Convolutional Neural Networks for Character-level Classification

Dae-Gun Ko, Su-Han Song, Ki-Min Kang, Seong-Wook Han
2017 IEIE Transactions on Smart Processing and Computing  
Optical character recognition (OCR) automatically recognizes text in an image. OCR is still a challenging problem in computer vision. A successful solution to OCR has important device applications, such as text-to-speech conversion and automatic document classification. In this work, we analyze character recognition performance using the current state-of-the-art deep-learning structures. One is the AlexNet structure, another is the LeNet structure, and the other one is the SPNet structure. For
more » ... his, we have built our own dataset that contains digits and upper-and lowercase characters. We experiment in the presence of salt-and-pepper noise or Gaussian noise, and report the performance comparison in terms of recognition error. Experimental results indicate by five-fold cross-validation that the SPNet structure (our approach) outperforms AlexNet and LeNet in recognition error.
doi:10.5573/ieiespc.2017.6.1.053 fatcat:ewuywwkmerfjrh4jd7dbtfrjvi