A Survey of Image Segmentation Based on Soft Computing Approaches

Abstract

Image segmentation is an essential step in image processing, that aim to make image analyzing easier. Image segmentation can be done by grouping or partitioning the image pixels into identical sets (regions) by relying on some qualities such as color, or texture, etc. It's have been used in many field such as object detection, recognition tasks, medical imaging and much more. there are basically two primary approaches for image segmentation which are traditional approaches and soft computing approaches (SCA), a lot of methods have been proposed based on these two approaches. SCA have many advantage over traditional approaches like flexibility, cost-effective, high performance. SCA involve using fuzzy logic, Artificial Neural Network (ANN) and Genetic Algorithm. This paper focus on providing a state-of-the-art new review and summaries of researchers' work on image segmentation based on different SCA. That will help the new researchers to learn about these methods and then choosing a certain method from these for improving or developing it to produce a new method for image segmentation.