Muhammad Hussein
Universitas Muhammadiyah Malang

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : Jurnal Repositor

Segmentasi Citra X-Ray Paru dengan Deep Learning Muhammad Hussein; Agus Eko Minarno; Yufis Azhar
Jurnal Repositor Vol 5 No 1 (2023): Februari 2023
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/repositor.v5i1.1498

Abstract

Image segmentation is one of the main things in the study of computer vision and image processing. One example is the processing of lung x-ray images to find out diseases in the lungs. U-net is a segmentation model that has been created to make it easier for someone to build a model for image segmentation. U-net can be used on any image. From its advantages, the researchers tried to use U-net in combination with Inception, MobileNet and EfficientNet to segment medical x-ray images of the lungs. The image is resized to 512 x 512 pixels. Augmentation that is done is zoom range, height shift, width shift and horizontal flip. Epoch is 200 and batch size is 4. The best scenario in this research is to use U-net Efficientnetb0 with dice value of 0.967, Jaccard of 0.937.