书法
汉字
性格(数学)
计算机科学
人工智能
字符识别
风格(视觉艺术)
中国文化
特征提取
模式识别(心理学)
图像(数学)
自然语言处理
绘画
中国
数学
艺术
文学类
视觉艺术
历史
考古
几何学
标识
DOI:10.1109/aeeca52519.2021.9574199
摘要
Chinese characters and calligraphy are important components of Chinese traditional culture, which constitute rich and colorful Chinese history and culture. When users browse and appreciate books with Chinese characters, especially when browsing and appreciating ancient books with many traditional Chinese characters, they will encounter many unfamiliar Chinese characters, which brings obstacles to users' appreciation of works. In view of the above problems, this paper attempts to combine deep learning with traditional recognition methods to recognize calligraphy images. From the perspective of calligraphy style diversity, the deep convolution neural network is used to classify the calligraphy image in style, and then the traditional classification algorithm is used to identify the calligraphy image and get its corresponding Chinese characters. Through experimental comparison, the calligraphy character recognition algorithm proposed in this paper is more efficient. The Chinese character recognition algorithm is applied to the calligraphy scoring system to realize the interface of the Chinese character scoring system. Taking the scoring of three different fonts as an example, the calligraphy Chinese character scoring of the practitioners of three fonts is completed.
科研通智能强力驱动
Strongly Powered by AbleSci AI