Guqin Notation and Music Style Recognition

motivated by the Convolutional Neural Networks about digit recognition and ImageNet deep neural network by Krizhevsky et al. [1], I did this project on Guqin notation recognition, which classified reduced characters with positioned 1-10 (一 -十) in handwritten Chinese characters and translated to other music recording scores. I built a four-layer convolutional neural network using adjusted CaffeNet CNN [2] model to classify 8000 images from handwritten Guqin notations into 10 distinct classes. The network achieves a test set error rate of 15.7%. The model is trained using dropout on the fully-connected layers and performed PCA with assigned randomized variable for weight decay to reduce overfitting. Keywords—CNN; Guqin;Notation; CaffeNet; Text detection; Handwritten character recognition; Convolutional neural networks;