Research on single-character image classification of Tibetan ancient books based on deep learning
Zhongnan Zhao · 2022 3rd International Conference on Computer Vision, Image and Deep Learning & International Conference on Computer Engineering and Applications (CVIDL & ICCEA) · 2022
With the development of deep learning, classification neural networks have been widely used in the field of computer vision, and have achieved remarkable results in the direction of image classification. In order to realize the classification of Tibetan ancient books and characters, this paper uses LeNet, VGGNet, RestNet, and Wide-ResNet models to classify images of Tibetan ancient books and characters, and compare and analyze the experimental results. Experiments show that the above neural network models have achieved good classification results, and the Wide-ResNet model has the best classification effect on Tibetan single-character images, with an accuracy rate of about 94%. The experimental data in this paper provides a reference value for the classification research of Tibetan ancient books and images.