Traditional Chinese Painting Images Research Articles

Multi-Instance Learning Algorithm Based on LSTM for Chinese Painting Image Classification

Aiming at the problem of weakly supervised learning in traditional Chinese painting image classification, a novel multi-instance learning algorithm based on Long and Short-Term Memory neural network with attention mechanism (ALSTM-MIL) is proposed. Firstly, by using the Pyramid Overlapping Grid Division (POGP), a multi-instance modeling scheme is designed to convert Chinese painting images into multi-instance bag, thereby transforming the problem of Chinese painting image classification into a MIL problem. Secondly, an efficient sequence generator is designed. It selects discriminative instances from the positive bags, construct a discriminative instance set (DIS), and convert multi-instance bags into equal-length ordered sequences. Thirdly, an LSTM network model with an attention mechanism is designed to perform semantic analysis on multi-instance bags to obtain their memory coding features, and then combined with the Softmax classifier to achieve semantic classification of traditional Chinese painting images. Experimental results on the Chinese painting (CP) image set show that the LSTM network built on the visual feature set is feasible, and the performance of the proposed MIL algorithm is also superior to other classification algorithms.

基于Gabor理论的山水画皴法分类

提取具有代表性的特征进行纹理描述和分类一直是纹理分析的热点。本文针对不同皴法的国画，运用了一种Gabor滤波器技术进行分类：通过纹理特征提取，利用几何归一化和光线归一化方法将国画图像进行预处理，再对Gabor滤波器组滤波后组成的高维特征矢量通过主成分分析(PCA)进行降维，最后采用支持向量机(SVM)方法进行纹理分类。这种分类方法的准确率可达95.5%。 Extracting the effective features for texture description and classification has always been the hot spot of the texture analysis. In this paper, according to different texture of traditional Chinese painting, we use a kind of Gabor filter technique to classify the painting. By texture feature extraction, first of all, we preprocess the traditional Chinese painting images with geometric normalization and light normalization, after that we process the group of the Gabor filter of high dimensional feature vectors by principal component analysis (PCA) for dimension reduction. Finally, support vector machine (SVM) method is employed for texture classification. The accuracy rate of this classification method can reach 95.5%.

Aging and Reverse-aging Traditional Chinese Painting Images Based on Web-Mining

Traditional Chinese painting has a long history more than thousands of year. No matter when these paintings are drawn, all suffer deterioration caused both by aging and external forces. Restoring damaged artworks involves very professional issues beyond the scope of computer science. However, retouching their images with aging/reverse-aging effects throws down a very interesting challenge to us.