US5165101A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Methods and apparatus for optically determining the acceptability of products
Assignee: Philip Morris USA Inc
•
Filing: 1991-05-06
•
Granted: 1992-11-17
•
Expires: 2011-05-06
Abstract
Abstract The appearance of a product is determined to be acceptable or unacceptable by computing a discriminant function or image from a plurality of two-dimensional images of products of the kind to be inspected. A two-dimensional image of the product is then formed and processed using the discriminant image to produce an output value which can be compared to a predetermined value used in computing the discriminant image. If the output value deviates from the predetermined value by more than a predetermined amount, the product has been found to have an unacceptable appearance.
Inventors
- 👤 Kenneth A. Cox
- 👤 Henry M. Dante
- 👤 Charles N. Harward
- 👤 Robert J. Maher
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-01 19:49
CPC Classifications
B; B07; B07C; B07C5/00; B07C5/04; B07C5/10; G; G01; G01N; G01N21/00; G01N21/84; G01N21/88; G06; G06F; G06F18/00; G06F18/20; G06F18/21; G06T; G06T7/00; G06T7/0002; G06T7/0004; G06T7/001; G06V; G06V10/00; G06V10/70; G07; G07C; G07C3/00; G07C3/14; G01N21/8851; G06T2207/00; G06T2207/30; G06T2207/30108; G06T2207/30164
Claims Analysis
Total Claims: 54
Claim 1
Independent Claim
269 words
The method of determining whether or not a surface of an object has substantially the same appearance as the corresponding surface of a plurality of training objects, said method comprising the steps of: for each training object, (a) forming a two-dimensional image of said surface, (b) using computer processor means to subdivide said two-dimensional image into a plurality of pixels, (c) using said computer processor means to digitize each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixel, and (d) using said computer processor means to binarize each of said output digital values by associating a first binary value with each of said output digital values which is on one side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value; using said computer processor means to associate a discriminant value with each of said pixels, each discriminant value being chosen so that, for each training object, the dot product of the discimrinant values and the binary values is approximately a preetermined nonzero constant value; repeating steps (a)-(d) for said object; and using said computer processor means to compare the dot product of said discriminant values and the binary values for said object to said predetermined nonzero constant value; said first and second binary values and said predetermined nonzero constant value being selected so that the data resulting from step (d) for each training object and for said object is at least effectively rendered bipolar.
Claim 2
Independent Claim
269 words
The method of determining whether or not a surface of an object has substantially the same appearance as the corresponding surface of a plurality of training objects, said method comprising the steps of: for each training object, (a) forming a two-dimensional image of said surface, (b) using computer processor means to subdivide said two-dimensional image into a plurality of pixels, (c) using said computer processor means to digitize each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixel, and (d) using said computer processor means to binarize each of said output digital values by associating a first binary value with each of said output digital values which is on one side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value; using said computer processor means to associate a discriminant value with each of said pixels, each discriminant value being chosen so that, for each training object, the dot product of the discimrinant values and the binary values is approximately a preetermined nonzero constant value; repeating steps (a)-(d) for said object; and using said computer processor means to compare the dot product of said discriminant values and the binary values for said object to said predetermined nonzero constant value; said first and second binary values and said predetermined nonzero constant value being selected so that the data resulting from step (d) for each training object and for said object is at least effectively rendered bipolar.
Claim 3
Dependent Claim
17 words
The method defined in claim 1 wherein said first and second binary values are -1 and 1.
Claim 4
Dependent Claim
75 words
The method defined in claim 1 wherein said first and second binary values are 0 and 1, and wherein said discriminant values are selected such that [2I.sup.(i) -1]·F=1 for all acceptable images, and such that [2I.sup.(i) -1]·F<<1 for all unacceptable images, where I.sup.(i) is the data that results from step (d) for the ith training object, 1 is a matrix in which all entries are 1, and F is the matrix of all discriminant values.
Claim 5
Dependent Claim
20 words
The method defined in claim 3 wherein said predetermined nonzero constant value is the value p selected such that p=∥+1·F]/2.
Claim 6
Dependent Claim
239 words
Apparatus for determining whether or not a surface of an object has substantially the same appearance as the corresponding surface of a plurality of training objects comprising: means for, for each training object, (a) forming a two-dimensional image of said surface, (b) subdividing said two-dimensional image into a plurality of pixels, (c) digitizing each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixel, and (d) binarizing each of said output digital values by associating a first binary value with each of said output digital values which is on one side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value; value with each of said pixels, each discriminant value being chosen so that, for each training object, the dot product of the discriminant values and the binary values is approximately a predetermined nonzero constant value; means for repeating functions (a)-(d) with respect to said object; and means for comparing the dot product of said discriminant values and the binary values for said object to said predetermined nonzero constant value; said first and second binary values and said predetermined nonzero constant value being selected so that the data resulting from function (d) for each training object and for said object is at least effectively rendered bipolar.
Claim 7
Dependent Claim
239 words
Apparatus for determining whether or not a surface of an object has substantially the same appearance as the corresponding surface of a plurality of training objects comprising: means for, for each training object, (a) forming a two-dimensional image of said surface, (b) subdividing said two-dimensional image into a plurality of pixels, (c) digitizing each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixel, and (d) binarizing each of said output digital values by associating a first binary value with each of said output digital values which is on one side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value; value with each of said pixels, each discriminant value being chosen so that, for each training object, the dot product of the discriminant values and the binary values is approximately a predetermined nonzero constant value; means for repeating functions (a)-(d) with respect to said object; and means for comparing the dot product of said discriminant values and the binary values for said object to said predetermined nonzero constant value; said first and second binary values and said predetermined nonzero constant value being selected so that the data resulting from function (d) for each training object and for said object is at least effectively rendered bipolar.
Claim 8
Dependent Claim
17 words
The apparatus defined in claim 5 wherein said first and second binary values are -1 and 1.
Claim 9
Dependent Claim
75 words
The apparatus defined in claim 5 wherein said first and second binary values are 0 and 1, and wherein said discriminant values are selected such that [2I.sup.(i) -1]·F=1 for all acceptable images, and such that [2I.sup.(i) -1]·F<<1 for all unacceptable images, where I.sup.(i) is the data that results from function (d) for the ith training object, 1 is a matrix in which all entires are 1, and F is the matrix of all discriminant values.
Claim 10
Dependent Claim
20 words
The apparatus defined in claim 7 wherein said predetermined nonzero constant value is the value p selected such that p=[1+1·F]/2.
Claim 11
Independent Claim
313 words
The method of determining whether of not a surface of an object has substantially the same appearance as the corresponding surface of a plurality of training objects, said method comprsiing the steps of: for each training object, (a) forming a two-dimensional image of said surface, (b) using computer processor means to subdivide said two-dimensional image into a plurality of pixels, (c) using said computer processor means to digitize each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixel, and (d) using said computer processor means to binarize each of said output digital values by associating a first binary value with each of said output digital values which is on one side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value; using said computer processor means to associate a discriminant value with each of said pixels, each discriminant value being chosen so that, for each training object, the dot product of the discimrinant values and the binary values is approximately a preetermined nonzero constant value; repeating steps (a)-(d) for said object; and using said computer processor means to compare the dot product of said discriminant values and the binary values for said object to said predetermined nonzero constant value; wherein said image of said object may be shifted in at least one direction from a reference position, and wherein said method further comprises the steps of: using said computer processor means to associate a filter value with at least some of said pixels, each filter value being chosen so that the dot product of the filter values and the binary values for said object is proportional to the amount by which said object is shifted from said reference position.
Claim 12
Independent Claim
313 words
The method of determining whether of not a surface of an object has substantially the same appearance as the corresponding surface of a plurality of training objects, said method comprsiing the steps of: for each training object, (a) forming a two-dimensional image of said surface, (b) using computer processor means to subdivide said two-dimensional image into a plurality of pixels, (c) using said computer processor means to digitize each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixel, and (d) using said computer processor means to binarize each of said output digital values by associating a first binary value with each of said output digital values which is on one side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value; using said computer processor means to associate a discriminant value with each of said pixels, each discriminant value being chosen so that, for each training object, the dot product of the discimrinant values and the binary values is approximately a preetermined nonzero constant value; repeating steps (a)-(d) for said object; and using said computer processor means to compare the dot product of said discriminant values and the binary values for said object to said predetermined nonzero constant value; wherein said image of said object may be shifted in at least one direction from a reference position, and wherein said method further comprises the steps of: using said computer processor means to associate a filter value with at least some of said pixels, each filter value being chosen so that the dot product of the filter values and the binary values for said object is proportional to the amount by which said object is shifted from said reference position.
Claim 13
Dependent Claim
44 words
The method defined in claim 9 further comprising the step of: using said computer processor means to utilize the dot product of the filter values and the binary values for said object to register the binary values for said product within said reference position.
Claim 14
Dependent Claim
145 words
The method defined in claim 9 wherein said filter values collectively comprise a filter iamge F, and wherein said method further comprises the steps of: using said computer processor means to select a training object which is in registration with said reference position; using said computer processor means to select at least a portion of the binary values associated with said selected training object to produce a plurality of data sets in each of which said selected binary values are shifted by a respective one of a plurality of predetermined amounts parallel to said direction; using said computer processor means to convert each of said data sets to a respective one of a plurality of eigenimages Φ.sup.(j), all of said eigenimages being orthogonal to one another; and using said computer processor means to form said filter image from a truncated linear combination of said eigenimages.
Claim 15
Dependent Claim
56 words
The method defined in claim 11 wherein the ith data set is given by J.sup.(i), and wherein said method further comprises the step of: using said computer processor means to compute each of said eigenimages as in the equation: ##EQU26## where m is the total number of data sets and the transformation matrix B is orthogonal.
Claim 16
Dependent Claim
52 words
The method defined in claim 12 wherein said step of forming said filter image from a truncated linear combination of said eigenimages comprises the step of: using said computer processor means to compute said filter image as in the equation: ##EQU27## where s is an integer less than or equal to m.
Claim 17
Dependent Claim
57 words
The method defined in claim 13 further comprising the step of: using said computer processor means to select the value of s for the equation in claim 13 which gives the minimum value for V(s) in the equation: ##EQU28## where δ i is the amount by which the ith data set is shifted from said reference position.
Claim 18
Dependent Claim
82 words
The method defined in claim 14 further comprising the step of: using said computer processor means to compute the value C j for the equation of claim 13 as in the equation: ##EQU29## where the eigenvalues γ j are the diagonal elements of a matrix G given by the equation A·B=B·G, where A is the real, symmetric matrix given by the dot product of every possible data set pair, and the transformation matrix B is such that B·B + =B + ·B=1.
Claim 19
Dependent Claim
25 words
The method defined in claim 15 wherein said eigenimages are ordered such that γ 1 >γ 2 >γ 3 >. . . ≦γ n .
Claim 20
Dependent Claim
27 words
The method defined in claim 9 wherein said filter values are additionally chosen so that the sum of the squares of the filter values is approximately minimized.
Claim 21
Dependent Claim
281 words
Apparatus for determining whether or not a surface of an object has substantially the same appearance as the corresponding surface of a plurality of training objects comprising: means for, for each training object, (a) forming a two-dimensional image of said surface, (b) subdividing said two-dimensional image into a plurality of pixels, (c) digitizing each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixel, and (d) binarizing each of said output digital values by associating a first binary value with each of said output digital values which is on one side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value; means for associating a discriminant value with each of said pixels, each discriminant value being chosen so that, for each training object, the dot product of the discriminant values and the binary values is approximately a predetermined nonzero constant value; means for repeating functions (a)-(d) with respect to said object; and means for comparing the dot product of said discriminant values and the binary values for said object to said predetermined nonzero constant value; wherein said image of said object may be shifted in at least one direction from a reference position, and wherein said apparatus further comprises: means for associating a filter value with at least some of said pixels, each filter value being chosen so that the dot product of the filter values and the binary values for said object is proportional to the amount by which said object is shifted from said reference position.
Claim 22
Dependent Claim
281 words
Apparatus for determining whether or not a surface of an object has substantially the same appearance as the corresponding surface of a plurality of training objects comprising: means for, for each training object, (a) forming a two-dimensional image of said surface, (b) subdividing said two-dimensional image into a plurality of pixels, (c) digitizing each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixel, and (d) binarizing each of said output digital values by associating a first binary value with each of said output digital values which is on one side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value; means for associating a discriminant value with each of said pixels, each discriminant value being chosen so that, for each training object, the dot product of the discriminant values and the binary values is approximately a predetermined nonzero constant value; means for repeating functions (a)-(d) with respect to said object; and means for comparing the dot product of said discriminant values and the binary values for said object to said predetermined nonzero constant value; wherein said image of said object may be shifted in at least one direction from a reference position, and wherein said apparatus further comprises: means for associating a filter value with at least some of said pixels, each filter value being chosen so that the dot product of the filter values and the binary values for said object is proportional to the amount by which said object is shifted from said reference position.
Claim 23
Dependent Claim
37 words
The apparatus defined in claim 18 further comprising: means for using the dot product of the filter values and the binary values for said object to register the binary values for said product with said reference position.
Claim 24
Dependent Claim
126 words
The apparatus defined in claim 18 wherein said filter values collectively comprise a filter image F, and wherein said apparatus further comprises: means for selecting a training object which is in registration with said reference position; means for selecting at least a portion of the binary values associated with said selected training object to produce a plurality of data sets in each of which said selected binary values are shifted by a respective one of a plurality of predetermined amounts parallel to said direction; means for converting each of said data sets to a respective one of a plurality of eigenimages Φ.sup.(j), all of said eigenimages being orthogonal to one another; and means for forming said filter image from a truncated linear combination of said eigenimages.
Claim 25
Dependent Claim
49 words
The apparatus defined in claim 20 wherein the ith data set is given by J.sup.(i), and wherein said apparatus further comprises: means for computing each of said eigenimages as in the equation: ##EQU30## where m is the total number of data sets and the transformation matrix B is orthogonal.
Claim 26
Dependent Claim
45 words
The apparatus defined in claim 21 wherein said means for forming said filter image from a truncated linear combination of said eigenimages comprises: means for computing said filter image as in the equation: ##EQU31## where s is an integer less than or equal to m.
Claim 27
Dependent Claim
50 words
The apparatus defined in claim 22 further comprising: means for selecting the value of s for the equation in claim 22 which gives the minimum value for V(s) in the equation: ##EQU32## where δ i is the amount by which the ith data set is shifted from said reference position.
Claim 28
Dependent Claim
75 words
The apparatus defined in claim 23 further comprising: means for computing the value C j for the equation of claim 22 as in the equation: ##EQU33## where the eigenvalues γ j are the diagonal elements of a matrix G given by the equation A·B=B·G, where A is the real, symmetric matrix given by the dot product of every possible data set pair, and the transformation matrix B is such that B·B + =B + ·B=1.
Claim 29
Dependent Claim
25 words
The apparatus defined in claim 24 wherein said eigenimages are ordered such that γ 1 >γ 2 >γ 3 >. . . ≦γ n .
Claim 30
Dependent Claim
27 words
The apparatus defined in claim 18 wherein said filter values are additionally chosen so that the sum of the squares of the filter values is approximately minimized.
Claim 31
Independent Claim
260 words
The method of determining whether or not a surface of an object has substantially the same appearance as the corresponding surface of a plurality of training objects, said method comprising the steps of: for each training object, (a) forming a two-dimensional image of said surface, (b) using computer procsesor means to subdivide said two-dimensional image into a plurality of pixels, (c) using said computer processor means to digitize each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixle, and (d) using said computer processor means to binarize each of said output digital values by associating a first binary value with each of said output digital values which is on one side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value; using said computer processor means to associate a discriminant value with each of said pixels, each dicriminant value being chosen so that, for each training object, the dot product of the disciminant values and the binary values is approximately a predetermined nonzero constant value; repeating steps (a)-(d) for said object; and using said computer processor means to compare the dot product of said discriminant values and the binary values for said object to said predetermined nonzero constant value; wherein all of the foregoing steps are performed separately for each of a plurality of different segments of the surface of said training objects and said object.
Claim 32
Independent Claim
260 words
The method of determining whether or not a surface of an object has substantially the same appearance as the corresponding surface of a plurality of training objects, said method comprising the steps of: for each training object, (a) forming a two-dimensional image of said surface, (b) using computer procsesor means to subdivide said two-dimensional image into a plurality of pixels, (c) using said computer processor means to digitize each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixle, and (d) using said computer processor means to binarize each of said output digital values by associating a first binary value with each of said output digital values which is on one side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value; using said computer processor means to associate a discriminant value with each of said pixels, each dicriminant value being chosen so that, for each training object, the dot product of the disciminant values and the binary values is approximately a predetermined nonzero constant value; repeating steps (a)-(d) for said object; and using said computer processor means to compare the dot product of said discriminant values and the binary values for said object to said predetermined nonzero constant value; wherein all of the foregoing steps are performed separately for each of a plurality of different segments of the surface of said training objects and said object.
Claim 33
Independent Claim
245 words
The method of determining whether or not a surface of each of a plurality of successive objects has substantially the same appearance as the corresponding surface of at least some of the preceding objects in the succession, said method comprising the steps of: forming a two-dimensional image of said surface of each object in said succession; using computer processor means to digitize each of said two-dimensional images; using said computer processor means to utilize the digitized image of the surface of said first object in said succession to form a digitized discriminant image in which the digital values are related to the digital values in the digitized image of the surface of said first object by a predetermined proportionality value; for at least some of said objects after said first object in said succession, using said computer processor means to replace the digital values in said digitized discriminant image with updated digital values respectively equal to the discriminant image digital values prior to replacement plus an amount proportional to a function of the dot product of the digitized discirminant image prior to replacement and the digitized image of the surface of said object on which said replacement is being based; and for at least some of said objects after said first object in said succession, using said computer processor means to compare the dot product of said digitized disciminant image and said digitized image of the surface of said object to a predetermined constant value.
Claim 34
Independent Claim
245 words
The method of determining whether or not a surface of each of a plurality of successive objects has substantially the same appearance as the corresponding surface of at least some of the preceding objects in the succession, said method comprising the steps of: forming a two-dimensional image of said surface of each object in said succession; using computer processor means to digitize each of said two-dimensional images; using said computer processor means to utilize the digitized image of the surface of said first object in said succession to form a digitized discriminant image in which the digital values are related to the digital values in the digitized image of the surface of said first object by a predetermined proportionality value; for at least some of said objects after said first object in said succession, using said computer processor means to replace the digital values in said digitized discriminant image with updated digital values respectively equal to the discriminant image digital values prior to replacement plus an amount proportional to a function of the dot product of the digitized discirminant image prior to replacement and the digitized image of the surface of said object on which said replacement is being based; and for at least some of said objects after said first object in said succession, using said computer processor means to compare the dot product of said digitized disciminant image and said digitized image of the surface of said object to a predetermined constant value.
Claim 35
Dependent Claim
41 words
The method defined in claim 28 wherein said proportionality value is approximately equal to the reciprocal of the square root of the sum of the squares of the digital values in the digitized image of the surface of said first object.
Claim 36
Dependent Claim
111 words
The method defined in claim 29 wherein said amount proportional to a function of teh dot product of the digitized discriminant image prior to replacement and the digitized image of the surface of said object on which said replacement is being based is given by the expression: ##EQU34## where I.sup.(i) is the diqitized image on which said replacement is being based, where |I.sup.(i) |, is equal to the square root of the sum of the squares of the digitized values in the digitized image of the surface of said first object, where ##EQU35## where F.sup.(i-1) is the digitized discriminant image prior to replacement, and where δ is a predetermined learning parameter.
Claim 37
Dependent Claim
20 words
The method defined in claim 30 wherein δ has a value in the range from about 0 to about 2.
Claim 38
Dependent Claim
11 words
The method defined in claim 31 wherein δ is approximately 1.
Claim 39
Dependent Claim
77 words
The method defined in claim 28 wherein, for any object after said first object for which both said using said computer processor means to replace and using said computer processor means to comapre steps may be performed, said using said computer processor means to replace step is performed only if the results of said using said computer processor means to comapre step indicate that said dot product is within a predetermined range of said predetermined constant value.
Claim 40
Dependent Claim
28 words
The method defined in claim 28 wherein said using said computer processor means to compare step is performed only on objects following an initial subplurality of said objects.
Claim 41
Dependent Claim
30 words
The method defined in claim 28 wherein said using said computer processor means to replace step is performed only with respect to objects in an initial subplurality of said objects.
Claim 42
Dependent Claim
134 words
The method defined in claim 28 wherein said step of using said computer processor means to digitize each of said two-dimensional images comprises the steps of: using said computer processor means to subdivide said image into a plurality of pixels; using said computer processor means to digitize each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixel; and using said computer processor means to binarize each of said output digital values by associating a first binary value with each of said output digital values which is on a first side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value.
Claim 43
Dependent Claim
69 words
The method defined in claim 28 further comprising the step of: using said computer processor means to subdivide each of said two-dimensional images into a plurality of segments and using said computer processor means to perform said steps of using said computer processor means to utilize, using said computer processor means to replace, and usign said computer processor means to compare separately with respect to each of said segments.
Claim 44
Dependent Claim
228 words
Apparatus for determining whether or not a surface of each of a plurality of successive objects has substantially the same appearance as the corresponding surface of at least some of the preceding objects in the succession, said apparatus comprising: means for forming a two-dimensional image of said surface of each object in said succession; means for digitizing each of said twodimensional images; means for using the digitized image of the surface of said first object in said succession to form a digitized discriminant image in which the digital values are related to the digital values in the digitized image of the surface of said first object by a predetermined proportionality value; means for, for at least some of said objects after said first object in said succession, replacing the digital values in said digitized discriminant image with updated digital values respectively equal to the discriminant image digital values prior to replacement plus an amount proportional to a function of the dot product of the digitized discriminant image prior to replacement and the digitized image of the surface of said object on which said replacement is being based; and means for, for at least some of said objects after said first object in said succession, comparing the dot product of said digitized discriminant image and said digitized image of the surface of said object to a predetermined constant value.
Claim 45
Dependent Claim
228 words
Apparatus for determining whether or not a surface of each of a plurality of successive objects has substantially the same appearance as the corresponding surface of at least some of the preceding objects in the succession, said apparatus comprising: means for forming a two-dimensional image of said surface of each object in said succession; means for digitizing each of said twodimensional images; means for using the digitized image of the surface of said first object in said succession to form a digitized discriminant image in which the digital values are related to the digital values in the digitized image of the surface of said first object by a predetermined proportionality value; means for, for at least some of said objects after said first object in said succession, replacing the digital values in said digitized discriminant image with updated digital values respectively equal to the discriminant image digital values prior to replacement plus an amount proportional to a function of the dot product of the digitized discriminant image prior to replacement and the digitized image of the surface of said object on which said replacement is being based; and means for, for at least some of said objects after said first object in said succession, comparing the dot product of said digitized discriminant image and said digitized image of the surface of said object to a predetermined constant value.
Claim 46
Dependent Claim
41 words
The apparatus defined in claim 38 wherein said proportionality value is approximately equal to the reciprocal of the square root of the sum of the squares of the digital values in the digitized image of the surface of said first object.
Claim 47
Dependent Claim
111 words
The apparatus defined in claim 39 wherein said amount proportional to a function of the dot product of the digitized discriminant image prior to replacement and the digitized image of the surface of said object on which said replacement is being based is given by the expression: ##EQU36## where I.sup.(i) is the digitized image on which said replacement is being based, where |I.sup.(i) |, is equal to the square root of the sum of the squares of the digitized values in the digitized image of the surface of said first object, where ##EQU37## where F.sup.(i-l) is the digitized discriminant image prior to replacement, and where δ is a predetermined learning parameter.
Claim 48
Dependent Claim
20 words
The apparatus defined in claim 40 wherein δ has a value in the range from about 0 to about 2.
Claim 49
Dependent Claim
11 words
The apparatus defined in claim 41 wherein δ is approximately 1.
Claim 50
Dependent Claim
53 words
The apparatus defined in claim 38 wherein, for any object after said first object for which both said replacing and comparing functions may be performed, said replacing function is performed only if the results of said comparing function indicate that said dot product is within a predetermined range of said predetermined constant value.
Claim 51
Dependent Claim
22 words
The apparatus defined in claim 38 wherein said comparing function is performed only on objects following an initial subplurality of said objects.
Claim 52
Dependent Claim
24 words
The apparatus defined in claim 38 wherein said replacing function is performed only with respect to objects in an initial subplurality of said objects.
Claim 53
Dependent Claim
113 words
The apparatus defined in claim 38 wherein said means for digitizing each of said two-dimensional images comprises: means for subdividing said image into a plurality of pixels; means for digitizing each of said pixels by associating with each pixel an output digital value proportional to an optical characteristic of the associated pixel; and means for binarizing each of said output digital values by associating a first binary value with each of said output digital values which is on a first side of a first predetermined threshold value and by associating a second binary value with each of said output digital values which is on the other side of said first predetermined threshold value.
Claim 54
Dependent Claim
38 words
The apparatus defined in claim 38 further comprising: means for subdividing each of said twodimensional images into a plurality of segments and performing said functions of using, replacing, and comparing separately with respect to each of said segments.
References Cited (Backward Citations)
US3835332A
Cited
US4053056A
Cited
US4097845A
Cited
US4481664A
Cited
US4482971A
Cited
US4637054A
Cited
EP0155789A2
Cited
US4926491A
Cited
US4759074A
Cited
US4912554A
Cited
US5007096A
Cited
JPS63257083A
Cited
US4975971A
Cited
US4859863A
Cited
US4872024A
Cited
EP0330495A2
Cited
US4972494A
Cited
WO1989010596A1
Cited
US4952062A
Cited
US4975972A
Cited
US4972262A
Cited
US4974261A
Cited
US5046111A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1995-11-17 | 1996-05-18 | 1996-11-17 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1999-11-17 | 2000-05-18 | 2000-11-17 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2003-11-17 | 2004-05-18 | 2004-11-17 | $7700.00 | Window Closed / Lapsed |