Improving classification performance of sonar targets by applying general regression neural network with PCA

Pattern Recognition has become an attractive research oriented field of the computer vision and machine learning for the last few decades. Neural pattern recognition techniques are also being exercised for pattern recognition, showing promising results. In this paper, a comparison is made between statistical and neural pattern recognition techniques and tried to realize how neural techniques reveal far better results than statistical techniques. In this comparison, Discriminant Analysis (DA) and Principal Component Analysis (PCA) are used for pattern recognition, which are a statistical technique. Discriminant Analysis engrosses the problem of huge data dimensions and small sample size. To evade these problems, pattern recognition task is also implemented using Generalized Regression Neural Network (GRNN) and Back-propagation Neural Network (BPNN) techniques. The task of pattern recognition is conceded on a data base of face images of 400 people. Neural networks proved results for better than statistical methods.

show abstract

“…GRNN is also very unswerving and robust. In GRNN, as the size of data increases, the error approaches towards zero [5].…”

Section: B General Regression Neural Networkmentioning

confidence: 97%

“…GRNN has highly parallel structure in its architecture. In this highly parallel structure, learning is one fold that is from input to output side [5]. GRNN performs well on noisy data than Back-propagation; FFBP Neural Network does not work accurately if available data is large enough.…”

Section: B General Regression Neural Networkmentioning

confidence: 99%

Pattern recognition using statistical and neural techniques

Ahmad

Jameel

Ahmad

2011

International Conference on Computer Networks and Information Technology

View full text Add to dashboard Cite

show abstract

“…Filter methods have a higher efficiency compared with wrapped methods. Many Filter methods have been studied, such as Principal component analysis (PCA) [5][6][7] and ReliefF method. Theoretical and empirical analysis to the ReliefF and RReliefF methods [8] are given by R.Marko and K.Igor.…”

Section: Introductionmentioning

confidence: 99%

Feature selection based on ReliefF and PCA for underwater sound classification

Zeng

Wang

Zhang

et al. 2013

Proceedings of 2013 3rd International Conference on Computer Science and Network Technology

View full text Add to dashboard Cite

The performance of underwater noise classification system is highly related to the dimensions of the features and the size of the training set. However, underwater sound signals are difficult to obtain, the training sets are always in small size and the limited information are embedded in a few feature subspace. In this paper, MFCC features are extracted firstly, and then a feature selection method based on PCA and ReliefF is presented to find the most discriminating feature subset. PCA is used to project the original feature to a new feature space by maximizing the variance matrix. ReliefF method is applied to find the proper feature subset which has the maximum score. Experimental results show that our method performs well and achieves higher recognition accuracy than that of the original features in most cases.

show abstract

“…Applications range from control [25], prediction [26][27][28], modelling [29][30][31], fault diagnosis [32], feature extraction [33,34] to engine management [35] and data analysis [36].…”

Section: Introductionmentioning

confidence: 99%