Lutfia, Qonita (2024) Implementasi Local Adaptive Thresholding dan Watershed Dalam Segmentasi Sel Pap Smear Serviks Tumpang Tindih. Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
Cover.pdf Download (829kB) |
|
Text (Bab 1)
Bab 1.pdf Download (169kB) |
|
Text (Bab 2)
Bab 2.pdf Restricted to Repository staff only until 19 July 2026. Download (485kB) |
|
Text (Bab 3)
Bab 3.pdf Restricted to Repository staff only until 19 July 2026. Download (3MB) |
|
Text (Bab 4)
Bab 4.pdf Restricted to Repository staff only until 19 July 2026. Download (3MB) |
|
Text (Bab 5)
Bab 5.pdf Download (171kB) |
|
Text (Daftar Pustaka)
Daftar Pustaka.pdf Download (103kB) |
|
Text (Lampiran)
Lampiran.pdf Restricted to Repository staff only until 19 July 2026. Download (14MB) |
Abstract
Cervical cancer is the third most common cancer in women and a serious concern in Indonesia. Lack of awareness of symptoms and lack of access to early screening are major factors in the high mortality rate. Pap smear examination is important for early detection of cervical cancer, but manual analysis is prone to human error and difficulty in separating abnormal cells that are stacked on top of each other. To separate the stacked abnormal cervical cells, a special segmentation method such as Thresholding is required. This study aims to segment the abnormal cervical cell region using Local Adaptive Thresholding and Watershed. This computational research focuses on developing and evaluating image segmentation algorithms to identify and separate abnormal cervical cells in Pap smear images. The Local Adaptive Thresholding method is used to locally adjust the threshold on the image, enabling more accurate segmentation of regions with varying illumination. Meanwhile, the Watershed method is applied to separate stacked cells with additional techniques such as the use of markers to overcome over-segmentation. The results show that the combination of Local Adaptive Thresholding and Watershed method provides good performance in segmenting stacked Pap smear cervical cells. Evaluation using 5-fold and 7-fold cross validation methods showed the success of this approach with a very high average accuracy of 90.93% for both cross validation methods. However, analysis of the precision, recall, and F1-Score metrics showed that although the precision was very high (97.97%), the recall was still relatively low (49.22%), indicating the method is very good at identifying positive cells but less effective at identifying all positive cases that actually exist.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics Q Science > QA Mathematics > QA76.6 Computer Programming T Technology > T Technology (General) |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Qonita Lutfia | ||||||||||||
| Date Deposited: | 19 Jul 2024 07:25 | ||||||||||||
| Last Modified: | 19 Jul 2024 07:25 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/26614 |
Actions (login required)
 |
View Item |