Fan Zhu — Page 4 | Search arXiv

arXiv2018

CG-DIQA: No-reference Document Image Quality Assessment Based on Character Gradient

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.

cs.CV↗

arXiv2013

Ultrastrong and Ultrastable Metallic Glass

The lack of thermal stability, originating from their metastable nature, has been one of the paramount obstacles that hinder the wide range of applications of metallic glasses. We report that the stability of a metallic glass can be dramatically improved by slow deposition at high temperatures. The glass transition and crystallization temperatures of the ultrastable metallic glass can be increased by 51 K and 203 K, respectively, from its ordinary glass state. The ultrastable metallic glass also shows ultrahigh strength and hardness, over 30 % higher than its ordinary counterpart. Atomic structure characterization reveals that the exceptional properties of the ultrastable glass are associated with abundance of medium range order. The finding of the ultrastable metallic glass sheds light on atomic mechanisms of metallic glass formation and has important impact on the technological applications of metallic glasses.

cond-mat.mtrl-sci↗