Semi-supervised-transfer-learning character recognition method and system based on convolutional neural network

Abstract

The invention provides a semi-supervised-transfer-learning character recognition method based on a convolutional neural network. The method includes: using batch character image samples, which are ina target domain and have no class label, as a test sample set to input the same to the convolutional neural network after semi-supervised transfer learning, and identifying character images of the test sample set, wherein the convolutional neural network after semi-supervised transfer learning is obtained by joint training of batch character image samples which are in a source domain and have class labels, batch character image samples which are in the target domain and have class labels, and batch character image samples which are in the target domain and have no class label, and recognitionprecision is improved. According to the semi-supervised-transfer-learning character recognition method based on the convolutional neural network provided and a semi-supervised-transfer-learning character recognition system based on the convolutional neural network provided by the invention, the large number of source-domain samples with the class labels, the small number of target-domain sampleswith the class labels and the relatively easily obtained target-domain samples without a class label can be utilized for semi-supervised transfer learning, and adaptation ability of the convolutionalneural network on the target-domain samples is improved.

Description

technical field [0001] The present invention relates to the technical field of information processing, and more specifically, to a semi-supervised transfer learning character recognition method and system based on a convolutional neural network. Background technique [0002] Due to the strong nonlinear description ability of deep neural network, it has achieved better results than traditional pattern recognition and machine learning methods in text recognition and other aspects. However, there are also some problems in the application of deep neural networks. First of all, due to the complex structure and many parameters of the deep neural network, a large number of samples with class labels are required for supervised learning; in addition, when using the deep neural network for specific text recognition or other tasks, it is necessary to assume the feature vectors of the test set and the training set Subject to the same statistical distribution, but in practical applicati...

AI Technical Summary

Problems solved by technology

[0004] However, the transfer learning method provided by the existing technology only uses samples with class labels in the target domain for supervised learning. It is difficult to obey the same statistical distribu

Images