Skripsi
DETEKSI OBJEK THALAMI DAN HYPOCAMPAL GYRUS PADA KEPALA JANIN MENGGUNAKAN METODE U-NET DAN YOLO V3
Ultrasonography (USG) is a process or procedure in taking an object in a certain part of the body by taking high-frequency sound waves. Objects obtained from ultrasound results will be used in this study. This study uses a Convolutinonal Neural Network (CNN) to detect Transthalamic objects. This study aims to compare the detection results from previous studies using the Faster R-CNN with a new detection method which will make it easier for medical personnel to examine fetal head objects. The first process that is carried out is segmentation to simplify the process of annotating data in the detection process. This study uses the U-Net and YOLOv3 methods in the segmentation and detection processes. Of the 15 models tested, model 5 was found to be the best result with 98.3% having been validated using unseen data which yielded 97.2%. The conclusion obtained is that the YOLOv3 architecture obtains a higher final result compared to the previous detection results using Faster R-CNN of 89%.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2307005448 | T128334 | T1283342023 | Central Library (Referens) | Available |
No other version available