Skripsi

PENERAPAN ARSITEKTUR U-NET DAN YOLO V3 DALAM MENDETEKSI OBJEK TRANSVENTRIKULAR PADA CITRA KEPALA JANIN

---

Penilaian

---

The content is a research summary about the process of examining the contents of the uterus using Ultrasonography (USG) conducted by doctors. USG is a device that utilizes high-frequency sound waves to produce video images used by experts to identify vital objects within the uterus. The resulting USG images are used for this particular research, which utilizes Convolutional Neural Network technology to detect objects of transventricular fetal head USG. The research's objective is to compare the accuracy of transventricular object detection on fetal heads using the Faster-RCNN architecture from a previous study with the detection method using the YOLOv3 architecture. The segmentation process is performed beforehand to aid in labeling during the detection process. U-Net and YOLOv3 architectures are selected for segmentation and detection processes. Among the 20 models created, Model 11 yields the best results with 92.1% accuracy, and when validated with unseen data, it achieves 88.1% accuracy. The conclusion drawn from the study is that the YOLOv3 architecture outperforms the Faster-RCNN architecture, achieving a final accuracy of 92.1%, compared to Faster-RCNN's 65%.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2307003703	T127738	T1277382023	Central Library (REFERENS)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1277382023

Publisher

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

Collation

xv, 138 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

003.5 07

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Teori komunikasi dan kontrol
Jurusan Sistem Komputer

Specific Detail Info

-

Statement of Responsibility

ANUG

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment