(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Image enhancement of endoscopic gastritis images for feature extraction

Image enhancement of endoscopic gastritis images for feature extraction / Tan Kok Chun
Sejak beberapa dekad yang lalu, teknologi kejuruteraan telah berkembang dari semasa ke semasa. Kebanyakan aplikasi kejuruteraan telah digunakan dalam sektor perubatan. Dengan menggunakan teknologi ini, masalah-masalah yang dihadapi oleh doktor-doktor boleh diselesaikan. Endoskopi merupakan satu aplikasi kejuruteraan dalam sektor perubatan. Endoskopi adalah satu proses memasukkan endoskop ke dalam badan untuk memerhati secara visual dan mengenal pasti penyakit pesakit-pesakit. Proses endoskopi itu boleh dicatatkan dan gambar-gambar tersebut boleh disimpankan. Biasanya, gambar-gambar itu mempunyai kontras yang rendah. Oleh itu, matlamat projek ini adalah untuk meningkatkan kualiti gambar-gambar endoskopi gastritis itu. Penapis “Gaussian” dan penapis “Wiener” telah digunakan. Selepas itu, teknik-teknik iaitu CHE, AHE, BPDFHE dan NMHE telah digunakan untuk meningkatkan kontras gambar-gambar tersebut. Beberapa teknik pentaksiran juga digunakan untuk menganalisa algoritma-algoritma tersebut. Keputusan menunjukkan penapis “Gaussian” bersaiz 3×3 dan sisihan piawai 1.0 mempunyai keputusan yang bagus. Penapis “Gaussian” boleh mengurangkan hingar dari 40% hingga 70% dalam gambar-gambar tersebut. BPDFHE telah digunakan untuk meningkatkan kontras gambar-gambar tersebut. BPDFHE menghasilkan kontras yang paling bagus. Berdasarkan teknik pentaksiran, BPDFHE juga mempunyai keputusan yang bagus berbanding dengan algoritma lain. Akhirnya, antara muka grafik pengguna (GUI) telah direkakan untuk menyenangkan pengguna-pengguna. _______________________________________________________________________________________________________ For the past few decades, engineering technologies have become more advanced. A lot of engineering applications have been applied in the field of medicine. With these advanced technologies, the problems faced by doctors can be solved. A common application of engineering in medicine is endoscopy. Endoscopy is a process where an endoscope is inserted into the human’s body to observe visually and to identify diseases of patients. The whole endoscopy procedure can be recorded and images can be captured. Usually, the images have low contrast as they are distorted by noise. In this project, the objective is to enhance endoscopic gastritis images. Gaussian filter and Wiener filter have been applied to the images. Then, contrast enhancement have been carried out by using several histogram equalization techniques, which are CHE, AHE, BPDFHE and NMHE. Qualitative and quantitative assessments have also been carried out to analyze the algorithms. It has been shown that Gaussian filter with kernel 3×3 and standard deviation of 1.0 gave the best result. It helped to reduce about 40% to 70% of the noise in the images. BPDFHE has been employed for contrast enhancement. From qualitative assessment, it has clearly been shown that BPDFHE was able to produce images with the best contrast. On top of that, it also produced a good quantitative assessment results as compared to other methods. Lastly, a GUI has been designed to enable the proposed techniques to be used easily.
Contributor(s):
Tan Kok Chun - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875006291
Language:
English
Subject Keywords:
technologies; Endoscopy; patients
First presented to the public:
6/1/2014
Original Publication Date:
8/26/2020
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 93
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2020-08-26 10:02:44.841
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Image enhancement of endoscopic gastritis images for feature extraction1 2020-08-26 10:02:44.841