BELLA, SINTA and Nurmaini, Siti (2021) DETEKSI CARA PANDANG JANTUNG ANAK MENGGUNAKAN METODE FASTER-RCNN. Undergraduate thesis, Sriwijaya University.
Text
RAMA_56201_09011181722072.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (40MB) | Request a copy |
|
Text
RAMA_56201_09011181722072_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy |
|
Preview |
Text
RAMA_56201_09011181722072_ 0002085908_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (18MB) | Preview |
Text
RAMA_56201_09011181722072_ 0002085908_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (364kB) | Request a copy |
|
Text
RAMA_56201_09011181722072_ 0002085908_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (527kB) | Request a copy |
|
Text
RAMA_56201_09011181722072_ 0002085908_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy |
|
Text
RAMA_56201_09011181722072_ 0002085908_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (55kB) | Request a copy |
|
Text
RAMA_56201_09011181722072_ 0002085908_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (190kB) | Request a copy |
|
Text
RAMA_56201_09011181722072_ 0002085908_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (19MB) | Request a copy |
Abstract
Deteksi objek merupakan salah satu hal penting di bidang komputer dan telah berkembang dari pengenalan objek tunggal hingga pengenalan multi-objek. Dalam tugas akhir ini mengusulkan metode Faster-RCNN dengan menggunakan backbone VGG16, Resnet50 dan Mobilenet. Dataset yang digunakan adalah data frame jantung anak dengan sudut pandang Four chamber view, Five chamber view, Subcostal, Long Axis dan Short Axis. Penelitian yang dilakukan berfokus pada tingkat akurasi yang dihasilkan pada data frame untuk membangun model Faster R-CNN yang efektif dalam mendeteksi jantung bayi. Hasil mAP (mean average precision) nantinya berfungsi sebagai tolak ukur dalam mengetahui tingkat keberhasilan metode dalam mendeteksi objek terkhusus pada jantung anak. Hasil terbaik didadapatkan pada model VGG16 dengan Learning Rate 0,001 dan Learning rate 0,0001 dengan mAP sebesar 100% dan pada data unseen hasil terbaik didapatkan pada model VGG16 dengan mAP sebesar 63,46%.
Item Type: | Thesis (Undergraduate) |
---|---|
Uncontrolled Keywords: | Deteksi Objek, Faster R-CNN, VGG16, Resnet50, Mobilenet, mAP |
Subjects: | Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation. |
Divisions: | 09-Faculty of Computer Science > 56201-Computer Systems (S1) |
Depositing User: | Sinta Bella |
Date Deposited: | 04 Aug 2021 07:21 |
Last Modified: | 04 Aug 2021 07:21 |
URI: | http://repository.unsri.ac.id/id/eprint/51556 |
Actions (login required)
View Item |