(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Development of digital image re-colouring Algorithm

Development of digital image re-colouring Algorithm / Noorfairuse Mohamad Noor
Pewarnaan semula imej atau pewarnaan adalah proses menambah warna kepada imej monokrom atau imej skala kelabu yang biasanya melibatkan peruasan imej ke kawasan-kawasan dan menjejaki kawasan ini di seluruh urutan imej. Pewarnaan telah dikembangkan dari masa ke semasa dalam usaha untuk meningkatkan prestasi dan memastikan bahawa sistem yang dibangunkan adalah mesra pengguna. Kaedah pewarnaan pertama adalah menentukan kawasan piksel secara manual dan menambah warna ke kawasan ini secara manual. Kaedah pewarnaan ini menlibatkan kos yang tinggi dan memakan masa. Kaedah lain adalah berdasarkan contoh-pewarnaan dengan memindahkan warna dari imej warna rujukan. Kawalan seni kaedah ini adalah agak tidak langsung kerana pengguna diperlukan untuk mencari imej rujukan yang mengandungi warna yang dikehendaki di rantau.Tesis ini akan membentangkan kaedah pewarnaan mudah iaitu kaedah contengan yang berdasarkan permis mudah yang jiran piksel dalam ruang masa yang mempunyai keamatan yang sama perlu mempunyai warna yang sama. Pengguna hanya perlu menganotasi imej dengan beberapa contengan warna dan warna-warna yang ditunjukkan secara automatik merebak dalam kedua-dua ruang dan masa untuk menghasilkan imej berwarna sepenuhnya. Analisis kualiti dan analisis kuantiti digunakan bagi mengenalpasti imej yang terhasil. Imej yang terhasil telah dibanding berdasarkan kualiti warna yang terhasil dan nilai PSNR. Nilai PSNR yang tinggi menunjukkan imej yang terhasil sama dengan imej yang asal. _______________________________________________________________________________________________________ Re-colouring image or colorization is the process of adding colour to monochrome image or greyscale image which it is typically involve segmentation of images into regions and tracking these regions across image sequence. Colorization have been developed from time to time in order to improve its performance and to make sure that the system developed is user friendly. First colorization method was by determined the pixels region manually and added colour to the region manually. This colorization method spends a lot of cost and time consume. Another method is examplebased colorization by transferring colour from a reference colour image. The artistic control of this method is quite indirect because the user is required to find a references image containing the desired colour over region. This thesis will present the simple colorization method which is scribble colorization method that based on a simple premise which is neighbouring pixels in space time that have similar intensities should have similar colour. The user only needs to annotate the image with a few colour scribbles and the indicated colours are automatically propagated in both space and time to produce a fully colorized image. Qualitative analysis and quantitative analysis are used to verify the result images. The result images are compared based on the quality of the colour and the value of PSNR. High value of PNSR indicated that the result image and the ground truth image are similar.
Contributor(s):
Noorfairuse Mohamad Noor - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875007705
Language:
English
Subject Keywords:
Re-colouring; colorization; greyscale image
First presented to the public:
6/1/2018
Original Publication Date:
8/10/2018
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 50
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2018-08-14 12:12:30.386
Date Last Updated
2019-01-07 11:24:32.9118
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Development of digital image re-colouring Algorithm1 2018-08-14 12:12:30.386