Methods and apparatus for hinting a font for controlling stem width as font size and resolution of output device vary

Claim 2

In an automatic typography system, a method of hinting a font having glyphs with a plurality of stems, said hinting being for reproduction of said glyphs on a pixel based device using a scaling factor equal to a number of design units per pixel, said method comprising the steps of: measuring the width of each of said plurality of stems to produce a plurality of stem width measurements; assigning a weight to each of said plurality of stem width measurements, said weight being at least a function of the number of times said stem width measurement was detected in said step of measuring the width of each of said plurality of stems; clustering said stem width measurements to provide a set of clustered measurements by: (a) selecting as a first stem width measurement a particular one of said plurality of stem width measurements having the greatest said weight assigned thereto; (b) adding said first stem width measurement to said set of clustered measurements a next clustered measurement and deleting said first stem width measurement from said plurality of stem width measurements; (c) deleting from said plurality of stem width measurements all of said stem width measurements having values within a range of values determined by said first stem width measurement and a stem cluster tolerance value; (d) repeating sub-steps (a) through (c) above, until all of said plurality of stem width measurements have been deleted; partitioning said set of clustered measurements into one or more blocks of clustered measurements as a function of said scaling factor and a predetermined threshold ε so that each said block corresponds to a unique, with respect to others of said blocks, integer number of said pixels; and hinting each of the blocks of clustered measurements to one of said plurality of integer numbers of pixels.

Claim 3

The method of claim 1 wherein said weight of each of said plurality of measurements m is a function of both said number of occurrences of said stem width measurement m and the number of occurrences of said stem width measurements that are within said predetermined stem cluster tolerance value of said stem width measurement m.

Claim 4

The method of claim 1, wherein said hinting step comprises the steps of: selecting one member of said set of clustered measurements as a representative measurement from each of said plurality of blocks; hinting each said representative measurement from each of said plurality of blocks to a particular one of said integer numbers of pixels; and hinting each of said clustered measurements that is in the same block as each said respective representative measurement to said particular one of said plurality of integer numbers of pixels.

Claim 5

The method of claim 3, wherein said weight of each of said plurality of stem width measurements m is a function of both said number of occurrences of said stem width measurement m and the number of occurrences of said stem width measurements that are within said predetermined stem cluster tolerance value of said stem width measurement m, and said representative measurement from each of the plurality of blocks is the clustered measurement in said each block with the greatest frequency, wherein said frequency of said clustered measurement m is the sum of the number of occurrences of each of said stem width measurements for stem width measurements that were clustered into said representative clustered measurement during said step of clustering said stem width measurements.

Claim 6

The method of claim 3, wherein said step of hinting each said representative measurement to a particular one of said integer number of pixels comprises: scaling each said representative measurement according to said scaling factor to provide a scaled representative measurement; and rounding said scaled representative measurement to the nearest integer number of pixels.

Claim 7

The method of claim 1, wherein said predetermined threshold ε is one half a pixel.

Claim 8

A method of hinting a font as recited in claim 1 wherein: said step of clustering said stem width measurements includes a substep of: providing a frequency value for said next clustered measurement as a function of the sum of the number of occurrences of said first stem width measurement in said plurality of stem width measurements and the number of occurrences of said stem width measurements deleted in substep (c); and said step of partitioning said set of clustered measurements into one or more blocks of clustered measurements includes a step of selecting a representative clustered measurement for each said block as the clustered measurement within said block having the highest said frequency value of all of said clustered measurements within said block.

Claim 9

An autohinter for hinting a font having glyphs with a plurality of stems, said hinting being for reproduction of said glyphs on a pixel based device using a scaling factor equal to a number of design units per pixel, comprising: a measuring element for measuring the width of each of said plurality of stems to produce a plurality of stem width measurements; a counter for assigning a weight to each of said plurality of stem width measurements, said weight being at least a function of the number of times said stem width measurement was detected by said measuring element; a clustering element for clustering said stem width measurements to provide a set of clustered measurements by iteratively (a) selecting as a first stem width measurement a particular one of said plurality of stem width measurements having the greatest said weight assigned thereto; (b) adding said first stein width measurement to said set of clustered measurements as a next clustered measurement and deleting said first stem width measurement from said plurality of stem width measurements; (c) deleting from said plurality of stem width measurements all of said stem width measurements having values within a range of values determined by said first stem width measurement and a stem cluster tolerance value; until all of said plurality of stem width measurements having been deleted; a partitioning clement for partitioning said set of clustered measurements into one or more blocks of clustered measurements as a function of said scaling factor and a predetermined threshold ε so that each said block corresponds to a unique, with respect to others of said blocks, integer number of said pixels; and a hinting element for hinting each of the blocks of clustered measurements to one of said plurality of integer numbers of pixels.

Claim 10

An autohinter for hinting a font having glyphs with a plurality of stems, said hinting being for reproduction of said glyphs on a pixel based device using a scaling factor equal to a number of design units per pixel, comprising: a measuring element for measuring the width of each of said plurality of stems to produce a plurality of stem width measurements; a counter for assigning a weight to each of said plurality of stem width measurements, said weight being at least a function of the number of times said stem width measurement was detected by said measuring element; a clustering element for clustering said stem width measurements to provide a set of clustered measurements by iteratively (a) selecting as a first stem width measurement a particular one of said plurality of stem width measurements having the greatest said weight assigned thereto; (b) adding said first stein width measurement to said set of clustered measurements as a next clustered measurement and deleting said first stem width measurement from said plurality of stem width measurements; (c) deleting from said plurality of stem width measurements all of said stem width measurements having values within a range of values determined by said first stem width measurement and a stem cluster tolerance value; until all of said plurality of stem width measurements having been deleted; a partitioning clement for partitioning said set of clustered measurements into one or more blocks of clustered measurements as a function of said scaling factor and a predetermined threshold ε so that each said block corresponds to a unique, with respect to others of said blocks, integer number of said pixels; and a hinting element for hinting each of the blocks of clustered measurements to one of said plurality of integer numbers of pixels.

Claim 11

The autohinter of claim 8, wherein said hinting element is operative to: select one member of said set of clustered measurements as a representative measurement from each of said plurality of blocks; hint each said representative measurement from each of said plurality of blocks to a particular one of said integer numbers of pixels; and hint each of said clustered measurements that is in the same block as each said respective representative measurement to said particular one of said plurality of integer numbers of pixels.

Claim 12

In an automatic typography system, a method of hinting a subset of more than one character of a font, each character in the font having glyphs with a plurality of stems, said hinting being for reproduction of said glyphs on a pixel based device using a scaling factor equal to a number of design units per pixel, said method comprising the steps of: measuring the width of each of said plurality of steins to produce a plurality of stem width measurements; assigning a weight to each of said plurality of stem width measurements, said weight being at least a function of the number of times said stem width measurement was detected in said step of measuring the width of each of said plurality of stems; clustering said stem width measurements to provide a set of clustered measurements by; (a) selecting as a first stem width measurement a particular one of said plurality of stem width measurements having the greatest said weight assigned thereto; (b) adding said first stem width measurement to said sat of clustered measurements a next clustered measurement and deleting said first stem width measurement from said plurality of stem width measurements; (c) deleting from said plurality of stem width measurements all of said stem width measurements having values within a range of values determined by said first stem width measurement and a stem cluster tolerance value; (d) repeating sub-steps (a) through (c) above, until all of said plurality of stem width measurements have been deleted; partitioning said set of clustered measurements into one or more blocks of clustered measurements as function of said scaling factor and a predetermined threshold e so that each said block corresponds to a unique, with respect to others of said blocks, integer number of said pixels; and hinting each of the blocks of clustered measurements to one of said plurality of integer numbers of pixels.

Claim 13

In an automatic typography system, a method of hinting a subset of more than one character of a font, each character in the font having glyphs with a plurality of stems, said hinting being for reproduction of said glyphs on a pixel based device using a scaling factor equal to a number of design units per pixel, said method comprising the steps of: measuring the width of each of said plurality of steins to produce a plurality of stem width measurements; assigning a weight to each of said plurality of stem width measurements, said weight being at least a function of the number of times said stem width measurement was detected in said step of measuring the width of each of said plurality of stems; clustering said stem width measurements to provide a set of clustered measurements by; (a) selecting as a first stem width measurement a particular one of said plurality of stem width measurements having the greatest said weight assigned thereto; (b) adding said first stem width measurement to said sat of clustered measurements a next clustered measurement and deleting said first stem width measurement from said plurality of stem width measurements; (c) deleting from said plurality of stem width measurements all of said stem width measurements having values within a range of values determined by said first stem width measurement and a stem cluster tolerance value; (d) repeating sub-steps (a) through (c) above, until all of said plurality of stem width measurements have been deleted; partitioning said set of clustered measurements into one or more blocks of clustered measurements as function of said scaling factor and a predetermined threshold e so that each said block corresponds to a unique, with respect to others of said blocks, integer number of said pixels; and hinting each of the blocks of clustered measurements to one of said plurality of integer numbers of pixels.

Claim 14

The method of claim 10 wherein said weight of each of said plurality of measurements m is a function of both said number of occurrences of said stem width measurement m and the number of occurrences of said stem width measurements that are within said predetermined stem cluster tolerance value of said stem width measurement m.

Claim 15

The method of claim 10, wherein said hinting step comprises the steps of: selecting one member of said set of clustered measurements as a representative measurement from each of said plurality of blocks; hinting each said representative measurement from each of said plurality of blocks to a particular one of said integer numbers of pixels; and hinting each of said clustered measurements that is in the same block as each said respective representative measurement to said particular one of said plurality of integer numbers of pixels.

Claim 16

The method of claim 12, wherein said weight of each of said plurality of stem width measurements m is a function of both said number of occurrences of said stem width measurement m and the number of occurrences of said stem width measurements that are within said predetermined stem cluster tolerance value of said stem width measurement m, and said representative measurement from each of the plurality of blocks is the clustered measurement in said each block with the greatest frequency, wherein said frequency of said clustered measurement m is the sum of the number of occurrences of each of said stem width measurements for stem width measurements that were clustered into said representative clustered measurement during said step of clustering said stem width measurements.

Claim 17

The method of claim 12, wherein said step of hinting each said representative measurement to a particular one of said integer number of pixels comprises: scaling each said representative measurement according to said scaling factor to provide a scaled representative measurement; and rounding said scaled representative measurement to the nearest integer number of pixels.

Claim 18

The method of claim 10 wherein: said step of clustering said stem width measurements includes a substep of: providing a frequency value for said next clustered measurement as a function of the sum of the number of occurrences of said first stem width measurement in said plurality of stem width measurements and the number of occurrences of said stem width measurements deleted in substep (c); and said step of partitioning said set of clustered measurements into one or more blocks of clustered measurements includes a step of selecting a representative clustered measurement for each said block as the clustered measurement within said block having the highest said frequency value of all of said clustered measurements within said block.

Claim 19

An autohinter for hinting a subset of more than one character of a font, each character in the font having glyphs with a plurality of stems said hinting being for reproduction of said glyphs on a pixel based device using a scaling factor equal to a number of design units per pixel, comprising: a measuring element for measuring the width of each of said plurality of stems to produce a plurality of stem width measurements; a counter for assigning a weight to each of said plurality of stem width measurements, said weight being at least a function of the number of times said stem width measurement was detected by said measuring element; a clustering element for clustering said stem width measurements to provide a set of clustered measurements by iteratively: (a) selecting as a first stem width measurement a particular one of said plurality of stem width measurements having the greatest said weight assigned thereto; (b) adding said first stem width measurement to said set of clustered measurements as a next clustered measurement and deleting said first stem width measurement from said plurality of stem width measurements; (c) deleting from said plurality of stem width measurements all of said stem width measurements having values within a range of values determined by said first stem width measurement and a stem cluster tolerance value; until all of said plurality of stem width measurements having been deleted; a partitioning element for partitioning said set of clustered measurements into one or more blocks of clustered measurements as a function of said scaling factor and a predetermined threshold ε so that each said block corresponds to a unique, with respect to others of said blocks, integer number of said pixels; and a hinting element for hinting each of the blocks of clustered measurements to one of said plurality of integer numbers of pixels.

Claim 20

An autohinter for hinting a subset of more than one character of a font, each character in the font having glyphs with a plurality of stems said hinting being for reproduction of said glyphs on a pixel based device using a scaling factor equal to a number of design units per pixel, comprising: a measuring element for measuring the width of each of said plurality of stems to produce a plurality of stem width measurements; a counter for assigning a weight to each of said plurality of stem width measurements, said weight being at least a function of the number of times said stem width measurement was detected by said measuring element; a clustering element for clustering said stem width measurements to provide a set of clustered measurements by iteratively: (a) selecting as a first stem width measurement a particular one of said plurality of stem width measurements having the greatest said weight assigned thereto; (b) adding said first stem width measurement to said set of clustered measurements as a next clustered measurement and deleting said first stem width measurement from said plurality of stem width measurements; (c) deleting from said plurality of stem width measurements all of said stem width measurements having values within a range of values determined by said first stem width measurement and a stem cluster tolerance value; until all of said plurality of stem width measurements having been deleted; a partitioning element for partitioning said set of clustered measurements into one or more blocks of clustered measurements as a function of said scaling factor and a predetermined threshold ε so that each said block corresponds to a unique, with respect to others of said blocks, integer number of said pixels; and a hinting element for hinting each of the blocks of clustered measurements to one of said plurality of integer numbers of pixels.

Claim 21

The autohinter of claim 16, wherein said hinting element is operative to: select one member of said set of clustered measurements as a representative measurement from each of said plurality of blocks; hint each said representative measurement from each of said plurality of blocks to a particular one of said integer numbers of pixels; and hint each of said clustered measurements that is in the same block as each said respective representative measurement to said particular one of said plurality of integer human of pixels.