International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 117 - Number 22 |
Year of Publication: 2015 |
Authors: Harshshikha Nandan, Manisha Jindal, Arsh, Debanjan Konar |
10.5120/20687-3554 |
Harshshikha Nandan, Manisha Jindal, Arsh, Debanjan Konar . A Comparative Study on the Development of Binary Object Extraction System using Different Self Organizing Neural Network. International Journal of Computer Applications. 117, 22 ( May 2015), 24-29. DOI=10.5120/20687-3554
Accurately Extraction of a binary object from a noisy perspective has been a daunting task in the field of pattern recognition. Several techniques have been tried to optimally solve the problem of denoising of object over the decades. In this paper, different binary object extraction methods are reviewed which are basically guided by different Self-Organizing Neural Networks (SONN) architectures as Bi-Directional Self Organizing Neural Network (BDSONN), multi-Layer Self Organizing neural Network (MLSONN) and quantum version of MLSONN (QMLSONN). The result shows that QMLSONN outperforms over other network architectures in terms of time and also it restores shape of the object with great accuracy.