2016/04/20 by Sounak Dey, Anguelos Nicolaou, Dey, Sounak +5
Computer Science · #Advanced Image and Video Retrieval Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Handwritten Text Recognition Techniques #Image Retrieval and Classification Techniques
paper · pdf · doi:10.48550/arxiv.1604.05907
openalex publication_date 2016/04/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Digital libraries store images which can be highly degraded and to index this\nkind of images we resort to word spot- ting as our information retrieval\nsystem. Information retrieval for handwritten document images is more\nchallenging due to the difficulties in complex layout analysis, large\nvariations of writing styles, and degradation or low quality of historical\nmanuscripts. This paper presents a simple innovative learning-free method for\nword spotting from large scale historical documents combining Local Binary\nPattern (LBP) and spatial sampling. This method offers three advantages:\nfirstly, it operates in completely learning free paradigm which is very\ndifferent from unsupervised learning methods, secondly, the computational time\nis significantly low because of the LBP features which are very fast to\ncompute, and thirdly, the method can be used in scenarios where annotations are\nnot available. Finally we compare the results of our proposed retrieval method\nwith the other methods in the literature.\n