Investigations on edge preservation and smoothening of frequency domain filters for speckle removal in breast ultrasound images
by Bikesh kumar Singh; Kesari Verma; A.S. Thoke
International Journal of Biomedical Engineering and Technology (IJBET), Vol. 20, No. 2, 2016

Abstract: Spatial domain methods for reducing speckle noise are more efficient but their performance is highly sensitive to size, shape of window and threshold used during filtering process. Hence frequency domain filtering is more appropriate if suitable and straightforward kernel or threshold cannot be found in the spatial domain. Unfortunately, they result in over smoothening, loss of edges and sharp details. In this paper various frequency domain techniques for the suppression of speckle noise are compiled and investigated. Further, they were evaluated using suitable image quality metrics and visual perception evaluation by two experts. Experiments were conducted on 37 breast ultrasound images. The results of comparative study using various image quality metrics show that wavelet filter with first level decomposition and eliminated HH band [Wav(HH/1)] outperformed others. On the other hand, homomorphic filter with wavelet filter function Wav(HH/1) outperformed others in visual assessment carried out by two experts.

Online publication date: Sat, 16-Jan-2016

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Biomedical Engineering and Technology (IJBET):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com