Date of Award

1-1-2012

Thesis Type

masters

Document Type

Thesis

Divisions

science

Department

Institute of Biological Sciences

Institution

University of Malaya

Abstract

Cyanobacteria can be used as indicators to offer relatively exclusive information pertaining to ecosystem condition. Cyanobacteria react quickly and predictably to a broad range of pollutant. Thus provides potentially constructive early caution signals of worsening environment and the possible causes. Therefore the aim of this study is to develop an image processing and pattern recognition methods to detect and classify oscillatoria genus from Cyanobacteria found on tropical Putrajaya Lake (Malaysia). Computer-based image analysis and pattern recognition methods were used to construct a system that is able to identify, and classify selected Cyanobacteria genus automatically. An image analysis algorithm was implemented to contrast, filter, isolate and recognize objects from microscope images. Image preprocessing module used to contrast images, to remove the noise, and to improve image quality. Segmentation module used to isolate the different objects found in input image. A combination of Feed forward artificial neural network (ANN) with feature extraction module was used to train and recognize oscillator images. System accuracy was measured by using manual and automated classifying methods, and developed system showed a great accuracy system reach to 90%.

Note

Submitted in partial fulfillment of the requirements for the degree Of Master Of Bioinformatics

Share

COinS