Faulty Object Detection by Image Processing Based on Modified Back-Propagation Algorithm of Artificial Neural Network
The conventional algorithms such as canny edge detection require so much computation which makes it more time consuming. Hence it is not used for industrial applications such as damage detection because each product should be analyzed in a fraction of seconds so that the manufacturing rate of the industry should not be affected. Hence Artificial Neural Network can be used for edge detection. Though it requires a lot of time for training, it is very fast at simulation, i.e., at function time which fulfills industry requirement for such application.
Cite this Article
Samadhan Jogdand, U.L. Bombale. Faulty Object Detection by Image Processing Based on Modified Back-propagation Algorithm of Artificial Neural Network. Journal of Image Processing & Pattern Recognition Progress. 2016; 3(2): 19–25p.
Canny J. A Computational Approach to Edge Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1986; 8(6): 679–98p.
Freund Y, Schapire RE. Large margin classification using the perceptron algorithm. Machine Learning. 1999; 37(3): 277–96p.
Bastain MR. Neural Networks and Natural Gradient. Ph.D Thesis. Logan, Utah: Utah State University; 2009.
Yao X. Evolving Artificial Neural Networks. Proceedings of the IEEE. 1999; 87(9): 1423–47p.
Werbos PJ. Back-propagation: Past and Future. Proceeding of International Conference on Neural Networks; 1988 Jul 24–27; San Diego, CA. 343–54p.
Russell I. The Delta Rule. USA: University of Hartford; 2012 (Retrieved on 2012 Nov 5).
Snyman JA. Practical Mathematical Optimization: An Introduction to Basic Optimization Theory and Classical and New Gradient-Based Algorithms. Berlin, Germany: Springer Publishing; 2005.
pdf Cross-platform C++ development using Qt.
- There are currently no refbacks.