Content-based similar document image retrieval using fusion of CNN features

Abstract

Rapid increase of digitized document images give birth to high demand of document image retrieval. While conventional document image retrieval approaches depend on complex OCR-based text recognition and text similarity detection, this paper proposes a new content-based approach, in which more attention is paid to feature extraction and feature fusion methods. In the proposed approach, multiple features of document images are extracted by different CNN models. After that, the extracted CNN features are reduced and fused into weighted average feature. Finally, the document images are ranked based on the feature similarity to query image. Experimental procedure is performed on a group of document images that transformed from academic papers, which contain both English and Chinese document, the results show that the proposed approach has good ability to retrieve document images with similar text content, and the fusion of CNN features can improve the retrieval accuracy.