No-reference image quality assessment (NR-IQA) aims at computing an image quality score that best correlateswith either human perceived image quality or an objective quality measure, without any prior knowledge ofreference images. Although learning-based NR-IQA methods have achieved the best state-of-the-art results sofar, those methods perform well only on the datasets on which they were trained. The datasets usually containhomogeneous documents, whereas in reality, document images come from different sources. It is unrealistic tocollect training samples of images from every possible capturing device and every document type. Hence, weargue that a metric-based IQA method is more suitable for heterogeneous documents. We propose a NR-IQAmethod with the objective quality measure of OCR accuracy. The method combines distortion-specific qualitymetrics. The final quality score is calculated taking into account the proportions of, and the dependency amongdifferent distortions. Experimental results show that the method achieves competitive results with learning-basedNR-IQA methods on standard datasets, and performs better on heterogeneous documents.

EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Exploration

HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000026 | SxmlIndent | more

HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000026 | SxmlIndent | more

{{Explor lien
   |wiki=    Ticri/CIDE
   |area=    OcrV1
   |flux=    Main
   |étape=   Exploration
   |type=    RBID
   |clé=     Hal:hal-01319901
   |texte=   Metric-based No-reference Quality Assessment of Heterogeneous Document Images
}}