Image analysis – Pattern recognition – Feature extraction
Reexamination Certificate
2006-07-11
2006-07-11
Bali, Vikkram (Department: 2623)
Image analysis
Pattern recognition
Feature extraction
Reexamination Certificate
active
07076098
ABSTRACT:
An image area extract unit includes a modeling unit for generating a pixel evaluation model used for extraction, and an extract unit for extracting a characteristic portion from an original image using the pixel evaluation model. The modeling unit sequentially generates a plurality of partial polynomials according to a modified Group Method of Data Handling using a training image taken under the same condition under which the original image is taken and a supervisory image that designates the characteristic portion of the training image. Each of the generated partial polynomials is outputted to the extract unit in the form of its coefficients only when a square error satisfies a predetermined criterion. The extract unit calculates the feature values for each pixel of the original image using the pixel evaluation model, and defines the extractive area that includes the characteristic portion based on the feature values.
REFERENCES:
patent: 5774129 (1998-06-01), Poggio et al.
patent: 6148099 (2000-11-01), Lee et al.
patent: 6611613 (2003-08-01), Kang et al.
patent: A-2-215276 (1990-08-01), None
patent: A-6-233761 (1994-08-01), None
patent: A-7-160881 (1995-06-01), None
patent: A-2000-22929 (2000-01-01), None
patent: A-2000-194859 (2000-07-01), None
Ishikawa Takahiro
Kawashima Takeshi
Bali Vikkram
Denso Corporation
Posz Law Group , PLC
LandOfFree
Adaptive image processing unit for extracting characteristic... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Adaptive image processing unit for extracting characteristic..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Adaptive image processing unit for extracting characteristic... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3575037