(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Development of contrast enhancement method for digital images

Development of contrast enhancement method for digital images / Yeoh Seng Cheng
Gambar yang ditangkap dalam persekitaran gelap, yang keadaan pencahayaannya tidak mencukupi atau tidak sekata, mungkin membawa kepada imej dengan beza jelas yang rendah. Imej malam kelihatan gelap dan tidak jelas berbanding dengan imej siang. Keadah peningkatan imej boleh digunakan untuk meningkatkan kualiti imej. Kaedah penyeragaman histogram (HE) adalah kaedah peningkatan imej yang biasa. Walaupun penyelidik telah mencadangkan pelbagai kaedah peningkatan imej termasuk penyeragaman histogram sejagat dan setempat, namun masih terdapat beberapa masalah yang dihadapi termasuk lebih peningkatan, peralihan kecerahan purata, dan kehilangan butiran. Oleh itu, dua kaedah peningkatan imej telah dibangunkan dengan melata penyeragaman histogram sub-imej dedahan (ESIHE) dan penyeragaman histogram penyesuaian kontras terhad (CLAHE). ESIHE adalah kaedah berdasarkan penyeragaman histogram sejagat, manakala CLAHE adalah kaedah berdasarkan penyeragaman histogram setempat. Kemudian, kedua-dua keadah yang dicadangkan dibandingkan dengan kaedah sedia ada yang berdasarkan HE secara kualitatif dan kuantitatif. Penilaian kualitatif merupakan kaji selidik penilaian visual, manakala penilaian kuantitatif adalah sisihan piawai hingar (NSD), varians imej (IV) indeks rintik (SI) dan beza jelas setiap piksel (CPP). Berdasarkan penilaian, kaedah yang digunakan ESIHE kemudian diikuti oleh CLAHE mampu meningkatkan imej lebih baik daripada kaedah digunakan CLAHE dan diikuti dengan ESIHE. Imej keluaran mempunyai penampilan semula jadi, beza jelas tinggi dan butir-butir imej adalah jelas. _______________________________________________________________________________________________________ Photos captured in the dark environments, which have insufficient or uneven lighting conditions, might lead to low contrast images. The night images are looked dark and not clear as compared to day images. Image enhancement methods can be applied to improve the image quality. Histogram equalization (HE) method is a common image enhancement method. Although researchers had proposed many enhancement methods which including global and local histogram equalization, there are still some problems faced which include over enhancement, shift of mean brightness and loss of details. Hence, two image enhancement methods were developed by cascading exposure sub-image histogram equalization (ESIHE) and contrast limited adaptive histogram equalization (CLAHE) in different sequences. ESIHE is a global histogram equalization based method, while CLAHE is a local histogram equalization based method. Then, these two proposed methods were compared with existing HE based methods qualitatively and quantitatively. The qualitative assessment is visual assessment survey, while quantitative assessments are. noise standard deviation (NSD), image variance (IV), speckle index (SI) and contrast per pixel (CPP). Based on the assessments, the method that applied ESIHE then followed by CLAHE is able to enhance images better than the method applied CLAHE first and followed by ESIHE. The output image have a natural appearance, high contrast, and the details of image are clear.
Contributor(s):
Yeoh Seng Cheng - Author
Primary Item Type:
Final Year Project
Identifiers:
Barcode : 00003107033
Accession Number : 875007155
Language:
English
Subject Keywords:
Photos captured; low contrast images; Image enhancement methods
First presented to the public:
6/1/2017
Original Publication Date:
4/19/2018
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 89
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2018-04-19 15:53:51.007
Date Last Updated
2019-01-07 11:24:32.9118
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Development of contrast enhancement method for digital images1 2018-04-19 15:53:51.007