FPGA based implementation of low complex adaptive speckle suppression filter for B-mode medical ultrasound images

R, Bharath and Akkala, V and P, Rajalakshmi and P, Kumar (2015) FPGA based implementation of low complex adaptive speckle suppression filter for B-mode medical ultrasound images. In: 3rd IEEE Conference on Biomedical Engineering and Sciences, IECBES 2014, 8-10 December, 2014, Kuala Lumpur; Malaysia.

Text (Author version pre-print)
edas.final-1570017085.pdf - Accepted Version

Download (280kB) | Preview


Speckles are considered as noise, which masks the fine information present in B-mode ultrasound images. Speckles appears as small snakes and dense granular like structures which has serious impact on visual perception of an image. Adaptive filter based on local statistics of an image is used to enhance the image by suppressing the noise. Adaptive speckle suppression filter enhance the image by reducing the variance between intrapixel intensities in homogeneous regions and preserving variance across interpixel intensities across the nonhomogeneous regions. In this paper, we implemented low complex adaptive speckle suppression filter on FPGA based kintex7 board. The performance of the filter is evaluated by plotting the pixel variations of original image with filtered image of an ultrasound phantom. The results show that proposed algorithm can be implemented on mobile ultrasound platforms due to 50% less computations needed per pixel compared to traditional adaptive speckle suppression algorithms, which aids better diagnosis for healthcare.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Adaptive filter; FPGA; Image enhancement; Multiplicative noise; Speckle
Subjects: Others > Electricity
Divisions: Department of Electrical Engineering
Depositing User: Team Library
Date Deposited: 07 Apr 2015 11:20
Last Modified: 26 Aug 2015 09:05
URI: http://raiith.iith.ac.in/id/eprint/1441
Publisher URL: http://dx.doi.org/10.1109/IECBES.2014.7047561
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 1441 Statistics for this ePrint Item