No-Reference Quality Assessment Using the Entropy of First Derivative of Blurred Images in HSV Color Space

No-Reference Quality Assessment Using the Entropy of First Derivative of Blurred Images in HSV Color Space Ahmed Majeed Hameed* , Moaz H. Ali International Journal of Modern Physics and Applications 1,4, 175-180
Al-Safwa University College, Department of Computer Technics, Karbala, Iraq
Abstract
Quality assessment of No-Reference (NR) images is the process of finding a novel metric via comparable results with the
results of Full-Reference (FR) metrics. Otherwise, it is the process of finding a computational model that can predict the
human perceptual system. This research paper focused on the process of NR images quality assessment using the Entropy of
First Derivative (EFD). Four color images are used as a sample in the Hue-Saturation-Value (HSV) system. The images were
distorting manually with Gaussian blur, and the quality of distorted images was measured using the Normalize Mean Square
Error (NMSE) as a FR metric. Then the EFD metric was used to assess the quality of distorted images. The results are
compared with the results of the FR to find the efficiency of the NR metric. Therefore, it can contribute that EFD metric could
be used in image quality assessment, and the HSV color space is an appropriate color space for this NR metric.
Keywords
Blurring, EFD, No Reference, HSV, Gaussian Blurring, Quality Assessment, IQA, Blurred Images
Received: June 27, 2015 / Accepted: July 10, 2015 / Published online: July 23, 2015
@ 2015 The Authors. Published by American Institute of Science. This Open Access article is under the CC BY-NC license.