A Multi-Stage Approach for Automatic Classification of Environmental Microorganisms

• Published in: Proceedings of the International Conference on Image Processing, Computer Vision, and Pattern Recognition (IPCV) p. 1
• Main Authors: Li, Chen, Shirahama, Kimiaki, Czajkowska, Joanna, Grzegorzek, Marcin, Ma, Fangshu, Zhou, Beihai
• Format: Conference Proceeding
• Language: English
• Published: Athens The Steering Committee of The World Congress in Computer Science, Computer Engineering and Applied Computing (WorldComp) 01.01.2013

Environmental Microorganisms (EMs) are currently recognised using molecular biology (DNA, RNA) or morphological methods. The first ones are very time-consuming and expensive. The second ones require a very experienced laboratory operator. To overcome these problems, we introduce an automatic classification method for EMs in the framework of content-based image analysis. In order to efficiently segment the EM structure, six segmentation approaches are tested. A Sobel edge detector based semi-automatic segmentation approach achieves the best evaluation performance on the testing data. To describe the shapes of EMs in microscopic images, we use Edge Histograms, Fourier Descriptors, extended Geometrical Features, as well as introduce Internal Structure Histograms. For classification, multi-class Support Vector Machine is applied to EMs represented by the above features. In order to quantitatively evaluate discriminative properties of features, we perform comprehensive experiments with a ground truth of manually segmented microscopic EM images. The classification result of 89.7% proves a high robustness of our method. [PUBLICATION ABSTRACT]