A Novel Hybrid Approach to Image Segmentation: Integrating Cellular Automata with U-Net

View at publisher Summarise with VIRAT
Authors: A. Kamatchi, R. Arulprakasam
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Image segmentation is essential in medical imaging because it identifies and delineates regions of interest, including organs and pathological segments. To produce precise segmentation masks, traditional methods depend on complex neural network-based models such as U-Net, Res U-Net, and U-Net++, which require a large amount of time and computing power to achieve significant accuracy. This study integrates cellular automata along with U-Net to further enhance the segmentation process. We initially use a CA method to generate a rough mask for each image, which enables the next deep-learning segmentation model to begin. We contend that the segmentation model can converge faster and provide more accurate final segmentation masks with this early mask. According to evaluation outcomes, our hybrid approach enhances the overall accuracy of the generated masks while simultaneously speeding up the training process. This approach has the potential to improve the efficiency and precision of image segmentation in medical imaging, as well as in other fields.

Fields of Study & Subjects

Medical Image Segmentation Techniques Advanced Neural Network Applications Brain Tumor Detection and Classification

Publication details

Journal / source
CRC Press eBooks
Pages
251–257
Publication date
26 March 2026
Language
EN
Document type
Book chapter