Blind Quality Assessment for Slice of Microtomographic Image
| Autor: | Anton Kornilov, Ivan Yakimchuk, Ilia V. Safonov |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Smoothness
blind quality assessment Total quality management business.industry Computer science Image quality media_common.quotation_subject Distortion (optics) Contrast (statistics) Pattern recognition Sample (graphics) X-ray microtomography imaging lcsh:Telecommunication Image (mathematics) quality metrics Digital Rock lcsh:TK5101-6720 artifacts detection Quality (business) Artificial intelligence business media_common |
| Zdroj: | FRUCT Proceedings of the XXth Conference of Open Innovations Association FRUCT, Vol 854, Iss 24, Pp 170-178 (2019) |
| DOI: | 10.23919/fruct.2019.8711938 |
| Popis: | The paper considers a new algorithm for blind quality assessment of a slice of X-ray microtomographic image. We selected the following factors impacting on micro-CT image quality with respect to Digital Rock technology: smoothness, sharpness, contrast, absence of high-density regions and ring artifacts. We propose algorithms for estimation of partial quality measures for named factors inside Region-of-Interest, that is in area associated with a sample of rock or granular material. Total quality metrics is calculated as a product of these partial measures. Our method for quality assessment provides reasonable outcomes for synthetic and real slices of micro-CT images. We collected experts' judgments about quality of slices. Proposed solution has a high correlation with scores of experts and outperforms existing blind quality metrics. An application of developed method to all slices allows to obtain quality estimation for 3D micro-CT image. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|