ANALISIS METODE PREPROCESSING PADA CITRA MAMMOGRAM

SURAHMAT LAGUNI, NIM. 12651044 (2016) ANALISIS METODE PREPROCESSING PADA CITRA MAMMOGRAM. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

[img]
Preview
Text (ANALISIS METODE PREPROCESSING PADA CITRA MAMMOGRAM)
12651044_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf

Download (3MB) | Preview
[img] Text (ANALISIS METODE PREPROCESSING PADA CITRA MAMMOGRAM)
12651044_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf
Restricted to Registered users only

Download (3MB)

Abstract

Mammogram image is an image result from an X-ray examination process called mammography. Mammographic examination is performed to detect breast cancer. This enables a doctor to see any abnormality in a patient as early as possible. The problem is the similarity between abnormalities and non-abnormalities in mammogram images due to nearly the same contrast level. Therefore, a serious preprocessing stage is required to improve the accuracy of detection and clarify parts of images. Preprocessing stage includes ROI selection and image conversion to grayscale. After grayscaling, filtering is performed using Adaptive Median Filter method. Then, image contrast is modified using CLAHE method. After improving the contrast, an image is segmented using Region Growing method. The results of analysis and test showed that the implementation of preprocessing method had adequate result. It’s evidence in AMF method where the average value of PSNR is 42,3092 dB with standard deviation of 1,98 and the average value of MSE is 4,1386 with standard deviation of 1,36. While using MF method, the average value of PSNR is 9.0918 with standard deviation of 3,08. The result of improving contrast by CLAHE method showed the best result at clip limit 2 and region size 2 x 2 as evidenced by the completeness of image features and accuracy of segmentation result.

Item Type: Thesis (Skripsi)
Additional Information: Dr. Shofwatul ‘Uyun M.Kom.
Uncontrolled Keywords: Preprocessing on mammogram image, Contrast Limited Histogram Equalization, Adaptive Media Filter, Region growing
Subjects: Tehnik Informatika
Divisions: Fakultas Sains dan Teknologi > Teknik Informatika (S1)
Depositing User: Miftahul Ulum [IT Staff]
Date Deposited: 03 Aug 2016 11:53
Last Modified: 03 Aug 2016 11:53
URI: http://digilib.uin-suka.ac.id/id/eprint/21309

Share this knowledge with your friends :

Actions (login required)

View Item View Item
Chat Kak Imum