Skripsi

SEGMENTASI SEMANTIK MULTICLASS PADA CITRA PRA-KANKER SERVIKS MENGGUNAKAN DEEP LEARNING

---

Penilaian

---

Artificial Intelligence (AI) technology in the field of Computer Vision (CV) can be utilized to detect cervical cancer using the Image Segmentation approach. With the advancement of technology, image segmentation processes can be implemented using Deep Learning (DL) models. This research utilizes the U-Net architecture with backbones for pre-cervical cancer semantic image segmentation. There are 12 backbones combined with the U-Net architecture: Vgg16, Vgg19, ResNet50, ResNext50, EfficientNetb7, InceptionResNetv2, DenseNet201, Inceptionv3, Mobilenetv2, Se-ResNet50, SE-ResNext50, and SE-Net154. The best performance is achieved by the U-Net model using the SENet154 backbone. The evaluation results of the U-Net model with the SENet154 backbone on the metrics of Pixel Accuracy, Intersection Over Union (IoU), and Dice Coefficient are 81.82%, 71.69%, and 82.29%, respectively.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2307005124	T124963	T1249632023	Central Library (Referens)	Available

Detail Information

Series Title

-

Call Number

T1249632023

Publisher

Indralaya : Jurusan Sistem Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya., 2023

Collation

xiii, 69 hlm.; Ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

004.678

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Internet
Jurusan Sistem Komputer

Specific Detail Info

-

Statement of Responsibility

FIRA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment