Fusion on Citrus Image Data from Cold Mirror Acquisition System

Fusion on Citrus Image Data from Cold Mirror Acquisition System

Peilin Li (School of Advanced Manufacturing and Mechanical Engineering, University of South Australia, Mawson Lakes, SA, Australia), Sang-Heon Lee (School of Advanced Manufacturing and Mechanical Engineering, University of South Australia, Mawson Lakes, SA, Australia) and Hung-Yao Hsu (School of Advanced Manufacturing and Mechanical Engineering, University of South Australia, Mawson Lakes, SA, Australia)
Copyright: © 2012 |Pages: 14
DOI: 10.4018/ijcvip.2012100102
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

In this paper, an image fusion is presented to improve the citrus identification by filtering the incoming data from two cameras. The citrus image data has been photographed by using a portable bi-camera cold mirror acquisition system. The prototype of the customized fixture has been manufactured to position and align a classical cold mirror with two CCD cameras in relative kinematic position. The algorithmic registration on the pairwise images has been bypassed by both the spatial alignment of two cameras with recourse of software calibration and the triggering synchronization in temporal during the photographing. The pairwise frames have been fused by using the Daubechies wavelets decomposition filters. The pixel level fusion index rule is proposed to combine the low pass coefficients of the visible image and the low pass coefficients of the near-infrared image convoluted by the complementary of entropy filter from the visible low pass coefficients. In the study, the fused artifact color image and the non-fused color image have been processed and compared by some classification methods such as low dimensional projection, self-organizing map and the support vector machine.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 7: 4 Issues (2017)
Volume 6: 2 Issues (2016)
Volume 5: 2 Issues (2015)
Volume 4: 2 Issues (2014)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing