Microorganism image clustering based on type-2 fuzzy sets and restricted equivalent functions

A microorganism is an organism that can be unicellular, multicellular, and acellular type and visible only under a microscope, which includes bacteria, protozoa, algae, and fungi. They play a significant role in many processes like biogeochemical cycling, climate regulation, food security, bioremediation, and medicine. Therefore, a broad and deep understanding of microorganisms and their ecosystem or inhabited areas are extremely important. Due to great advantages of automatic image processing techniques over the traditional methods, they are broadly applied for analysis, enhancing microorganism’s classification, segmentation, and identification. Clustering methods are playing very important roles in microorganism image monitoring, object recognition, segmentation, key point detection, and image categorization or classification. Clustering or segmentation procedures use exceptionally simple algorithm, and they are very efficient and easy to apply. But the segmentation of microorganisms is difficult due to its uncertainties present in the images. The uncertainties occur due to the difficulty in the image capturing, noises, and other perturbations. In this method, we offer a type-2 fuzzy set (T1FS)-based technique to cluster the microorganism. The T1FS is helpful to reduce the vagueness or uncertainty effectively. In the proposed method, the fuzzy set is extracted from a set of restricted equivalent functions. The method is found to be quite efficient compared to leading-edge approach.