Graph Based Image Segmentation by Dinic Algorithm
Prajjwal Singh, Priti Chakurkar, Varsha Naik · 2023
In Computer Vision objects are represented by images and analyzing labeling segmenting them is important for processing them. To solve this need image segmentation comes into the picture. There are numerous approaches to implementing image segmentation deep learning, graph-based, threshold-based, etc. Each has its pros and cons and is suited for the particular use case. In this paper, we will be looking at the graph-based approach which is an advanced one and fairly new in digital image processing.As of now, we have a lot of a number of methods present that can extract the required foreground from the background.But, most of these methods are solely based on the boundary or regional information which has limited the segmentation result. Thus graph cut based approaches have been proposed which has obtained a lot of popularity because it utilizes both boundary and regional information. Furthermore, graph cut based method is efficient and accepted throughout the world since it can achieve the globally optimal result for the energy function.Graph-based image segmentation approaches itself has implementation based upon the algorithm used like Ford Fulkerson, Edmond Karp, Dinic Algorithm, etc. We will be focusing on Dinic here as it is the fastest and most efficient among its peers and there is hardly any literature shedding light on it. We will look into comparative analysis, algorithm, and how it can be applied to image segmentation problems.