Chapter 3
Speckle Noise Reduction
and Enhancement for OCT Images
Zahra Amini, Raheleh Kafieh and Hossein Rabbani
The OCT imaging produces speckle noise due to multiple forward and backward
scattered waves. The most important step of OCT preprocessing is noise reduction,
which is helpful for more sophisticated processes like segmentation. Three OCT
despeckling and enhancement methods are presented in this chapter, which are based
on statistical modeling, data adaptive and non data adaptive transform based models,
respectively.
3.1 OCT Background
Noninvasive methods for determination of anatomic information are of interest for
many years. Some familiar methods are magnetic resonance imaging (MRI), com-
puted tomography (CT scan) and ultrasound. Optical Coherence Tomography (OCT)
is recently introduced to replace the ultrasound source with infrared light source and
measure reflectance properties of tissue [1]. Each reflectance beam is called A-scan
and a row of these beams make 2 dimensional B-scans. Series of such 2D images
can be combined to produce a 3D volumetric measure of the anatomical area. Trans-
parent anatomy of the retina to OCT energy source (light), makes this tissue a nice
place to be scanned by this method. This transparency is easy to be understood since
the light beam needs to pass through the whole retina to reach photoreceptors (on
posterior surface) which have the responsibility of turning the light to electric signal
of the brain. Figure 3.1 depicts the standard OCT scheme.
Z. Amini · R. Kafieh · H. Rabbani (B )
Department of Bioelectrics & Biomedical Engineering, School of Advanced
Technologies in Medicine, Medical Image & Signal Processing Research Center, Isfahan
University of Medical Sciences, Isfahan, Iran
e-mail: hoss_rab@yahoo.com
© Science Press and Springer Nature Singapore Pte Ltd. 2019
X. Chen et al. (eds.), Retinal Optical Coherence Tomography Image Analysis,
Biological and Medical Physics, Biomedical Engineering,
https://doi.org/10.1007/978-981-13-1825-2_3
39