Image Segmentation of Microalgae Cells Using Fuzzy Logic for Edge Detection

Abstract

One of the relatively unexplored field in microalgae technology is the use of digital image processing. Images of microalgae cells can be digitized to be able to perform various procedures to characterize and approximate their population counts. In digital image processing, images can be segmented to specify regions of interest that can be used for cell counting. However, microalgae cell structures differ from one specie to another. They vary in shapes, color, and even sizes. This results to declines in robustness of commonly used algorithm in image segmentation. A problem in choosing the suitable procedure to do so arises. In this paper, a method of segmenting images using fuzzy image analysis is proposed. The aim is to apply fuzzy logic in order to create a simple method of segmentation that can be useable for a wide variety of microalgae specimens based on the image gradient and color. Fuzzification is employed in order to overcome inhomogeneity in the images. The hardware is composed of a microscope, a webcam for image acquisition, and a computer for the image analysis. Although the target species of the study are microalgae cells that are used in the production of biofuel, the system will be implemented to multiple sample images and microalgae varieties.