All The Claims

1. A process of manufacturing cellulose acetate, the process comprising:
steaming a material selected from one or more of bagasse, kenaf, reeds and rice straw in a pressure vessel in a sub-critical state at a temperature of about 150 to 350\xb0 C. and a pressure of about 15 to 29 MPa for about 10 to 30 minutes to obtain a mixture of a solid product and a solution product, wherein the solid product comprises cellulose, wherein the solution product comprises soluble hemicelluloses and polyphenol converted from lignin separated from lignocelluloses, and wherein the sub-critical state is such that the solid product and the solution product are obtained from the material solely due to the temperature and pressure conditions of the state;
filtering the mixture to separate the solid product; and
dehydrating and acetylating the solid product in the presence of acetic anhydride and sulfuric acid to obtain cellulose acetate from the solid product.
2. The process of claim 1, wherein the steaming is performed at a temperature of about 150 to 250\xb0 C. and a pressure of about 15 to 25 MPa, respectively.
3. The process of claim 1, wherein the steaming is performed at a temperature of about 180 to 200\xb0 C. and a pressure of about 25 to 28 MPa, respectively.
4. The process of claim 1, wherein the steaming is performed for about 15 to 20 minutes.

The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.

1. A method for repositioning images in a video data stream, said method comprising:
storing encoded video data in a first buffer, said data including the representation of a first image at a first position in a displayed image;
determining whether repositioning of the first image to a second position in the displayed image would result in a change of bit positions of the encoded first image data, said bit positions being determined with respect to a first number of bits;
reading said video data from said first buffer;
modifying said video data to reposition said first image;
modifying said video data by generating one or more stuffing bits configured to restore said encoded first image data to said bit positions, in response to determining said repositioning would result in said change of bits positions;
coding said stuffing bits such that upon decode said stuffing bits will not materially affect said displayed image; and
storing said modified video data in a second buffer.
2. The method as recited in claim 1, wherein the video data stream is an MPEG-2 data stream, the video data stored in the first buffer comprises an intra-coded P frame, and wherein the modified video data in the second buffer is a modified version of the P frame.
3. The method as recited in claim 2, wherein the first image represented by the P frame is overlaid upon a background image in the displayed image, wherein the background image is conveyed in a separate reference frame.
4. The method as recited in claim 3, wherein said stuffing bits are coded as non-intra data, and wherein the method further comprises generating a non-intra quantization matrix in response to determining said matrix is not represented by said P frame.
5. The method as recited in claim 4, further comprising generating empty slices for slices of the modified P frame which do not include the repositioned first image data, and wherein for each slice of the modified P frame that includes the first image data, the method further comprises:
generating a new slice header with a modified vertical position code to reposition the slice to the second location, in response to determining the second position represents a change in the vertical position of the first image as compared to the first position;
generating a stuffing macroblock, wherein said stuffing macroblock is generated with a first DCT coefficient with length equal to said number of stuffing bits, and wherein upon reconstruction said coefficient is zero;
generate a new address increment of a first macroblock in order to horizontally reposition the first image to the second location;
inserting said stuffing macroblock before said first macroblock;
appending macroblocks which follow said first macroblock from said first buffer to said second buffer; and
generating a last macroblock.
6. The method as recited in claim 5, further comprising:
calculating a number of said stuffing bits to be equal to 7 minus the modular length of said address increment, said modular length being determined with respect to said number of bits, in response to determining said first macroblock is coded with a quantizer_scale_code;
calculating a number of said stuffing bits to be equal to the modular length of one minus the address increment, said modular length being determined with respect to said number of bits, in response to determining said first macroblock is not coded with a quantizer_scale_code;
adding a quantizer_scale_code to the first macroblock if the macroblock does not already include a quantizer_scale_code, wherein a value of said quantizer_scale_code is selected to be the same as that of the slice header which corresponds to the macroblock; and
setting a macroblock_type of said first macroblock equal to binary value of 000001.
7. The method as recited in claim 5, further comprising:
generating said matrix with all coefficients with a value less than 32, in response to determining a picture header of said P frame has a q_scale_type=1; and
generating all said coefficients with a value less than 16, in response to determining said picture header has a q_scale_type=0.
8. The method as recited in claim 5, wherein said stuffing macroblock is generated with macroblock_type equal to binary value 00001 and quantizer_scale_code equal to binary value 00001.
9. A device configured to reposition images in a video data stream, said device comprising:
a storage device configured to store encoded video data, said data including the representation of a first image at a first position in a displayed image; and
a repositioning mechanism configured to:
determine whether repositioning of the first image to a second position in the displayed image would result in a change of bit positions of the encoded first image data, said bit positions being determined with respect to a first number of bits;
read said video data from said first buffer;
modify said video data to reposition said first image to said second position;
modify said video data by generating one or more stuffing bits configured to restore said encoded first image data to said bit positions, in response to determining said repositioning would result in said change of bits positions;
code said stuffing bits such that upon decode said stuffing bits will not materially affect said displayed image; and
store said modified video data in a second buffer.
10. The device as recited in claim 9, wherein the video data stream is an MPEG-2 data stream and wherein the video data stored in the first buffer comprises an intra-coded P frame, and wherein the modified video data in the second buffer is a modified version of the P frame.
11. The device as recited in claim 10, wherein the first image represented by the P frame is overlaid upon a background image in the displayed image, wherein the background image is conveyed in a separate reference frame.
12. The device as recited in claim 11, wherein said repositioning mechanism is further configured to:
code said stuffing bits as non-intra data; and
generate a non-intra quantization matrix in response to determining said matrix is not represented by said P frame.
13. The device as recited in claim 12, wherein said repositioning mechanism is further configured to generate empty slices for slices of the modified P frame which do not include the repositioned first image data, and wherein for each slice of the modified P frame that includes the first image data, the mechanism is further configured to:
generate a new slice header with a modified vertical position code to reposition the slice to the second location, in response to determining the second position represents a change in the vertical position of the first image as compared to the first position;
generate a stuffing macroblock, wherein said stuffing macroblock is generated with a first DCT coefficient with length equal to said number of stuffing bits, and wherein upon reconstruction said coefficient is zero;
generate a new address increment of a first macroblock in order to horizontally reposition the first image to the second location;
insert said stuffing macroblock before said first macroblock;
append macroblocks which follow said first macroblock from said first buffer to said second buffer; and
generate a last macroblock.
14. The device as recited in claim 13, wherein said repositioning mechanism is further configured to:
calculate a number of said stuffing bits to be equal to 7 minus the modular length of said address increment, said modular length being determined with respect to said number of bits, in response to determining said first macroblock is coded with a quantizer_scale_code;
calculate a number of said stuffing bits to be equal to the modular length of one minus the address increment, said modular length being determined with respect to said number of bits, in response to determining said first macroblock is not coded with a quantizer_scale_code;
add a quantizer_scale_code to the first macroblock if the macroblock does not already include a quantizer_scale_code, wherein a value of said quantizer_scale_code is selected to be the same as that of the slice header which corresponds to the macroblock; and
set a macroblock_type of said first macroblock equal to binary value of 000001.
15. The device as recited in claim 13, wherein said repositioning mechanism is further configured to:
generate said matrix with all coefficients having a value less than 32, in response to determining a picture header of said P frame has a q_scale_type=1; and
generate all said coefficients with a value less than 16, in response to determining said picture header has a q_scale_type=0.
16. The device as recited in claim 13, wherein said mechanism is configured to generate said stuffing macroblock with macroblock_type equal to binary value 00001 and quantizer_scale_code equal to binary value 00001.
17. The device as recited in claim 13, wherein said video data stream is obtained by said device from one of the group consisting of: a television broadcast signal; the Internet; and a local storage medium.
18. A carrier medium comprising program instructions, wherein said program instructions are executable to:
store encoded video data in a first buffer, said data including the representation of a first image at a first position in a displayed image;
determine whether repositioning of the first image to a second position in the displayed image would result in a change of bit positions of the encoded first image data, said bit positions being determined with respect to a first number of bits;
read said video data from said first buffer;
modify said video data to reposition said first image to said second position;
modify said video data by generating one or more stuffing bits configured to restore said encoded first image data to said bit positions, in response to determining said repositioning would result in said change of bits positions;
code said stuffing bits such that upon decode said stuffing bits will not materially affect said displayed image; and
store said modified video data in a second buffer.
19. The carrier medium as recited in claim 18, wherein the video data stream comprises an MPEG-2 data stream, and wherein the displayed image comprises the first image overlaid upon a background, the first image being represented by an intra-coded P frame and the background image being represented by an I frame.
20. The carrier medium as recited in claim 19, wherein said program instructions are further executable to:
generate a non-intra quantization matrix in response to determining said matrix is not represented by said P frame;
generate empty slices for slices of the modified P frame which do not include the repositioned first image data; and
for each slice of the modified P frame that includes the first image data:
generate a new slice header with a modified vertical position code to reposition the slice to the second location, in response to determining the second position represents a change in the vertical position of the first image as compared to the first position;
generate a stuffing macroblock coded as non-intra data, wherein said stuffing macroblock is generated with a first DCT coefficient with length equal to said number of stuffing bits, and wherein upon reconstruction said coefficient is zero;
modify an address increment of a first macroblock in order to horizontally reposition the first image to the second location;
insert said stuffing macroblock before said first macroblock;
append macroblocks which follow said first macroblock from said first buffer to said second buffer; and
generate a last macroblock.

1. A photocurable composition for a liquid crystal sealant, comprising a phosphoric acid (meth)acrylate and a compound represented by formula (1):
A1-Y1\u2014B1\u2014Y2-A2\u2003\u2003(1)

wherein A1 and A2 each represents, independently, a monovalent group having a cyclic \u03c0-electron conjugated structure;
Y1 and Y2 each represents, independently, a divalent group represented by formula (4):
wherein y1 and y2 represent a linking group selected from a single bond, an ester bond, a urethane bond, and \u2014NR\u2014, wherein R represents an alkyl group of 1 to 4 carbon atoms, or a benzyl group, or a group in which one of the linking group and one of an alkylene group of 1 to 2 carbon atoms are linked, wherein the alkylene group is bonded to A1 or A2, in which y1 is bonded to A1 or A2 and y2 is bonded to B1, y3 represents an alkyltriyl group of 1 to 3 carbon atoms, y4 represents a single bond, an ether bond, an ester bond, or a urethane bond, y5 represents a single bond, an alkylene group of 1 to 5 carbon atoms, or an oxyalkylene group of 1 to 6 carbon atoms, wherein 2 to 4 oxyalkylene groups may be contiguous, or a phenylene group, wherein y4 is invariably a single bond when y5 is a single bond, and y6 represents a polymerizable group selected from the group consisting of a (meth)acryloyl group, a (meth)acryloyloxy group, a (meth)acrylamide group, a vinyl group, and a vinyloxy group; and
B1 represents (i) a divalent alicyclic hydrocarbon group, (ii) a divalent aromatic hydrocarbon group, (iii) a divalent heterocyclic group, (iv) a divalent aliphatic hydrocarbon group of 1 to 6 carbon atoms, or (v) a divalent group in which two or more groups selected from the group consisting of a divalent alicyclic hydrocarbon group, a divalent aromatic hydrocarbon group, a divalent heterocyclic group, a divalent aliphatic hydrocarbon group of 1 to 6 carbon atoms, a carbonyl group, an ether group, a thioether group, and a SO2 group are linked.
2. The photocurable composition for the liquid crystal sealant according to claim 1, wherein
in the formula (1), A1 and A2 each represents, independently, a monovalent group of a condensed ring structure composed of 2 to 4 rings, wherein the ring constituting the condensed ring is an aromatic ring, a heterocyclic ring, or an alicyclic ring of 3 to 10 carbon atoms), or a group represented by formula (3):
wherein a2 represents a single bond or an alkylene group of 1 to 10 carbon atoms, a3 represents a hydrogen atom, an aromatic hydrocarbon group of 6 to 12 carbon atoms, an aromatic heterocyclic group of 6 to 12 carbon atoms, or an alicyclic hydrocarbon group of 3 to 10 carbon atoms, and m represents an integer from 1 to 2; and
B1 represents a divalent alicyclic hydrocarbon group of 3 to 10 carbon atoms, a divalent aromatic hydrocarbon group of 6 to 20 carbon atoms, an alkylene group of 1 to 6 carbon atoms, or a structure represented by -b1-b2-b3- wherein b1 and b3 represent a divalent alicyclic hydrocarbon group of 3 to 10 carbon atoms or a divalent aromatic hydrocarbon group of 6 to 12 carbon atoms, and b2 represents a single bond, an alkylene group of 1 to 6 carbon atoms, a carbonyl group, an ether group, a thioether group, or a SO2 group.
3. The photocurable composition for the liquid crystal sealant according to claim 1, wherein
in the formula (1), A1 and A2 each represents, independently, a group selected from structural formulas A-1 to A-23 shown below:
Y1 and Y2 each represents, independently, a divalent group represented by the formula (4), and a structure represented by structural formula (y123) shown below in the formula (4) is a group selected from structural formulas y-1 to y-31 shown below:
wherein R represents a hydrocarbon group of 1 to 3 carbon atoms or a phenyl group, A1,2 represents A1 or A2 in the formula (1), B1 represents B1 in the formula (1), y4 represents y4 in the formula (4), and each group is described as a group to which a structure represented by the structural formula (y123) binds;
a structure of -y4 -y5 -y6 which binds to y3 is a group selected from structural formulas y-51 to y-56 shown below:
wherein R\u2032 represents a hydrogen atom or a methyl group, y3 represents y3 in the formula (4) and is described as a group to which -y4 -y5 -y6 binds, and n represents an integer of 1 or 2; and
B1 represents a group selected from structural formulas B-1 to B-16 shown below:
wherein Y1 and Y2 represent Y1 and Y2 in the formula (1), and are described as groups to which B binds.
4. The photocurable composition for the liquid cyrstal sealant according to claim 1, further comprising a photopolymerization initiator having an absorbance at 365 nm of 50 M\u22121 cm\u22121 or more.
5. The photocurable composition for the liquid crystal sealant according to claim 1, wherein the sum of atomic weights of atoms constituting the structures of A1, A2 and B1 is 40% by mass or more based on the molecular weight of the compound represented by formula (1).
6. A liquid crystal panel wherein the components of the panel are sealed by applying and photocuring a photocurable composition for a liquid crystal sealant, and
the photocurable composition for the liquid crystal sealant comprises a compound represented by formula (1):
A1-Y1\u2014B1\u2014Y2-A2\u2003\u2003(1)
wherein A1 and A2 each represents, independently, a monovalent group having a cyclic \u03c0-electron conjugated structure;
Y1 and Y2 each represents, independently, a divalent group represented by formula (4):
wherein y1 and y2 represent a linking group selected from a single bond, an ether bond, an ester bond, a urethane bond, and \u2014NR\u2014, wherein R represents an alkyl group of 1 to 4 carbon atoms, or a benzyl group, or a group in which one of the linking group and one of an alkylene group of 1 to 2 carbon atoms are linked, wherein the alkylene group is bonded to A1 or A2, in which y1 is bonded to A1 or A2 and y2 is bonded to B1, y3 represents an alkyltriyl group of 1 to 3 carbon atoms, y4 represents a single bond, an ether bond, an ester bond, or a urethane bond, y5 represents a single bond, an alkylene group of 1 to 5 carbon atoms, or an oxyalkylene group of 1 to 6 carbon atoms, wherein 2 to 4 oxyalkylene groups may be contiguous, or a phenylene group, wherein y4 is invariably a single bond when y5 is a single bond, and y6 represents a polymerizable group selected from the group consisting of a (meth)acryloyl group, a (meth)acryloyloxy group, a (meth)acrylamide group, a vinyl group, and a vinyloxy group; and
B1 represents (i) a divalent alicyclic hydrocarbon group, (ii) a divalent aromatic hydrocarbon group, (iii) a divalent heterocyclic group, (iv) a divalent aliphatic hydrocarbon group of 1 to 6 carbon atoms, or (v) a divalent group in which two or more groups selected from the group consisting of a divalent alicyclic hydrocarbon group, a divalent aromatic hydrocarbon group, a divalent heterocyclic group, a divalent aliphatic hydrocarbon group of 1 to 6 carbon atoms, a carbonyl group, an ether group, a thioether group, and a SO2 group are linked.

The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.

1. A method comprising:
receiving an existing language model for a first domain;
receiving labeled data associated with a second domain; and
modifying the existing language model by incorporating the labeled data to yield a modified language model associated with the second domain.
2. The method of claim 1, wherein the act of modifying the existing language model further comprises:
determining a distance from the existing language model to the labeled data; and
modifying the existing language model using the distance.
3. The method of claim 2, wherein the existing language model is a speech processing model and wherein the distance comprises a logistic loss function.
4. The method of claim 1, further comprising:
obtaining confidence scores from the existing language model and the modified language model; and
engaging in active learning using the confidence scores.
5. The method of claim 1, wherein modifying the existing language model utilizes at least one of a Boosting algorithm, a Na\xefve Bayes classifier, a linear model interpolation, and a Bayesian adaptation.
6. The method of claim 5, wherein probability distributions correspond to an existing language model probability distribution and a modified language model probability distribution.
7. The method of claim 1, further comprising labeling future utterances using the modified language model.
8. A system comprising:
a processor;
a storage device storing instructions for controlling the processor to perform steps comprising:
receiving an existing language model for a first domain;
receiving labeled data associated with a second domain; and
modifying the existing language model by incorporating the labeled data to yield a modified language model associated with the second domain.
9. The system of claim 8, wherein the act of modifying the existing language model further comprises:
determining a distance from the existing language model to the labeled data; and
modifying the existing language model using the distance.
10. The system of claim 9, wherein the existing language model is a speech processing model and wherein the distance comprises a logistic loss function.
11. The system of claim 8, the steps further comprising:
obtaining confidence scores from the existing language model and the modified language model; and
engaging in active learning using the confidence scores.
12. The system of claim 8, wherein modifying the existing language model utilizes at least one of a Boosting algorithm, a Na\xefve Bayes classifier, a linear model interpolation, and a Bayesian adaptation.
13. The system of claim 12, wherein probability distributions correspond to an existing language model probability distribution and a modified language model probability distribution.
14. The system of claim 8, further comprising labeling future utterances using the modified language model.
15. A non-transitory computer-readable storage medium storing instructions which, when executed by a computing device, cause the computing device to perform steps comprising:
receiving an existing language model for a first domain;
receiving labeled data associated with a second domain; and
modifying the existing language model by incorporating the labeled data to yield a modified language model associated with the second domain.
16. The non-transitory computer-readable storage medium of claim 15, wherein the act of modifying the existing language model further comprises:
determining a distance from the existing language model to the labeled data; and
modifying the existing language model using the distance.
17. The non-transitory computer-readable storage medium of claim 16, wherein the existing language model is a speech processing model and wherein the distance comprises a logistic loss function.
18. The non-transitory computer-readable storage medium of claim 15, the steps further comprising:
obtaining confidence scores from the existing language model and the modified language model; and
engaging in active learning using the confidence scores.
19. The non-transitory computer-readable storage medium of claim 15, wherein modifying the existing language model utilizes at least one of a Boosting algorithm, a Na\xefve Bayes classifier, a linear model interpolation, and a Bayesian adaptation.
20. The non-transitory computer-readable storage medium of claim 19, wherein probability distributions correspond to an existing language model probability distribution and a modified language model probability distribution.