Ahmadreza Baghaie

Sparse and low rank decomposition based batch image alignment for speckle reduction of retinal OCT images

Optical Coherence Tomography (OCT) is an emerging technique in the field of biomedical imaging, with applications in ophthalmology, dermatology, coronary imaging etc. Due to the underlying physics, OCT images usually suffer from a granular pattern, called speckle noise, which restricts the process of interpretation. Here, a sparse and low rank decomposition based method is used for speckle reduction in retinal OCT images. This technique works on input data that consists of several B-scans of the same location. The next step is the batch alignment of the images using a sparse and low-rank decomposition based technique. Finally the denoised image is created by median filtering of the low-rank component of the processed data. Simultaneous decomposition and alignment of the images result in better performance in comparison to simple registration-based methods that are used in the literature for noise reduction of OCT images.

Structure tensor based image interpolation method

Abstract Feature preserving image interpolation is an active area in image processing field. In this paper a new direct edge directed image super-resolution algorithm based on structure tensors is proposed. Using an isotropic Gaussian filter, the structure tensor at each pixel of the input image is computed and the pixels are classified to three distinct classes; uniform region, corners and edges, according to the eigenvalues of the structure tensor. Due to application of the isotropic Gaussian filter, the classification is robust to noise presented in image. Based on the tangent eigenvector of the structure tensor, the edge direction is determined and used for interpolation along the edges. In comparison to some previous edge directed image interpolation methods, the proposed method achieves higher quality in both subjective and objective aspects. Also the proposed method outperforms previous methods in case of noisy and JPEG compressed images. Furthermore, without the need for optimization in the process, the algorithm can achieve higher speed.

3DSEM++: Adaptive and intelligent 3D SEM surface reconstruction.

Structural analysis of microscopic objects is a longstanding topic in several scientific disciplines, such as biological, mechanical, and materials sciences. The scanning electron microscope (SEM), as a promising imaging equipment has been around for decades to determine the surface properties (e.g., compositions or geometries) of specimens by achieving increased magnification, contrast, and resolution greater than one nanometer. Whereas SEM micrographs still remain two-dimensional (2D), many research and educational questions truly require knowledge and facts about their three-dimensional (3D) structures. 3D surface reconstruction from SEM images leads to remarkable understanding of microscopic surfaces, allowing informative and qualitative visualization of the samples being investigated. In this contribution, we integrate several computational technologies including machine learning, contrario methodology, and epipolar geometry to design and develop a novel and efficient method called 3DSEM++ for multi-view 3D SEM surface reconstruction in an adaptive and intelligent fashion. The experiments which have been performed on real and synthetic data assert the approach is able to reach a significant precision to both SEM extrinsic calibration and its 3D surface modeling.

State-of-the-art in retinal optical coherence tomography image analysis.

Optical coherence tomography (OCT) is an emerging imaging modality that has been widely used in the field of biomedical imaging. In the recent past, it has found uses as a diagnostic tool in dermatology, cardiology, and ophthalmology. In this paper we focus on its applications in the field of ophthalmology and retinal imaging. OCT is able to non-invasively produce cross-sectional volumetric images of the tissues which can be used for analysis of tissue structure and properties. Due to the underlying physics, OCT images suffer from a granular pattern, called speckle noise, which restricts the process of interpretation. This requires specialized noise reduction techniques to eliminate the noise while preserving image details. Another major step in OCT image analysis involves the use of segmentation techniques for distinguishing between different structures, especially in retinal OCT volumes. The outcome of this step is usually thickness maps of different retinal layers which are very useful in study of normal/diseased subjects. Lastly, movements of the tissue under imaging as well as the progression of disease in the tissue affect the quality and the proper interpretation of the acquired images which require the use of different image registration techniques. This paper reviews various techniques that are currently used to process raw image data into a form that can be clearly interpreted by clinicians.

Curvature-Based Registration for Slice Interpolation of Medical Images

Slice interpolation is a fast growing field in medical image processing. Intensity-based interpolation and object-based interpolation are two major groups of methods in the literature. In this paper an object based method for slice interpolation using a modified version of curvature registration is proposed. Due to non-linear nature of image registration the results of forward and backward registration can be different. Therefore assuming a linear displacement between corresponding pixels of reference and moving image, a functional is minimized and the displacement fields for both reference and moving images with respect to the missing in-between slice are computed and used for reconstruction of the missing slice. The proposed approach is evaluated quantitatively by using the Mean Squared Difference (MSD) as metric. The produced results show significant visual improvement in preserving sharp edges in images.

Fast Mesh-Based Medical Image Registration

In this paper a fast triangular mesh based registration method is proposed. Having Template and Reference images as inputs, the template image is triangulated using a content adaptive mesh generation algorithm. Considering the pixel values at mesh nodes, interpolated using spline interpolation method for both of the images, the energy functional needed for image registration is minimized. The minimization process was achieved using a mesh based discretization of the distance measure and regularization term which resulted in a sparse system of linear equations, which due to the smaller size in comparison to the pixel-wise registration method, can be solved directly. Mean Squared Difference (MSD) is used as a metric for evaluating the results. Using the mesh based technique, higher speed was achieved compared to pixel-based curvature registration technique with fast DCT solver. The implementation was done in MATLAB without any specific optimization. Higher speeds can be achieved using C/C++ implementations.

Dense Correspondence and Optical Flow Estimation Using Gabor, Schmid and Steerable Descriptors

In this paper, the use of three dense descriptors, namely Schmid, Gabor and steerable descriptors, is introduced and investigated for optical flow estimation and dense correspondence of different scenes and compared with the well-known dense SIFT/SIFTFlow. Several examples of optical flow estimation and dense correspondence across scenes with high variations in the intensity levels, difference in the presence of features and different misalignment models (rigid, deformable, homography etc.) are studied and the results are quantitatively/qualitatively compared with dense SIFT/SIFTFlow. The proposed dense descriptors provide comparable or better results than dense SIFT/SIFTFlow which shows the high potential in this area for more thorough investigations.

Dense Descriptors for Optical Flow Estimation: A Comparative Study

Estimating the displacements of intensity patterns between sequential frames is a very well-studied problem, which is usually referred to as optical flow estimation. The first assumption among many of the methods in the field is the brightness constancy during movements of pixels between frames. This assumption is proven to be not true in general, and therefore, the use of photometric invariant constraints has been studied in the past. One other solution can be sought by use of structural descriptors rather than pixels for estimating the optical flow. Unlike sparse feature detection/description techniques and since the problem of optical flow estimation tries to find a dense flow field, a dense structural representation of individual pixels and their neighbors is computed and then used for matching and optical flow estimation. Here, a comparative study is carried out by extending the framework of SIFT-flow to include more dense descriptors, and comprehensive comparisons are given. Overall, the work can be considered as a baseline for stimulating more interest in the use of dense descriptors for optical flow estimation.

A comparative study on the application of SIFT, SURF, BRIEF and ORB for 3D surface reconstruction of electron microscopy images

Image feature detector and descriptor algorithms have made a big advance in almost every area of computer vision applications including object localisation, object tracking, mobile robot mapping, watermarking, panorama stitching and 3D surface reconstruction by assisting the detection and description of feature points in a set of given images. In this paper, we evaluate the performance of four robust feature detection algorithms namely SIFT, SURF, BRIEF and ORB on multi-view 3D surface reconstruction of microscopic samples obtained by a scanning electron microscope (SEM), a widely used equipment in biological and materials sciences for determining the surface attributes of micro objects. To this end, we first develop an optimised multi-view framework for SEM extrinsic calibration and its 3D surface reconstruction. We design a Differential Evolutionary-based algorithm to solve the problem in a global optimisation platform. Several qualitative and quantitative comparisons such as reliability on SEM extrinsic calibration and validity on 3D visualisation performed on real microscopic objects as well as a synthetic model. The present evaluation is expected to provide better insights and consideration to determine which algorithm is well deserved for multi-view 3D SEM surface reconstruction.

Merging computational fluid dynamics and 4D Flow MRI using proper orthogonal decomposition and ridge regression

Time resolved phase-contrast magnetic resonance imaging 4D-PCMR (also called 4D Flow MRI) data while capable of non-invasively measuring blood velocities, can be affected by acquisition noise, flow artifacts, and resolution limits. In this paper, we present a novel method for merging 4D Flow MRI with computational fluid dynamics (CFD) to address these limitations and to reconstruct de-noised, divergence-free high-resolution flow-fields. Proper orthogonal decomposition (POD) is used to construct the orthonormal basis of the local sampling of the space of all possible solutions to the flow equations both at the low-resolution level of the 4D Flow MRI grid and the high-level resolution of the CFD mesh. Low-resolution, de-noised flow is obtained by projecting in vivo 4D Flow MRI data onto the low-resolution basis vectors. Ridge regression is then used to reconstruct high-resolution de-noised divergence-free solution. The effects of 4D Flow MRI grid resolution, and noise levels on the resulting velocity fields are further investigated. A numerical phantom of the flow through a cerebral aneurysm was used to compare the results obtained using the POD method with those obtained with the state-of-the-art de-noising methods. At the 4D Flow MRI grid resolution, the POD method was shown to preserve the small flow structures better than the other methods, while eliminating noise. Furthermore, the method was shown to successfully reconstruct details at the CFD mesh resolution not discernible at the 4D Flow MRI grid resolution. This method will improve the accuracy of the clinically relevant flow-derived parameters, such as pressure gradients and wall shear stresses, computed from in vivo 4D Flow MRI data.