TY - GEN
T1 - New optimized approach for written character recognition using symlest wavelet
AU - Munguía, R.
AU - Toscano, K.
AU - Sánchez, G.
AU - Nakano, M.
PY - 2009
Y1 - 2009
N2 - The technological changes over the time, have allowed today's society focuses on the acquisition of all types of electronic documents, which is why there is a need to implement new systems to help us in the handwriting characters recognition field, since 70's years have been made research in this area but there are still problems without a solution, especially in cursive handwriting characters recognition In recent years there have been various schemes aimed at hand written character recognition for automatic database applications creation in libraries, automatic reading checks, among others. That is why this research proposes an algorithm for cursive character recognition, which is to obtain the characteristic points of each character, which are interpolated using the Natural Spline Function. The handwriting characters recognition process is developed in inverse order using wavelet by its smoothing properties, also compare the performance system using three different classifiers: SVM (Support Vector Machines), GMM (Gaussian Mixture Model) and ANN (Artificial Neural Network).
AB - The technological changes over the time, have allowed today's society focuses on the acquisition of all types of electronic documents, which is why there is a need to implement new systems to help us in the handwriting characters recognition field, since 70's years have been made research in this area but there are still problems without a solution, especially in cursive handwriting characters recognition In recent years there have been various schemes aimed at hand written character recognition for automatic database applications creation in libraries, automatic reading checks, among others. That is why this research proposes an algorithm for cursive character recognition, which is to obtain the characteristic points of each character, which are interpolated using the Natural Spline Function. The handwriting characters recognition process is developed in inverse order using wavelet by its smoothing properties, also compare the performance system using three different classifiers: SVM (Support Vector Machines), GMM (Gaussian Mixture Model) and ANN (Artificial Neural Network).
UR - http://www.scopus.com/inward/record.url?scp=77950633501&partnerID=8YFLogxK
U2 - 10.1109/MWSCAS.2009.5235881
DO - 10.1109/MWSCAS.2009.5235881
M3 - Contribución a la conferencia
AN - SCOPUS:77950633501
SN - 9781424444793
T3 - Midwest Symposium on Circuits and Systems
SP - 766
EP - 769
BT - 2009 52nd IEEE International Midwest Symposium on Circuits and Systems, MWSCAS '09
T2 - 2009 52nd IEEE International Midwest Symposium on Circuits and Systems, MWSCAS '09
Y2 - 2 August 2009 through 5 August 2009
ER -