Research on Artificial Intelligence Machine Learning Character Recognition Algorithm Based on Feature Fusion

Abstract With the continuous progress of science and information technology, people begin to study in field intelligence, machine learning is one key contents. At present, human beings have made some intelligent robot, speech recognition network search. The method character based on great significance technology. In this paper, an improved CRNN algorithm feature fusion proposed, which combines Gabor features Zernike moment into a new vector, then uses generalized K-L transform compress dimension remove redundant information. After testing, accuracy training data set test as high 0.99, shows that neural model can perfectly fit Chinese recognition.