CG-DIQA: No-reference Document Image Quality Assessment Based on Character Gradient
This addresses the challenge of assessing document image quality without reference images, which is important for applications like OCR, but it appears incremental as it builds on existing methods like MSER.
The paper tackled the problem of no-reference document image quality assessment (DIQA) by proposing a method based on character gradient, using OCR accuracy as ground truth, and showed it outperforms state-of-the-art methods on a benchmark dataset.
Document image quality assessment (DIQA) is an important and challenging problem in real applications. In order to predict the quality scores of document images, this paper proposes a novel no-reference DIQA method based on character gradient, where the OCR accuracy is used as a ground-truth quality metric. Character gradient is computed on character patches detected with the maximally stable extremal regions (MSER) based method. Character patches are essentially significant to character recognition and therefore suitable for use in estimating document image quality. Experiments on a benchmark dataset show that the proposed method outperforms the state-of-the-art methods in estimating the quality score of document images.