A no-reference sharpness metric sensitive to blur and noise

Author

Zhu, Xiang ; Milanfar, Peyman

Author_Institution

Electr. Eng. Dept., Univ. of California at Santa Cruz, Santa Cruz, CA, USA

fYear

2009

fDate

29-31 July 2009

Firstpage

64

Lastpage

69

Abstract

A no-reference objective sharpness metric detecting both blur and noise is proposed in this paper. This metric is based on the local gradients of the image and does not require any edge detection. Its value drops either when the test image becomes blurred or corrupted by random noise. It can be thought of as an indicator of the signal to noise ratio of the image. Experiments using synthetic, natural, and compressed images are presented to demonstrate the effectiveness and robustness of this metric. Its statistical properties are also provided.

Keywords

data compression; gradient methods; image coding; random noise; blur detection; image compression; local gradient method; no-reference objective sharpness metric; noise detection; statistical property; Covariance matrix; Image analysis; Image edge detection; Image processing; Image quality; Matrix decomposition; Pixel; Signal to noise ratio; Singular value decomposition; Testing; Sharpness metric; blur; covariance; gradient; noise; singular value;

fLanguage

English

Publisher

ieee

Conference_Titel

Quality of Multimedia Experience, 2009. QoMEx 2009. International Workshop on

Conference_Location

San Diego, CA

Print_ISBN

978-1-4244-4370-3

Electronic_ISBN

978-1-4244-4370-3

Type

conf

DOI

10.1109/QOMEX.2009.5246976

Filename

5246976