Subspace-Based Striping Noise Reduction in Hyperspectral Images

ABSTRACT

In this paper, a new algorithm for striping noise reduction in hyperspectral images is proposed. The new algorithm exploits the orthogonal subspace approach to estimate the striping component and to remove it from the image, preserving the useful signal. The algorithm does not introduce artifacts in the data and also takes into account the dependence on the signal intensity of the striping component. The effectiveness of the algorithm in reducing striping noise is experimentally demonstrated on real data acquired both by airborne and satellite hyperspectral sensors.

Existing System

The existing system available for fuzzy filters for noise reduction deals with fat-tailed noise like impulse noise and median filter.

Ø      Only Impulse noise reduction using fuzzy filters

Ø      Gaussian noise is not specially concentrated

Ø      It does not distinguish local variation due to noise and due to image structure.

Proposed System

The proposed system presents a new technique for filtering narrow-tailed and medium narrow-tailed noise by a fuzzy filter. The system,

Ø      First estimates a “fuzzy derivative” in order to be less sensitive to local variations due to image structures such as edges

Ø      Second, the membership functions are adapted accordingly to the noise level to perform “fuzzy smoothing.”

Ø      For each pixel that is processed, the first stage computes a fuzzy derivative. Second, a set of 16 fuzzy rules is fired to determine a correction term. These rules make use of the fuzzy derivative as input.

Ø      Fuzzy sets are employed to represent the properties, while the membership functions for and are fixed, the membership function for is adapted after each iteration.

 

HARDWARE REQUIREMENTS

•                     SYSTEM                     : Pentium IV 2.4 GHz

•                     HARD DISK               : 40 GB

•                     MONITOR                  : 15 VGA colour

•                     MOUSE                      : Logitech.

•                     RAM                           : 256 MB

•                     KEYBOARD               : 110 keys enhanced.

SOFTWARE REQUIREMENTS

•                     Operating system          :           Windows XP Professional

•                     Front End                     :           JAVA.

•                     Tool                             :           Eclipse 3.3

MODULES USED

·        Pre Processing

·        Member function

·        Fuzzy Smoothing

·        Get Clear Gray Image

MODULE DESCRIPTION

Pre Processing

First estimates a “fuzzy derivative” in order to be less sensitive to local variations due to image structures such as edges

Second, the membership functions are adapted accordingly to the noise level to perform “fuzzy smoothing.”

Member function

For each pixel that is processed, the first stage computes a fuzzy derivative. Second, a set of fuzzy rules is fired to determine a correction term. These rules make use of the fuzzy derivative as input.

Fuzzy sets are employed to represent the properties, and while the membership functions for and is fixed, the membership function for are adapted after each iteration.

Fuzzy Smoothing

Set the calculated member function value from processing of gray scale Image to the negative pixel area

Get Clear Gray Image

To view the clear image by user this very particular module is used.

REFERENCE:

N.Acito, M.Diani, and G.Corsini, “Subspace-based striping noise reduction in hyperspectral images”, IEEE Transactions on Geoscience and Remote Sensing, Vol.49, April 2011.

Subspace-Based Striping Noise Reduction in Hyper Spectral Images