Hardware acceleration based connected component labeling algorithm in real-time ATR system

Fei Zhao; Zhi-yong Zhang

doi:10.1117/12.2014150

13 March 2013 Hardware acceleration based connected component labeling algorithm in real-time ATR system

Fei Zhao, Zhi-yong Zhang

Proceedings Volume 8784, Fifth International Conference on Machine Vision (ICMV 2012): Algorithms, Pattern Recognition, and Basic Technologies; 87841S (2013) https://doi.org/10.1117/12.2014150
Event: Fifth International Conference on Machine Vision (ICMV 12), 2012, Wuhan, China

Abstract

Aims at the requirement of real-time processing in Real-Time Automatic Target Recognition(RTATR) system, this paper presents a hardware acceleration based two-scan connected-component labeling algorithm. Conventional pixel and run based algorithm’s merits are combined, in the first scan, the pixel is processed scan unit while line as label unit, label equivalences are recorded while scanning the image by pixel. Lines with provisional label are outputted as the connected component labeling result. Then the union-find algorithm is used for resolving label equivalences and finds the representative label for each provisional label after the first scan. The labels are replaced in the second scan to complete the connected-component labeling. Experiments on RTATR platform demonstrate that the hardware acceleration implementation of algorithm reaches a higher performance and efficiency and consumes few resources. The implementation of proposed algorithm can meet the demand of real-time processing, and possesses a better practicability.

Citation Download Citation

Fei Zhao and Zhi-yong Zhang "Hardware acceleration based connected component labeling algorithm in real-time ATR system", Proc. SPIE 8784, Fifth International Conference on Machine Vision (ICMV 2012): Algorithms, Pattern Recognition, and Basic Technologies, 87841S (13 March 2013); https://doi.org/10.1117/12.2014150

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