Comparison of Neural Network Training Functions for Hematoma Classification in Brain CT Images
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2014, Vol 16, Issue 1
Abstract
Classification is one of the most important task in application areas of artificial neural networks (ANN).Training neural networks is a complex task in the supervised learning field of research. The main difficulty in adopting ANN is to find the most appropriate combination of learning, transfer and training function for the classification task. We compared the performances of three types of training algorithms in feed forward neural network for brain hematoma classification. In this work we have selected Gradient Descent based backpropagation, Gradient Descent with momentum, Resilence backpropogation algorithms. Under conjugate based algorithms, Scaled Conjugate back propagation, Conjugate Gradient backpropagation with Polak-Riebreupdates(CGP) and Conjugate Gradient backpropagation with Fletcher-Reeves updates (CGF).The last category is Quasi Newton based algorithm, under this BFGS, Levenberg-Marquardt algorithms are selected. Proposed work compared training algorithm on the basis of mean square error, accuracy, rate of convergence and correctness of the classification. Our conclusion about the training functions is based on the simulation results
Authors and Affiliations
Bhavna Sharma
Keywords
Related Articles
- EP ID EP126168
- DOI -
- Views 64
- Downloads 0
Virtual Mouse: Computer Vision Aided Pointing Device
Abstract: This paper puts forth the idea for an easy to use device to replace the conventional mouse. One such approach is a virtual pointing device. A web-camera based solution takes advantage of computer vision technol...
Comparing Ethernet and Soft RoCE for MPI Communication
Abstract: RDMA has pleasant consideration in the late 1990 when the Virtual Interface Architecture was introduced. This growth has accelerated with the introduction of Open Fabrics Alliance’s (OFA’s) Verb Interface...
Password-oriented Image Encryption with multiple dependent factors.
Abstract: The goal of this project was to develop a password-based image encryption algorithm that would be virtually safe from brute-force attacks, resulting in an image that would have no recognizable pattern. Th...
An Analysis of students’ performance using classification algorithms
In recent years, the analysis and evaluation of students‟ performance and retaining the standard of education is a very important problem in all the educational institutions. The most important goal of the paper...
Quality Enhancement of Video Denoising Algorithm withCombined Effect of Different Views
Abstract: An unwanted factor in digital images and videos called as Noise is responsible for hiding the detailsand destroying image information. Hence denoising has great importance in the detail restoration and to...