolution J using all sub-passes.
42. Apparatus for transcoding a first image file to a second image file, wherein the first image file comprises a single codestream and a table comprising references associated with different parsed forms of the single codestream for referencing segments of codestream header data and segments of the single codestream, the apparatus comprising:
means for selecting a parsed form of the single image codestream:
means for locating, by using the table, those codestream segments associated with the selected parsed form;
means for locating, by using the table, those codestream header data associated with the selected parsed form; and
means for writing the located codestream header data and codestream segments as a single codestream of the selected parsed form to the second image file.
43. Apparatus according to claim 42, wherein the first image file comprises a first coded representation of an original image and the second image file comprises a second coded representation of a reduced resolution of the original image, the image files each being a corresponding parsed form of the single image codestream, wherein the original image was previously encoded using an S-level DWT and blocks of DWT coefficients were previously coded in sub-passes to provide the first coded representation, the apparatus comprising:
means for determining a desired reduced resolution level J of the first coded representation, wherein J is less than S;
means for determining, using the table, a number of sub-passes to extract from each coded block of the first coded representation based on a predetermined image quality at the desired resolution level J; and
means for extracting those determined sub-passes to form the second coded representation of the original image, wherein a reconstructed image of the second coded representation of the original image is not significantly degraded visually compared with a reconstructed image of the first coded representation at the desired resolution level J.
44. Apparatus for generating a first image file from an original image, the apparatus comprising:
means for generating a single code stream representative of the original image, wherein the single codestream comprises a plurality of different parsed forms of the image; and
means for generating a first image file, wherein the first image file comprises segments of the single codestream and segments of code stream header data associated with different parsed forms of the original image, and a table comprising references associated with the different parsed forms of the original image for referencing segments of codestream header data and segments of the single codestream.
45. Apparatus according to claim 44, wherein the means for generating the single codestream comprises:
means for transforming the original image with an S-level DWT transform to produce a plurality of subbands of DWT coefficients;
means for quantizing the DWT coefficients of each subband according to a quantization table;
means for dividing each subband into one or more codeblocks of quantized DWT coefficients; and
means for entropy encoding the codeblocks to form a number of coded subpasses and for arranging the coded sub-passes into a number of block layers, wherein the codeblocks in resolution levels greater than resolution level l make no contributions to block layers 0,1 . . . ,l, so as to enable a reconstruction of a resolution of the original image of just visually acceptable quality.
46. Apparatus for decoding a first image file to provide an image, wherein the first image file comprises a single codestream and a table comprising references associated with different parsed forms of the single codestream for referencing segments of codestream header data and segments of the single codestream, the apparatus comprising:
means for selecting a parsed form of the single image codestream:
means for locating, by using the table, those codestream segments associated with the selected parsed form;
means for locating, by using the table, those codestream header data associated with the selected parsed form;
means for constructing a single codestream of the selected parsed form from the located codestream header data and codestream segments; and
means for decoding the constructed single codestream to produce an image of the selected parsed form.
47. Apparatus according to claim 46, wherein the first image file is a coded representation of an original image, wherein the original image was previously encoded using an S-level DWT transform and blocks of DWT coefficients were previously coded in sub-passes to provide the coded representation, being one of the parsed forms, the apparatus comprising:
means for determining a desired reduced resolution level J to decode, wherein J is less than S;
means for determining, using the table, a number of sub-passes to discard for each coded block based on a predetermined image quality at the resolution level J;
means for discarding the determined sub-passes from the code blocks; and
means for decoding those sub-passes not discarded to reconstruct the image at the desired reduced resolution level J, wherein the reconstructed image is not significantly degraded visually compared with an image reconstructed from the coded representation at the desired resolution level J using all sub passes.
48. Apparatus according to claim 46, the image being encoded to provide at least one parsed form using an S-level DWT, wherein S is an integer and wherein blocks of DWT coefficients are coded in sub-passes, the apparatus comprising:
means for determining a desired resolution J to decode, wherein J is less than S;
means for determining using the table a number of sub-passes to extract for each block code based on the resolution;
means for extracting determined sub-passes from the block codes; and
means for reconstructing the image from the extracted sub-passes at the desired reduced resolution J, wherein the reconstructed image is not significantly degraded visually compared with an image reconstructed from the coded representation at the desired resolution J using all sub-passes.
49. A program stored on a computer-readable medium which, when executed, performs a method for transcoding a first image file to a second image file, wherein the first image file comprises a single codestream and a table comprising references associated with different parsed forms of the single codestream for referencing segments of codestream header data and segments of the single codestream, the program comprising:
code for selecting a parsed form of the single image codestream:
code for locating, by using the table, those codestream segments associated with the selected parsed form;
code for locating, by using the table, those codestream header data associated with the selected parsed form; and
code for writing the located codestream header data and codestream segments as a single codestream of the selected parsed form to the second image file.
50. A program stored on a computer-readable medium according to claim 49, wherein the first image file comprises a first coded representation of an original image and the second image file comprises a second coded representation of a reduced resolution of the original image, the image files each being a corresponding parsed form of the single image codestream, wherein the original image was previously encoded using an S-level DWT and blocks of DWT coefficients were previously coded in sub-passes to provide the first coded representation, the program further comprising:
code for determining a desired reduced resolution level J of the first coded representation, wherein J is less than S;
code for determining using the table a number of sub-passes to extract from each coded block of the first coded representation based on a predetermined image quality at the desired resolution level J; and
code for extracting those determined sub-passes to form the second coded representation of the original image, wherein a reconstructed image of the second coded representation of the original image is not significantly degraded visually compared with a reconstructed image of the first coded representation at the desired resolution level J.
51. A program stored on a computer-readable medium which, when executed, performs a method for generating a first image file from an original image, the program comprising:
code for generating a single codestream representative of the original image, wherein the single codestream comprises a plurality of different parsed forms of the image; and
code for generating a first image file, wherein the first image file comprises segments of the single codestream and segments of codestream header data associated with different parsed forms of the original image, and a table comprising references associated with the different parsed forms of the original image for referencing segments of codestream header data and segments of the single codestream.
52. A program stored on a computer-readable medium according to claim 51, wherein the code for generating the single codestream comprises:
code for transforming the original image with an S-level DWT transform to produce a plurality of subbands of DWT coefficients;
code for quantizing the DWT coefficients of each subband according to a quantization table;
code for dividing each subband into one or more codeblocks of quantized DINT coefficients; and
code for entropy encoding the codeblocks to form a number of coded sub-passes and for arranging the coded sub passes into a number of block layers, wherein the codeblocks in resolution levels greater than resolution level l make no contributions to block layers 0,1 . . . ,l, so as to enable a reconstruction of a resolution of the original image of just visually acceptable quality.
53. A program stored on a computer-readable medium which, when executed, performs a method for decoding a first image file to provide an image, wherein the first image file comprises a single codestream and a table comprising references associated with different parsed forms of the single codestream for referencing segments of codestream header data and segments of the single codestream, the program comprising:
code for selecting a parsed form of the single image code stream:
code for locating, by using the table, those codestream segments associated with the selected parsed form;
code for locating, by using the table, those codestream header data associated with the selected parsed form;
code for constructing a single code stream of the selected parsed form from the located codestream header data and codestream segments; and
code for decoding the constructed single codestream to produce an image of the selected parsed form.
54. A program stored on a computer-readable medium according to claim 53, wherein the first image file is a coded representation of an original image, wherein the original image was previously encoded using an S-level DWT transform and blocks of DWT coefficients were previously coded in sub-passes to provide the coded representation, being one of the parsed forms, the program further comprising:
code for determining a desired reduced resolution level J to decode, wherein J is less than S;
code for determining using the table a number of sub-passes to discard for each coded block based on a predetermined image quality at the resolution level J;
code for discarding the determined sub-passes from the code blocks; and
code for decoding those sub-passes not discarded to reconstruct the image at the desired reduced resolution level J, wherein the reconstructed image is not significantly degraded visually compared with an image reconstructed from the coded representation at the desired resolution level J using all sub-passes.
55. A program stored on a computer-readable medium according to claim 53, the image being encoded to provide at least one parsed form using an S-level DWT, wherein S is an integer and wherein blocks of DWT coefficients are coded in sub-passes, the program further comprising:
code for determining a desired resolution J to decode, wherein J is less than S;
code for determining using the table a number of sub-passes to extract for each block code based on the resolution;
code for extracting determined sub-passes from the block codes; and
code for reconstructing the image from the extracted sub-passes at the desired reduced resolution J, wherein the reconstructed image is not significantly degraded visually compared with an image reconstructed from the coded representation at the desired resolution J using all sub-passes.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
That which is claimed is:
1. A dielectric insulation structure (30) formed in a silicon layer (31) by integrating a dielectric trench structure (32) effective to define an insulation well (33) to integrate semiconductor devices on a hollow region (34) completely surrounded by a dielectric area (35) forming also the side insulation of said dielectric trench structure (32) characterised in that said dielectric trench structure (32) is interrupted in a plurality of points to define a plurality of side support regions (36) for said insulation well (33).
2. A dielectric insulation structure (30) according to claim 1, characterised in that said plurality of side support regions (36) is formed through a phototechnique and photoetching step of said dielectric trench structure (32).
3. A dielectric insulation structure (30) according to claim 1, characterised in that said plurality of side support regions (36) is sized so as to ensure a complete dielectric insulation of said insulation well (33).
4. A dielectric insulation structure (30) according to claim 1, characterised in that said plurality of side support regions (36) is sized so as to ensure the oxidation.
5. A dielectric insulation structure (30) according to claim 3, characterised in that said plurality of side support regions (36) is sized through a masking step during an anisotropic etching step to form said dielectric trench structure (32).
6. A dielectric insulation structure (30) according to claim 1, characterised in that said plurality of side support regions (36) comprises a number of side support regions (36) being function of the overall dimensions of said insulation well (33).
7. A manufacturing process of a dielectric insulation structure (30) formed in a silicon layer (31) essentially comprising:
a manufacturing step of a dielectric trench structure (32);
a covering step of said dielectric trench structure (32) walls by means of a dielectric protection material;
a manufacturing step of a hollow region (34) underlying said dielectric trench structure (32);
an oxidising step to form a dielectric area (35) and to dielectrically insulate said hollow region (34), forming thus an insulation well (33) for integrating electronic devices;
a deposition step of a not very conductive material to fill said dielectric trench structure (32); and
a planarising step of said dielectric insulation structure (30) surface,
characterised in that said manufacturing step of said dielectric trench structure (32) also comprises a phototechnique and photoetching step of said dielectric trench structure (32) to define a plurality of side support regions (36) of said insulation well (33).
8. A process according to claim 7, characterised in that said plurality of side support regions (36) is sized so as to ensure a complete dielectric insulation of said insulation well (33).
9. A process according to claim 7, characterised in that said plurality of side support regions (36) is sized through masking during the anisotropic etching step of said manufacturing step of said dielectric trench structure (32).
10. A process according to claim 7, characterised in that said one manufacturing step of said hollow region (34) comprises an etching step effective to form a sole cavity intended as the union between adjacent cavities.
11. A process according to claim 7, characterised in that said insulation well (33) is defined with a same masking level used for said manufacturing step of said dielectric trench structure (32).
12. A process according to claim 7, characterised in that said oxidation step to form said dielectric area (35) affects said plurality of side support regions (36) completely oxidising them.