Abstract—A novel algorithm is proposed for background estimation using machine learning and statistical pattern recognition. Usually the segmentation of objects in images is achieved by identifying homogeneous regions in individual images or by finding motions of objects in videos. In this paper, we combine the advantages of these approaches for the estimation of background using only two images. The proposed algorithm uses the difference between images to obtain initial estimation of background and then to refine the estimation using machine learning and statistical pattern recognition. Experimental results have shown that the proposed algorithm can achieve promising performance in terms of accuracy and speed.
Index Terms—Background Estimation, Gaussian Mixture Models, Object Segmentation, Expectation-Maximization.
J. Cai is with the the Phenomics and Bioinformatics Research centre, School of Mathematics and Statistics, University of South Australia, Mawson Lakes, Adelaide SA5095, Australia. Corresponding author: Tel:+61 8 83025533; Fax: +61 8 83025785; Email: Jinhai.Cai@unisa.edu.au email@example.com.
M. R. Golzarian is with the Phenomics and Bioinformatics Research Centre, University of South Australia, Mawson Lakes, SA5095, Australia. He is also an academic member of Department of Agricultural Engineering-Agricultural Machinery, Faculty of Agriculture, Ferdowsi University of Mashhad, Iran. Email: Mahmood.Golzarian@unisa.edu.au.
S. J. Miklavcic is with the the Phenomics and Bioinformatics Research centre, School of Mathematics and Statistics, University of South Australia, Mawson Lakes, SA 5095, Australia. He is also a member of Department of Science and Technology, University of Linkoping, Norrkoping, Sweden, and Australian Centre for Plant Functional Genomics Pty Ltd, Adelaide, SA 5064, Australia. Email: Stan. Miklavcic@unisa.edu.au.
Cite: Jinhai Cai, Mahmood R. Golzarian, and Stan J. Miklavcic, "Novel Image Segmentation Based on Machine Learning and Its Application to Plant Analysis," International Journal of Information and Electronics Engineering vol. 1, no. 1, pp. 79-84, 2011.