Abstract
Handwriting is one of the natural biological characteristics of human beings. People have a high degree of acceptance of identity recognition technology based on handwriting. It has a wide application prospect in the fields of finance, government affairs, justice and so on. In this paper, firstly, the original document image is segmented into smaller regional samples by using the sliding window method. Secondly, multiple pre-training models are used to extract the sample features, and multiple features of the samples are fused. Finally, the euclidean distance is used to express the degree of difference between samples. Experiments show that this method has a high recognition rate and has a certain application value.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have