Thangka Religious Tools Classification and Detection Based on HOG+SVM

As a non-material cultural heritage of China, Thangka draws more and more attention. The religious tool is an important symbol for verifying the identity of Buddha. It is usually used to find the main Buddha in a Thangka image, and then judge Buddhas duty, vow, and way of spiritual practice. To build an automatic Thangka religious tools detector, Histogram of Oriented Gradient (HOG) and Support Vector Machine (SVM) are combined as the feature extractor and the classifier respectively. A large number of Thangka religious tools images are manually labeled as the positive samples. Thangka images without religious tools are used as negative samples. The final experimental results are presented through the system interface. Many factors that influence the final experimental results are considered for the best performance of Thangka religious tools detection.