(→References) |
|||
| (5 intermediate revisions by 3 users not shown) | |||
| Line 1: | Line 1: | ||
This page and its subtopics discusses everything about Artificial Neural Networks. | This page and its subtopics discusses everything about Artificial Neural Networks. | ||
| − | Lectures discussing Artificial Neural Networks: [[Lecture | + | Lectures discussing Artificial Neural Networks: [[Lecture_13_-_Kernel_function_for_SVMs_and_ANNs_introduction_OldKiwi|Lecture 13, ECE662 Spring 2010]] and [[Lecture_14_-_ANNs%2C_Non-parametric_Density_Estimation_(Parzen_Window)_OldKiwi|Lecture 14, ECE662, Spring 2010]] |
| − | + | Relevant homework: [[Homework 2_OldKiwi]] | |
== References == | == References == | ||
| + | * [http://www.ir.iit.edu/~nazli/cs422/CS422-Slides/DM-NeuralNetwork.pdf In-depth Lecture Slides] | ||
* [http://en.wikipedia.org/wiki/Neural_network Neural Network on Wikipedia] | * [http://en.wikipedia.org/wiki/Neural_network Neural Network on Wikipedia] | ||
| − | * http://www.sciencedirect.com/science/journal/08936080 | + | * [http://www.sciencedirect.com/science/journal/08936080 Special Issue on Advances in Neural Networks Research: IJCNN’07] |
* [http://en.wikipedia.org/wiki/Artificial_neural_network Artificial Neural Network on Wikipedia] | * [http://en.wikipedia.org/wiki/Artificial_neural_network Artificial Neural Network on Wikipedia] | ||
* [http://uhaweb.hartford.edu/compsci/neural-networks-tutorial.html Neural Network Tutorial] | * [http://uhaweb.hartford.edu/compsci/neural-networks-tutorial.html Neural Network Tutorial] | ||
| Line 14: | Line 15: | ||
* [http://www.dsi.unifi.it/ANNPR/CR/23.pdf Probabilistic Neural Network] | * [http://www.dsi.unifi.it/ANNPR/CR/23.pdf Probabilistic Neural Network] | ||
* An Implementation of Neural Network: [http://www.rgu.ac.uk/files/chapter3%20-%20bp.pdf Back Propagation Algorithm] | * An Implementation of Neural Network: [http://www.rgu.ac.uk/files/chapter3%20-%20bp.pdf Back Propagation Algorithm] | ||
| + | |||
| + | |||
| + | == Application of Neural Networks to Color Calibration == | ||
| + | The below is a link to a paper which employs Neural Network in calibrating scanner. | ||
| + | * [https://ritdml.rit.edu/dspace/bitstream/1850/3035/1/PAndersonArticle04-01-1992.pdf] | ||
| + | |||
| + | Summary: The sanner calibrtion has largely two procedures, gray balancing and transfrom linear RGB data to device independent XYX data. The purpose of this paper is to improve the performance of transformation from RGB to XYZ. The traditional method to transfrom linear RGB to XYZ is to find 3x3 linear transform matrix by minimizing the perceptual error. The author argue that by using Neural Network more precise transform form linear RGB to XYZ can be achieved, as expected, since Neural Network provide more complex nonlinear transformation from input and output. He measuerd the transform error in perceptually uniform domain, and prove the strength of Neural Network in scanner calibration process. | ||
| + | |||
| + | [[Category:ECE662]] | ||
Latest revision as of 09:46, 16 April 2010
This page and its subtopics discusses everything about Artificial Neural Networks.
Lectures discussing Artificial Neural Networks: Lecture 13, ECE662 Spring 2010 and Lecture 14, ECE662, Spring 2010
Relevant homework: Homework 2_OldKiwi
References
- In-depth Lecture Slides
- Neural Network on Wikipedia
- Special Issue on Advances in Neural Networks Research: IJCNN’07
- Artificial Neural Network on Wikipedia
- Neural Network Tutorial
- MATLAB Neural Network Toolbox
- Probabilistic Neural Network
- An Implementation of Neural Network: Back Propagation Algorithm
Application of Neural Networks to Color Calibration
The below is a link to a paper which employs Neural Network in calibrating scanner.
Summary: The sanner calibrtion has largely two procedures, gray balancing and transfrom linear RGB data to device independent XYX data. The purpose of this paper is to improve the performance of transformation from RGB to XYZ. The traditional method to transfrom linear RGB to XYZ is to find 3x3 linear transform matrix by minimizing the perceptual error. The author argue that by using Neural Network more precise transform form linear RGB to XYZ can be achieved, as expected, since Neural Network provide more complex nonlinear transformation from input and output. He measuerd the transform error in perceptually uniform domain, and prove the strength of Neural Network in scanner calibration process.
