A robust stamp detection framework on degraded documents

摘要: Detecting documents with a certain stamp instance is an effective and reliable way to retrieve associated specific source. However, this unique problem has essentially remained unaddressed. In paper, we present novel detection framework based on parameter estimation of connected edge features. Using robust basic-shape detectors, the approach for stamps analytically shaped contours, when only limited samples are available. For elliptic/circular stamps, it efficiently exploits orientation information from pairs points determine its center position area, without computing all five parameters ellipse. our approach, considered set characteristics patterns. Specifically, introduced algorithms address that often spatially overlay their background contents. These give significant advantages in accuracy computation complexity over traditional Hough transform method locating candidate ellipse regions. Experimental results real degraded demonstrated robustness retrieval large document database, which consists both printed text handwritten notes.