| Title |
Application of Convolutional Neural Network Classification Algorithm in Online-offline Blended Teaching of Chinese Painting |
| DOI |
https://doi.org/10.5573/IEIESPC.2023.12.3.243 |
| Keywords |
CNN; Chinese painting; Action recognition; Image classification; Blended teaching |
| Abstract |
In the post-epidemic era, some colleges and universities in China are still semi-closed. Some courses have adopted a mixed online and offline teaching model. This study attempts to solve the problem that Chinese painting online education is challenging to integrate into emotional education. In particular, this paper proposes an emotion-oriented hybrid-teaching model(Ed note: Compound adjectives that modify a noun are typically hyphenated but not when the first word of the adjective is an adverb ending with “-ly.”) based on an improved convolutional neural network (CNN). This mode can recognize students' movements and expressions online to judge their emotional state and improve the effectiveness of online teaching. After combining the mixed teaching mode and existing research on action recognition, the mainstream long-term and short-term memory network and attention mechanism are ineffective for emotion classification. The research first reduces the dimensionality of the input image. It then introduces an emotion-oriented weight shift module and uses the content analysis method. The experimental results showed that the teaching model proposed in the study does not improve theoretical knowledge and learning attitude significantly, but the emotional development index and emotional development quality are 110.15% and 16.62% higher, respectively, than the general method. Compared to the general teaching method, the method proposed in the research has a better effect on emotional development. |