Title :
Illumination compensation method for unevenly lighted document segmentation
Author :
Zhiyong Ju ; Guoqing Gu
Author_Institution :
School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and technology, 200093, China
Abstract :
This paper presents a new method for binarizing unevenly lighted document images. An illumination compensation algorithm is proposed to convert the unevenly lighted document to an evenly lighted document. The compensated image is then binarized by applying an improved locally adaptive approach. The method effectively overcomes the difficulty in handling uneven lighting and enhances segmentation quality considerably.
Keywords :
Document Segmentation; Illumination Compensation; Unevenly Lighted;
Conference_Titel :
Automatic Control and Artificial Intelligence (ACAI 2012), International Conference on
Conference_Location :
Xiamen
Electronic_ISBN :
978-1-84919-537-9
DOI :
10.1049/cp.2012.1184