Fitriana, Ida (2014) Peningkatan kualitas citra dental panoramic radiograph pada tulang mandibula menggunakan multi-histogram equalization. Undergraduate thesis, Universitas Islam Negeri Maulana Malik Ibrahim.
|
Text (Fulltext)
09650087.pdf - Accepted Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (3MB) | Preview |
Abstract
INDONESIA:
Tulang mandibula merupakan tulang rahang pembentuk wajah yang paling besar, berat dan kuat. Adanya hubungan antara kepadatan mineral tulang mandibula dengan bagian kerangka lain yang umumnya digunakan untuk densitometry tulang dalam mendeteksi osteoporosis. Penelitian ini meningkatkan kualitas citra Dental Panoramic Radiograph pada tulang mandibula dengan mengambil bagian tulang mandibula dan akar gigi. Memisahkan akar gigi dengan tulang mandibula untuk mendapatkan bagian tulang mandibula secara utuh. Metode yang digunakan yaitu Multi-Histogram Equalization dengan melakukan dekomposisi image menjadi beberapa sub-images, menerapkan histogram equalization dan threshold pada setiap sub-image agar peningkatan citra maksimal. Peneliti menggunakan metode segmentasi tambahan yaitu metode transformasi watershed. Hasil uji coba dilakukan dengan menggunakan metode Multi-Histogram Equalization dan tanpa metode. Kinerja metode Multi-Histogram Equallization untuk menunjang aplikasi ini memiliki akurasi, presisi, dan recall sebesar 79.48%, 92.37%, dan 27.64% saat image split 4. Sedangkan uji coba tanpa menggunakan metode Multi-Histogram Equalization memiliki akurasi, presisi, dan recall sebesar 59.89%, 100%, dan
0.5%.
ENGLISH:
Mandibula bone is the largest face shaper of jaw bone, heavy and strong. There is a relationship between mineral dense of mandibula bone and other sekeleton that generally used for bone densitometry in osteoporosis detection. This research improves the image quality of Dental Panoramic Radiograph in mandibular bone by taking part mandibular bone and tooth root. Separating the roots of teeth with mandibular bone to get the parts intact. The method used is Multi-Histogram Equalization to perform image decomposition into multiple sub-images, histogram equalization and threshold apply to each sub-image so that the maximum image enhancement. Researchers using additional segmentation method is the method of watershed transformation. The trial results were calculated using the Multi-Histogram Equalization and without method. Performance Multi-Histogram to support these applications have accuracy, precision, and recall of 79.48%, 92.37%, and 27.64% when the split image 4. While testing without using the Multi-Histogram Equalization has accuracy, precision, and recall of 59.89%, 100%, and 0.5%.
Item Type: | Thesis (Undergraduate) | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
Supervisor: | Kusumawati, Ririen and Abidin, Zainal | |||||||||
Contributors: |
|
|||||||||
Keywords: | Tulang Mandibula; Mandibula Bone; Osteoporosis; Dental Panoramic Radiograph; Multi-histogram Equalization; Thresholding Watershed | |||||||||
Departement: | Fakultas Sains dan Teknologi > Jurusan Teknik Informatika | |||||||||
Depositing User: | Dian Anesti | |||||||||
Date Deposited: | 13 Oct 2017 13:59 | |||||||||
Last Modified: | 13 Oct 2017 13:59 | |||||||||
URI: | http://etheses.uin-malang.ac.id/id/eprint/8039 |
Downloads
Downloads per month over past year
Actions (login required)
View Item |