Cell image segmentation using bacterial foraging optimization

Edge detection is the most commonly used method for cell image segmentation, where local search strategies are employed. Although traditional edge detectors are computationally efficient, they are heavily reliant on initialization and may produce discontinuous edges. In this paper, we propose a bacterial foraging-based edge detection (BFED) algorithm to segment cell images. We model the gradients of intensities as the nutrient concentration and propel bacteria to forage along nutrient-rich locations that mimic the behavior of Escherichia coli. Our nature-inspired evolutionary algorithm, can identify the desired edges and mark them as the tracks of bacteria. We have evaluated our algorithm against four edge detectors − the Canny, SUSAN, Verma's and an active contour model (ACM) technique − on synthetic and real cell images. Our results indicate that the BFED algorithm identifies boundaries more effectively and provides more accurate cell image segmentation.