Patent EP0737932A2 successfully refreshed from Google Patents.
EP0737932A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
An apparatus for extracting skin pattern features and a skin pattern image processor
Assignee: NEC Corp
•
Filing: 1996-04-12
•
Granted: —
•
Expires: 2016-04-12
Abstract
Abstract A skin pattern image is filtered by convolution calculation on a real plane or product calculation on a Fourier transformed plane with each of two-dimensional filters prepared according to kinds of features to be extracted of the skin pattern image. Image intensity of each subregion of each filtered data is calculated. Feature of each subregion is represented by a feature parameter corresponding to a filter that gives a highest value of the image intensity of the subregion. An initial pattern of features thus obtained is smoothed to minimize an evaluation function. A filtered skin pattern image having pixel values of each subregion smoothed by a set of filters prepared for extracting a kind of features is further filtered by another set of filters for extracting another kind of features. <IMAGE>
Inventors
- 👤 Toshio Kamei
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-11-09 14:25
CPC Classifications
G; G06; G06V; G06V40/00; G06V40/10; G06V40/12; G06V40/1347
Claims Analysis
Total Claims: 9
Claim 1
Independent Claim
136 words
An apparatus for extracting features of predetermined subregions of a skin pattern image, comprising: an image memory (11) for storing image data of the skin pattern image; a filter file (15) for storing a plurality of filters consisting of two-dimensional data prepared according to the features to be extracted; filtering means (12) for outputting filtered data sets, each of said filtered data sets obtained from said image data by convolution calculation with each of said plurality of filters; image intensity calculation means (13) for calculating image intensity of each of the predetermined subregions of each of said filtered data sets; and feature extracting means (14) for determining the features of the predetermined subregions of the skin pattern image referring to said image intensity of each of the predetermined subregions of each of said filtered data sets.
Claim 2
Independent Claim
158 words
An apparatus for extracting features of predetermined subregions of a skin pattern image, comprising: an image memory (11) for storing image data of the skin pattern image; a filter file (15) for storing a plurality of filters consisting of two-dimensional data prepared according to the features to be extracted; filtering means (24) for outputting filtered data sets, each of said filtered data sets being an inverse Fourier transform of a masked data set obtained from a Fourier transform of said image data by product calculation of each frequency component of said Fourier transform and each of said plurality of filters; image intensity calculation means (13) for calculating image intensity of each of the predetermined subregions of each of said filtered data sets; and feature extracting means (14) for determining the features of the predetermined subregions of the skin pattern image referring to said image intensity of each of the predetermined subregions of each of said filtered data sets.
Claim 3
Dependent Claim
71 words
An apparatus for extracting features of predetermined subregions of a skin pattern image, recited in one of claim 1 and claim 2; wherein: said feature extracting means (14) determine the features of the predetermined subregions of the skin pattern image according to feature parameters of each of said plurality of filters which gives a maximum value of said image intensity of each of the predetermined subregions among said plurality of filters.
Claim 4
Dependent Claim
121 words
An apparatus for extracting features of predetermined subregions of a skin pattern image, recited in one of claim 1 and claim 2; wherein: said feature extracting means (14) determine the features of the predetermined subregions of the skin pattern image according to feature parameters minimizing an evaluation function, said evaluation function being a linear function of said image intensity of each of the subregions of one of said filtered data sets corresponding to a feature parameter assigined to said each of the subregions, and a value corresponding to singularity of said feature parameter assigned to said each of the predetermined subregions compared to the feature parameters assigned to certain of the predetermined subregions neighbouring to said each of the predetermined subregions.
Claim 5
Dependent Claim
208 words
A skin pattern image processor for generating a filtered skin pattern image by smoothing and enhancing features of predetermined subregions of a skin pattern image, comprising: an image memory (11) for storing image data of the skin pattern image; a filter file (15) for storing a plurality of filters consisting of two-dimensional data prepared according to the features to be extracted; filtering means (12) for outputting filtered data sets, each of said filtered data sets obtained from said image data by convolution calculation with each of said plurality of filters; image intensity calculation means (13) for calculating image intensity of each of the predetermined subregions of each of said filtered data sets; and feature extracting means (14) for determining the features of the predetermined subregions of the skin pattern image referring to said image intensity of each of the predetermined subregions of each of said filtered data sets; and pixel value calculating means (75) for calculating a value of each pixel of the filtered skin pattern image from values of pixels of certain of said filtered data sets corresponding to said each pixel, each of said certain of said filtered data sets corresponding to each of the features extracted of the predetermind subregions where said each pixel belongs.
Claim 6
Dependent Claim
230 words
A skin pattern image processor for generating a filtered skin pattern image by smoothing and enhancing features of predetermined subregions of a skin pattern image, comprising: an image memory (11) for storing image data of the skin pattern image; a filter file (15) for storing a plurality of filters consisting of two-dimensional data prepared according to the features to be extracted; filtering means (24) for outputting filtered data sets, each of said filtered data sets being an inverse Fourier transform of a masked data set obtained from a Fourier transform of said image data by product calculation of each frequency component of said Fourier transform and each of said plurality of filters; image intensity calculation means (13) for calculating image intensity of each of the predetermined subregions of each of said filtered data sets; and feature extracting means (14) for determining the features of the predetermined subregions of the skin pattern image referring to said image intensity of each of the predetermined subregions of each of said filtered data sets; and pixel value calculating means (75) for calculating a value of each pixel of the filtered skin pattern image from values of pixels of certain of said filtered data sets corresponding to said each pixel, each of said certain of said filtered data sets corresponding to each of the features extracted of the predetermind subregions where said each pixel belongs.
Claim 7
Dependent Claim
62 words
A skin pattern processor recited in one of claim 5 and claim 6, wherein: said feature extracting means (14) determine the features of the predetermined subregions of the skin pattern image according to feature parameters of each of said plurality of filters which gives a maximum value of said image intensity of each of the predetermined subregions among said plurality of filters.
Claim 8
Dependent Claim
112 words
A skin pattern processor recited in one of claim 5 and claim 6, wherein: said feature extracting means (14) determine the features of the predetermined subregions of the skin pattern image according to feature parameters minimizing an evaluation function, said evaluation function being a linear function of said image intensity of each of the subregions of one of said filtered data sets corresponding to a feature parameter assigined to said each of the subregions, and a value corresponding to singularity of said feature parameter assigned to said each of the predetermined subregions compared to the feature parameters assigned to certain of the predetermined subregions neighbouring to said each of the predetermined subregions.
Claim 9
Dependent Claim
51 words
A skin pattern processor recited in one of claim 5 and claim 6, wherein said image data of the skin pattern image are image data of a filtered skin pattern image processed by another skin pattern processor by smoothing and enhancing the features of another kind of a skin pattern image.
References Cited (Backward Citations)
JPS5177138A
Cited
JPH0518195A
Cited
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.