Textural Image Segmentation based on Entropy and VGG16 Deep Neural Network Kernel Graph Cut

Document Type : Original Article

Authors

South Tehran Branch - Islamic Azad University

10.22034/jasp.2023.52704.1198

Abstract

Recently, image segmentation based on graph cut methods has shown remarkable performance on a set of image data. Although the kernel graph cut method provides good performance, its performance is highly dependent on the mapping of data to the transform space and image features. Entropy-based kernel graph cut method is suitable for segmentation of textured images. However, the quality of its segmentation is affected by the extracting kernel centers quality. This article examines the segmentation of textural images using the proposed weighted entropy and VGG16-based kernel graph cut method. Using the advantages of kernel space, the objective function consists of two data components to transfer the data standard deviation of each area in the segmented image and the adjustment component. The proposed method, while taking advantage of the appropriate computational load of graph-based methods, will be a suitable alternative for textural image segmentation. Experimental results have been taken on a set of well-known datasets that include textural shapes in order to evaluate the effectiveness of the algorithm in compared to state-of-the-art methods in the field of kernel graph cut.

Keywords

Main Subjects

Advanced Signal Processing

Articles in Press, Accepted Manuscript
Available Online from 21 May 2023

Files

Share

How to cite

Statistics