Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
- Discipline
- Institution
- Publication
- Publication Type
Articles 1 - 6 of 6
Full-Text Articles in Physical Sciences and Mathematics
A Neuromorphic Machine Learning Framework Based On The Growth Transform Dynamical System, Ahana Gangopadhyay
A Neuromorphic Machine Learning Framework Based On The Growth Transform Dynamical System, Ahana Gangopadhyay
McKelvey School of Engineering Theses & Dissertations
As computation increasingly moves from the cloud to the source of data collection, there is a growing demand for specialized machine learning algorithms that can perform learning and inference at the edge in energy and resource-constrained environments. In this regard, we can take inspiration from small biological systems like insect brains that exhibit high energy-efficiency within a small form-factor, and show superior cognitive performance using fewer, coarser neural operations (action potentials or spikes) than the high-precision floating-point operations used in deep learning platforms. Attempts at bridging this gap using neuromorphic hardware has produced silicon brains that are orders of magnitude …
Handwriting Recognition Through Distance-Based Morphing And Energy Minimization, Dr. William Barrett
Handwriting Recognition Through Distance-Based Morphing And Energy Minimization, Dr. William Barrett
Journal of Undergraduate Research
During the past year, I have had the opportunity to mentor two undergraduate students as we performed research for improving technologies used for family history. The specific projects each student worked on, the outcomes of the projects, and the mentoring are described below.
The Multimodal Brain Tumor Image Segmentation Benchmark (Brats), Bjoern H. Menze, Andras Jakab, Stefan Bauer, Jayashree Kalpathy-Cramer, Khan M. Iftekharuddin, Syed M.S. Reza
The Multimodal Brain Tumor Image Segmentation Benchmark (Brats), Bjoern H. Menze, Andras Jakab, Stefan Bauer, Jayashree Kalpathy-Cramer, Khan M. Iftekharuddin, Syed M.S. Reza
Electrical & Computer Engineering Faculty Publications
In this paper we report the set-up and results of the Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) organized in conjunction with the MICCAI 2012 and 2013 conferences. Twenty state-of-the-art tumor segmentation algorithms were applied to a set of 65 multi-contrast MR scans of low-and high-grade glioma patients-manually annotated by up to four raters-and to 65 comparable scans generated using tumor image simulation software. Quantitative evaluations revealed considerable disagreement between the human raters in segmenting various tumor sub-regions (Dice scores in the range 74%-85%), illustrating the difficulty of this task. We found that different algorithms worked best for different sub-regions …
Automatic Multi-Model Fitting For Blood Vessel Extraction, Xuefeng Chang
Automatic Multi-Model Fitting For Blood Vessel Extraction, Xuefeng Chang
Electronic Thesis and Dissertation Repository
Blood vessel extraction and visualization in 2D images or 3D volumes is an essential clinical task. A blood vessel system is an example of a tubular tree like structure, and fully automated reconstruction of tubular tree like structures remains an open computer vision problem. Most vessel extraction methods are based on the vesselness measure. A vesselness measure, usually based on the eigenvalues of the Hessian matrix, assigns a high value to a voxel that is likely to be a part of a blood vessel. After the vesselness measure is computed, most methods extract vessels based on the shortest paths connecting …
Advances In Graph-Cut Optimization: Multi-Surface Models, Label Costs, And Hierarchical Costs, Andrew T. Delong
Advances In Graph-Cut Optimization: Multi-Surface Models, Label Costs, And Hierarchical Costs, Andrew T. Delong
Electronic Thesis and Dissertation Repository
Computer vision is full of problems that are elegantly expressed in terms of mathematical optimization, or energy minimization. This is particularly true of "low-level" inference problems such as cleaning up noisy signals, clustering and classifying data, or estimating 3D points from images. Energies let us state each problem as a clear, precise objective function. Minimizing the correct energy would, hypothetically, yield a good solution to the corresponding problem. Unfortunately, even for low-level problems we are confronted by energies that are computationally hard—often NP-hard—to minimize. As a consequence, a rather large portion of computer vision research is dedicated to proposing …
Parameter Optimization In The Regularized Shannon's Kernels Of Higher-Order Discrete Singular Convolutions, Wei Xiong, Yibao Zhao, Yun Gu
Parameter Optimization In The Regularized Shannon's Kernels Of Higher-Order Discrete Singular Convolutions, Wei Xiong, Yibao Zhao, Yun Gu
Research Collection Lee Kong Chian School Of Business
The δ-type discrete singular convolution (DSC) algorithm has recently been proposed and applied to solve kinds of partial differential equations (PDEs). With appropriate parameters, particularly the key parameter r in its regularized Shannon's kernel, the DSC algorithm can be more accurate than the pseudospectral method. However, it was previously selected empirically or under constrained inequalities without optimization. In this paper, we present a new energy-minimization method to optimize r for higher-order DSC algorithms. Objective functions are proposed for the DSC algorithm for numerical differentiators of any differential order with any discrete convolution width. Typical optimal parameters are also shown. The …