focus on Microscopy Microtechniques & Frequency domain (Wavelet) investigation of OCT images of skin Mohammad R. N. Avanaki1 , hashem Jafaree2 , Ali Hojjat1, and Adrian Gh. Podeleanu3
1. Neurosciences and Medical image computing (NMIC) group, school of Biosciences, University of Kent, Canterbury, Kent, United Kingdom 2. Department of Computer Science, University of Tehran, Enghelab Ave., Tehran, Iran 3. Applied optics group (AOG), school of Physical sciences, University of Kent, Canterbury, Kent, United Kingdom
In this article we investigate the use of wavelet transform on the images produced by an optical coherence tomography (OCT) to obtain further information from the OCT images transformations in different levels of decomposition with two wavelet mother functions, in frequency domain. To transform the image from the spatial domain to the frequency domain, wavelet transformation of the image was used as it was found that the images obtained from the wavelet transform include more details than those obtained from Fourier transform. The OCT system employed for imaging was an en-Face time domain OCT which uses a dynamic focus scheme (Figure 1). With dynamic focus, the coherence gate is synchronised with the confocal gate; hence, the transverse resolution is conserved throughout the depth range and an enhanced signal is returned from all depths. Therefore, higher resolution images than those of standard OCT can be obtained. This OCT system is especially designed for use in applications where a high lateral resolution and a large depth range are required. The system uses a super luminescent diode (SLD) with a central wavelength of 1300 nm and a spectral bandwidth 54 nm.
Figure 3. Application of discreet wavelet transform on the fingertip OCT image with (a) dB1 level2, (b) sym4 level2
Figure 1. Dynamic focus time domain OCT (DF-OCT) optical set-up. SLD: super luminescent laser diode, BD: balance detection photodetector unit; EI: electronic conditioning signal interface; C1,2: 2 x 2 coupler, CL1,2,3: collimator lens, MPC: mirror positioning controller, PC1,2: polarisation controller, TS: translation stage, M: microscope objective, OF: optical fibre.
We utilised two wavelet transformation approaches: discreet wavelet transform (DWT), and stationary wavelet transform (SWT). The DWT is not a time-invariant transform which means that even with periodic signal extension, the DWT of a translated version of a signal X is not, in general, the translated version of the DWT of X. To restore the translation invariance, which is a desirable property lost by the classical DWT, SWT is employed. We applied the stationary and discrete wavelet transformations in different levels of decomposition with two wavelet mother functions, Daubechies and Symlet, more precisely sym4 and dB1, to the OCT images of fingertip of a 29 years-old Asian male (skin type III). The OCT images are cross section (B-scan) images. The results of imaging are illustrated in Figure 2 and 3. Of notice, that the images are enhanced using histogram equalisation before the wavelet transformations.
We then applied Canny edge detector with the threshold value 0.4 to compare the information obtained from the components of SWT and DWT with two different wavelet mother functions. In Figure 4 and 5, the gradient images of approximation, horizontal and vertical components are demonstrated.
Figure 4. Canny edge detector with threshold 0.4, applied on the SWT components. Gradient image of (a) approximation component obtained with dB1 level2, (b) horizontal component obtained with dB1 level2, (c) vertical component obtained with dB1 level2, (e) approximation component obtained with sym4 level2, (f) horizontal component obtained with sym4 level2, (g) vertical component obtained with sym4 level2.
Figure 2. Application of stationary wavelet transform on the fingertip OCT image with (a) dB1 level2, (b) sym4 level2
Figure 5. Canny edge detector with threshold 0.4, applied on the DWT components. Gradient image of (a) approximation component obtained with dB1 level2, (b) horizontal component obtained with dB1 level2, (c) vertical component obtained with dB1 level2, (e) approximation component obtained with sym4 level2, (f) horizontal component obtained with sym4 level2, (g) vertical component obtained with sym4 level2.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56
