(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Blur measurement in digital image

Blur measurement in digital image / Nurhamizah Kamaruddin
Kewujudan kabur di dalam gambar digital telah menyebabkan kesulitan dalam bidang pemprosesan gambar terutamanya apabila ada keperluan untuk pengumpulan maklumat di dalam gambar yang dipersembahkan. Kehadiran kabur sentiasa mengundang kepada masalah seperti sukar untuk mengenal pasti subjek di dalam gambar. Masalah ini boleh menyebabkan kurangnya keperincian di dalam maklumat sesuatu gambar untuk penyediaan bukti terutamanya jika digunakan di dalam kes penyiasatan yang sulit. Oleh itu, adalah penting untuk menyelesaikan masalah ini melalui pembinaan algoritma pengukuran kabur di dalam gambar digital. Di dalam projek ini, kerja pengukuran kabur di dalam gambar digital dijalankan menggunakan metrik lebar kabur pinggir dan analisis dua-lulus pinggir. Di dalam metrik lebar kabur pinggir, gambar digunakan dengan penapis pinggir untuk mencari lokasi pinggir. Daripada lokasi pinggir, lebar pinggir dikira untuk mencari purata lebar pinggir. Analisa dua-lulus pinggir mempunyai proses yang hampir sama. Akan tetapi, kaedah ini memerlukan gambar masukan dikenakan dengan penapis kabur. Kedua-dua gambar asli dan gambar versi kabur melalui proses yang sama menggunakan metrik lebar kabur pinggir untuk mencari peningkatan lebar pinggir. Hasil peningkatan lebar pinggir dikira jurus ke arah pengiraan purata lebar pinggir. Kedua-dua keputusan metrik lebar kabur pinggir dan analisis dua-lulus pinggir dibandingkan untuk memilih kaedah yang terbaik untuk pengukuran kabur di dalam gambar digital. Analisis dua-lulus pinggir mempunyai keputusan purata lebar kabur yang lebih baik di mana ia mengikut hipothesis bahawa pinggir lebih tersebar apabila gambar kabur. _______________________________________________________________________________________________________ The existence of blur in digital image is quite an inconvenient in the image processing field especially whenever there is a need to gather the information in the image presented. The presence of blur often leads to numerous difficulties such as hard to recognize the subject in the image. These problems can cause the lack of information to provide evidence especially if it is used for crime investigation. Therefore, it is important to address this problem through the development of blur measurement in digital images algorithm. In this project, the task blur measurement in digital image is done by edge sharpness analysis where the average edge width of the digital image is calculated by using an edge blur width metric and two-pass edge analysis metric. In edge blur width metric, the image is applied with edge filters to find the edge location. From the edge location, the edge width is calculated to find the average edge width. Two-pass edge analysis have almost the same process. However, this method require the input image to be applied with a blur filter. Both original and the blurred version of the image, went through the same process using the edge blur width metric to find the edge width increment. The difference between both result is calculated leading to the average edge width calculation. Both edge blur width metric and two-pass edge analysis results are compared to choose the best method proposed for blur measurement in digital image. Two-pass edge analysis has a better result of average edge width where it follows the proposed hypothesis where the edge spread more when the image is blur.
Contributor(s):
Nurhamizah Kamaruddin - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875008601
Language:
English
Subject Keywords:
blur; image; sharpness
First presented to the public:
6/1/2019
Original Publication Date:
2/26/2020
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 70
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2020-02-26 17:56:31.929
Date Last Updated
2020-12-02 12:04:21.694
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Blur measurement in digital image1 2020-02-26 17:56:31.929