Self-Similar Convolution Image Distribution Histograms as Invariant Identifiers

C. E. Tuke; S. E. M. O'Keefe; Jim Austin

Self-Similar Convolution Image Distribution Histograms as Invariant Identifiers

C. E. Tuke, S. E. M. O'Keefe, Jim Austin

Computer Science

Research output: Contribution to conference › Paper › peer-review

Abstract

This paper investigates the practical application of a new method called self-similar convolution masks to binary trademark images to produce affine invariant one dimensional histogram descriptions. Because the convolution mask is a scaled version of the original image, the normalised distribution histogram of the resulting grey scale image is an affine description based purely upon the image structure, which can be then be used for database search using a database of binary images. The effectiveness of this approach to trademark image identification is tested. Multiple methods of representing the distribution histograms within the database are also tested for robustness to noise, generalisation and reliability.

Original language	English
Pages	501-510
Number of pages	10
Publication status	Published - 2001

Access to Document

http://www.bmva.org/bmvc/2001/papers/17/17.html

Cite this

@conference{bf447f85a5e649c59b2b9982e6574d6a,

title = "Self-Similar Convolution Image Distribution Histograms as Invariant Identifiers",

abstract = "This paper investigates the practical application of a new method called self-similar convolution masks to binary trademark images to produce affine invariant one dimensional histogram descriptions. Because the convolution mask is a scaled version of the original image, the normalised distribution histogram of the resulting grey scale image is an affine description based purely upon the image structure, which can be then be used for database search using a database of binary images. The effectiveness of this approach to trademark image identification is tested. Multiple methods of representing the distribution histograms within the database are also tested for robustness to noise, generalisation and reliability.",

author = "Tuke, {C. E.} and O'Keefe, {S. E. M.} and Jim Austin",

year = "2001",

language = "English",

pages = "501--510",

}

TY - CONF

T1 - Self-Similar Convolution Image Distribution Histograms as Invariant Identifiers

AU - Tuke, C. E.

AU - O'Keefe, S. E. M.

AU - Austin, Jim

PY - 2001

Y1 - 2001

N2 - This paper investigates the practical application of a new method called self-similar convolution masks to binary trademark images to produce affine invariant one dimensional histogram descriptions. Because the convolution mask is a scaled version of the original image, the normalised distribution histogram of the resulting grey scale image is an affine description based purely upon the image structure, which can be then be used for database search using a database of binary images. The effectiveness of this approach to trademark image identification is tested. Multiple methods of representing the distribution histograms within the database are also tested for robustness to noise, generalisation and reliability.

AB - This paper investigates the practical application of a new method called self-similar convolution masks to binary trademark images to produce affine invariant one dimensional histogram descriptions. Because the convolution mask is a scaled version of the original image, the normalised distribution histogram of the resulting grey scale image is an affine description based purely upon the image structure, which can be then be used for database search using a database of binary images. The effectiveness of this approach to trademark image identification is tested. Multiple methods of representing the distribution histograms within the database are also tested for robustness to noise, generalisation and reliability.

M3 - Paper

SP - 501

EP - 510

ER -