An Intelligent Metadata Extraction Approach Based on Programming by Demonstration

摘要

Metadata extraction is an important prerequisite for remote sensing images management and sharing. Business users urgently need to extract metadata automatically and quickly. However, existing metadata extraction applications extract only one single type of image metadata. In this paper, we proposed a generic and extensible metadata extraction approach based on Programming by Demonstration (PbD). Data owners can specify the metadata items to be extracted and the extraction methods in an interactive and visual user interface. Such knowledge will be stored in the rule base in the form of rules. The advantage of this approach is that users do not need any programming knowledge, but they can handle new types of images themselves.