Hardware architecture for Fast 2D distance transformations

Vasiliki Giannakopoulou; Kostas Masselos

doi:10.1117/12.2064544

16 April 2014 Hardware architecture for Fast 2D distance transformations

Vasiliki Giannakopoulou, Kostas Masselos

Proceedings Volume 9159, Sixth International Conference on Digital Image Processing (ICDIP 2014); 91591X (2014) https://doi.org/10.1117/12.2064544
Event: Sixth International Conference on Digital Image Processing, 2014, Athens, Greece

Abstract

Distance transformation algorithms approximations use distance transformation with small neighborhood pixels, defined by masks. There are many different DT approximations applied, and some of them are proposed to be implemented in custom hardware, exploiting the possibilities offered for accelerating the actions required in the proposed algorithm. In this paper, a custom hardware architecture is proposed in order to implement fast 2D distance transformations that are independent of the distance function used. The current implementation concerns the Euclidean distance transform and Kintex7 evaluation board is used for this purpose. Area and synthesis results are presented for various image sizes, while it is observed a speed increase of 98% against the X86 architecture.

Citation Download Citation

Vasiliki Giannakopoulou and Kostas Masselos "Hardware architecture for Fast 2D distance transformations", Proc. SPIE 9159, Sixth International Conference on Digital Image Processing (ICDIP 2014), 91591X (16 April 2014); https://doi.org/10.1117/12.2064544

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available