Edge Detection using Enhanced Ant Colony System Algorithm and Standard Deviation in Medical Images

In the medical field, images are essential for both diagnosis of diseases and treatment. X-ray imaging is widely utilized by medical professionals to identify subtle bone fractures due to its painless procedure and cost-effectiveness compared to other imaging techniques. Edges in X-ray images play a guiding role in diagnosing fractures, which are identified by edge detection. It is a method for identifying locations in pictures where the intensity of the pixels varies suddenly. Ant colony optimization is a popular metaheuristic algorithm. The basic concept is that ants create a pheromone matrix by depositing a chemical pheromone while they are looking for food. The pheromone developed by ants dispatching on the image provides edge information available. The movements of ants depend on the local variance of the image intensity value. A novel method for image edge detection using an enhanced ant colony system (ACS) algorithm and standard deviation is presented in this paper. Standard deviation-based heuristic data is employed. In this work, the minimizing standard deviation-based objective function estimates the optimum threshold from the updated pheromone matrix further used to generate a binary image. The suggested method was examined with standard X-ray images. Results from experiments show that the suggested approach works better with existing conventional edge detectors in terms of performance parameters like Figure of Merit and Entropy for medical images. The proposed method demonstrates an approximate 29% improvement in accuracy compared to traditional ACS.