No-reference video monitoring image blur metric based on local gradient structure similarity

Authors:
Shurong Chen;Huijuan Jiao
Affiliations:
College of Information Engineering, Shanghai Maritime University, Shanghai, China;College of Information Engineering, Shanghai Maritime University, Shanghai, China
Venue:
AICI'11 Proceedings of the Third international conference on Artificial intelligence and computational intelligence - Volume Part II
Year:
2011

Citing 5
Cited 0

A no-reference objective image sharpness metric based on the notion of just noticeable blur (JNB)

IEEE Transactions on Image Processing
No-Reference Estimation of the Coding PSNR for H.264-Coded Sequences

IEEE Transactions on Consumer Electronics
Automatic functional TV set failure detection system

IEEE Transactions on Consumer Electronics
Image quality assessment: from error visibility to structural similarity

IEEE Transactions on Image Processing
No-Reference Quality Assessment for Networked Video via Primary Analysis of Bit Stream

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.01

Visualization

Abstract

No-Reference (NR) quality metric for monitoring video image is a challenging and meaningful research. In this paper, we propose a novel objective NR video blurriness metric for monitoring surveillance tapes based on human visual system (HVS) characteristics and the local structure similarity of gradient images. Firstly, according to the low-pass filter (LPF) characteristics of optical imaging system, we construct a reference frame image by passing an original (to be tested) frame through a LPF; secondly, weight the gradient images of reference frame and original frame respectively with contrast sensitivity function (CSF) of HVS, followed by extracting gradient informationrich blocks in the reference gradient image and then computing the local gradient structure similarity between the corresponding blocks of the original frame image and the reference one to assess the blurriness of single frame of original sequence; finally the blur quality of overall original video sequence is evaluated by employing a weighting method which calculates the quality of each frame with different weights. Experimental results show that the proposed metric model has a good correlation with subjective scores and achieves valid blurriness evaluation effects.