Electrical and Electronics Commons^™

A Deep Recurrent Neural Network With Iterative Optimization For Inverse Image Processing Applications, Masaki Ikuta

Theses and Dissertations

Many algorithms and methods have been proposed for inverse image processing applications, such as super-resolution, image de-noising, and image reconstruction, particularly with the recent surge of interest in machine learning and deep learning methods.

As for Computed Tomography (CT) image reconstruction, the most recently proposed methods are limited to image domain processing, where deep learning is used to learn the mapping between a true image data set and a noisy image data set in the image domain. While deep learning-based methods can produce higher quality images than conventional model-based algorithms, these methods have a limitation. Deep learning-based methods used in …

A Bibliometric Analysis On Recent Classification Techniques For Alzheimer’S Disease, Sumit Dhananjay Salunkhe, Mrinal Rahul Bachute Ph.D Guide And Associate Professor

Library Philosophy and Practice (e-journal)

Alzheimer's disease (AD) has been studied extensively to better understand the complexities of this disease and to address the numerous unanswered questions about prognosis and diagnosis. To be able to determine and allocate the resources appropriate to the research area, a detailed understanding of the research topic is much needed. Along with the tremendous expansion in the scope of neurodegenerative disease treatment research, the diversity of technologies to help the research continues to expand. Many studies have investigated into how AD affects different brain structures as the disease progresses, using various image processing methods to derive a variety of brain …

Machine Intelligence For Advanced Medical Data Analysis: Manifold Learning Approach, Fereshteh S Bashiri

Theses and Dissertations

In the current work, linear and non-linear manifold learning techniques, specifically Principle Component Analysis (PCA) and Laplacian Eigenmaps, are studied in detail. Their applications in medical image and shape analysis are investigated.

In the first contribution, a manifold learning-based multi-modal image registration technique is developed, which results in a unified intensity system through intensity transformation between the reference and sensed images. The transformation eliminates intensity variations in multi-modal medical scans and hence facilitates employing well-studied mono-modal registration techniques. The method can be used for registering multi-modal images with full and partial data.

Next, a manifold learning-based scale invariant global shape …

Per-Pixel Calibration For Rgb-Depth Natural 3d Reconstruction On Gpu, Sen Li

Theses and Dissertations--Electrical and Computer Engineering

Ever since the Kinect brought low-cost depth cameras into consumer market, great interest has been invigorated into Red-Green-Blue-Depth (RGBD) sensors. Without calibration, a RGBD camera’s horizontal and vertical field of view (FoV) could help generate 3D reconstruction in camera space naturally on graphics processing unit (GPU), which however is badly deformed by the lens distortions and imperfect depth resolution (depth distortion). The camera’s calibration based on a pinhole-camera model and a high-order distortion removal model requires a lot of calculations in the fragment shader. In order to get rid of both the lens distortion and the depth distortion …

Hyperspectral Tomographic Ftir Imaging Using Two Illumination Geometries For Polymer Phantoms, Zahrasadat Alavi

Theses and Dissertations

The purpose of this dissertation is to carry out non-destructive 3D imaging by applying Fourier Transform Infrared (FTIR) spectro-microtomographic techniques, and develop corresponding methods of data analysis. This is done by collecting 3D synchrotron-based and lab-based (Thermal) FTIR hyper spectral data at the Synchrotron Radiation Center (SRC) for the first time. Despite other 2D imaging techniques, this does not manipulate the sample, and suppresses the need to microtome 3D biological, material or biomedical samples into slices to study by spectroscopic imaging techniques. Spectro-micro-tomography is applicable for scientific, industrial, energy, biomedical samples such as stem cell characterization and materials such as …

Gaze Controlled Human-Computer Interface, Steve Petkovsek, Kevin Huang, Binay Poudel

Kevin Huang

The goal of the Gaze Controlled Human Computer Interface project is to design and construct a non-invasive gaze-tracking system that will determine where a user is looking on a computer screen in real time. To accomplish this, a fixed illumination source consisting of Infrared (IR) Light Emitting Diodes (LEDs) is used to produce corneal reflections on the user’s eyes. These reflections are captured with a video camera and compared to the relative location of the user’s pupils. From this comparison, a correlation matrix can be created and the approximate location of the screen that the user is looking at can …

Discovery And Correction Of Bias In Precision Landmark Location, Colin Foster

Theses and Dissertations

Precision Landmark Location (PLL) estimation is an integral part of 3D motion tracking. Circular landmark location estimation is one method of PLL. Current methods of estimation lead to systematic errors with a magnitude of up to .02 pixels. Estimation inaccuracies of this magnitude lead to unacceptable errors in depth measurement, the largest source of error. In the scope of this thesis, inadequacies in circular landmark location are uncovered and techniques to correct these errors are analyzed, tested, and demonstrated. Deviations in simulated images are seen to be reduced by a factor of three and the variances of real-world data were …

Gaze Controlled Human-Computer Interface, Steve Petkovsek, Kevin Huang, Binay Poudel

Senior Theses and Projects

The goal of the Gaze Controlled Human Computer Interface project is to design and construct a non-invasive gaze-tracking system that will determine where a user is looking on a computer screen in real time. To accomplish this, a fixed illumination source consisting of Infrared (IR) Light Emitting Diodes (LEDs) is used to produce corneal reflections on the user’s eyes. These reflections are captured with a video camera and compared to the relative location of the user’s pupils. From this comparison, a correlation matrix can be created and the approximate location of the screen that the user is looking at can …

A Highly Efficient Biometrics Approach For Unconstrained Iris Segmentation And Recognition, Yu Chen

FIU Electronic Theses and Dissertations

This dissertation develops an innovative approach towards less-constrained iris biometrics. Two major contributions are made in this research endeavor: (1) Designed an award-winning segmentation algorithm in the less-constrained environment where image acquisition is made of subjects on the move and taken under visible lighting conditions, and (2) Developed a pioneering iris biometrics method coupling segmentation and recognition of the iris based on video of moving persons under different acquisitions scenarios. The first part of the dissertation introduces a robust and fast segmentation approach using still images contained in the UBIRIS (version 2) noisy iris database. The results show accuracy estimated …

Filtration And Enhancement Of Environmental Characteristics Extracted From Sar Imagery Using Dynamic Kalman Technique, Ivan E. Villalon-Turrubiates

Iván Esteban Villalón Turrubiates

In this study, we propose a new computational paradigm based on the use of the Kalman filtering technique adjusted to reconstruct the dynamic behavior of the physical and electrical characteristics of different environmental monitoring data provided via reconstructive SAR imagery. As a matter of particular study we develop and report the Kalman filter-based algorithm for high-resolution filtration of the dynamic behavior of the hydrological indexes of the particular real-world SAR images of the test remotely sensed scenes. The simulation results verify the efficiency of the proposed approach.

2d And 3d Surface Image Processing Algorithms And Their Applications, Jianlin Gao

Dissertations

This doctoral dissertation work aims to develop algorithms for 2D image segmentation application of solar filament disappearance detection, 3D mesh simplification, and 3D image warping in pre-surgery simulation. Filament area detection in solar images is an image segmentation problem. A thresholding and region growing combined method is proposed and applied in this application. Based on the filament area detection results, filament disappearances are reported in real time. The solar images in 1999 are processed with this proposed system and three statistical results of filaments are presented.

3D images can be obtained by passive and active range sensing. An image registration …