Automatic classification of archaeological pottery sherds

摘要: This article presents a novel technique for automatic archaeological sherd classification. Sherds that are found in the field usually have little to no visible textual information such as symbols, graphs, or marks on them. makes manual classification an extremely difficult and time-consuming task conservators archaeologists. For bunch of sherds field, expert identifies different classes indicates at least one representative each class (training sample). The proposed uses order correctly classify remaining sherds. sherd, local features based color texture extracted then transformed into global vector describes whole image, using new bag words technique. Finally, feature selection algorithm is applied locates with high discriminative power. Extensive experiments were performed verify effectiveness show very promising results.