Affiliation:
1. State University of Rio de Janeiro (UERJ), Rio de Janeiro, Brazil
Abstract
Template matching is an important method used for object tracking, in order to find a given-pattern within a frame sequence. Pearson's Correlation Coefficient (PCC) is widely used to quantify the similarity between two images. Since this coefficient calculus is computed for each image pixel, it entails a computationally expensive process. In this article, an embedded co-design system is proposed, which implements the template matching, in order to accelerate this process. The dedicated co-processor, responsible for performing the PCC computation, is used in two configurations: serial and pipeline. Cuckoo Search (CS) is used to improve the search for the maximum correlation point of the image and the used template. The search process is implemented in software and is run by an embedded general-purpose processor. The performance results are compared to those obtained through Particle Swarm Optimization (PSO) using the same hardware approach.
Subject
Artificial Intelligence,Computational Theory and Mathematics,Computer Science Applications
Reference27 articles.
1. Object recognition by template matching using correlations and phase angle method.;K.Ahuja;International Journal of Advanced Research in Computer and Communication Engineering.,2013
2. Stable multi-target tracking in real-time surveillance video
3. UAV guidance using a monocular-vision sensor for aerial target tracking
4. An open source tracking testbed and evaluation web site.;R.Collins;IEEE International Workshop on Performance Evaluation of Tracking and Surveillance,2005
5. A comprehensive review of cuckoo search: variants and hybrids