- -

QMRNet: Quality Metric Regression for EO Image Quality Assessment and Super-Resolution

RiuNet: Repositorio Institucional de la Universidad Politécnica de Valencia

Compartir/Enviar a

Citas

Estadísticas

  • Estadisticas de Uso

QMRNet: Quality Metric Regression for EO Image Quality Assessment and Super-Resolution

Mostrar el registro sencillo del ítem

Ficheros en el ítem

dc.contributor.author Berga, David es_ES
dc.contributor.author Gallés, Pau es_ES
dc.contributor.author Takáts, Katalin es_ES
dc.contributor.author Mohedano, Eva es_ES
dc.contributor.author Riordan-Chen, Laura es_ES
dc.contributor.author García-Moll, Clara es_ES
dc.contributor.author Vilaseca, David es_ES
dc.contributor.author Marín, Javier es_ES
dc.date.accessioned 2024-04-11T10:00:40Z
dc.date.available 2024-04-11T10:00:40Z
dc.date.issued 2023-05-06 es_ES
dc.identifier.issn 2072-4292 es_ES
dc.identifier.uri http://hdl.handle.net/10251/203371
dc.description.abstract [EN] The latest advances in super-resolution have been tested with general-purpose images such as faces, landscapes and objects, but mainly unused for the task of super-resolving earth observation images. In this research paper, we benchmark state-of-the-art SR algorithms for distinct EO datasets using both full-reference and no-reference image quality assessment metrics. We also propose a novel Quality Metric Regression Network (QMRNet) that is able to predict the quality (as a no-reference metric) by training on any property of the image (e.g., its resolution, its distortions, etc.) and also able to optimize SR algorithms for a specific metric objective. This work is part of the implementation of the framework IQUAFLOW, which has been developed for the evaluation of image quality and the detection and classification of objects as well as image compression in EO use cases. We integrated our experimentation and tested our QMRNet algorithm on predicting features such as blur, sharpness, snr, rer and ground sampling distance and obtained validation medRs below 1.0 (out of N = 50) and recall rates above 95%. The overall benchmark shows promising results for LIIF, CAR and MSRN and also the potential use of QMRNet as a loss for optimizing SR predictions. Due to its simplicity, QMRNet could also be used for other use cases and image domains, as its architecture and data processing is fully scalable. es_ES
dc.description.sponsorship The project was financed by the Ministry of Science and Innovation (MICINN) and by the European Union within the framework of FEDER RETOS-Collaboration of the State Program of Research (RTC2019-007434-7), Development and Innovation Oriented to the Challenges of Society, within the State Research Plan Scientific and Technical and Innovation 2017¿2020, with the main objective of promoting technological development, innovation and quality research. es_ES
dc.language Inglés es_ES
dc.publisher MDPI AG es_ES
dc.relation.ispartof Remote Sensing es_ES
dc.rights Reconocimiento (by) es_ES
dc.subject Super-resolution es_ES
dc.subject Quality assessment es_ES
dc.subject Benchmark es_ES
dc.subject Denoising es_ES
dc.subject Regression es_ES
dc.subject Autoencoder networks es_ES
dc.subject Generative adversarial networks es_ES
dc.subject Self-supervision es_ES
dc.subject Regularization es_ES
dc.subject Optimization es_ES
dc.subject Earth observation es_ES
dc.title QMRNet: Quality Metric Regression for EO Image Quality Assessment and Super-Resolution es_ES
dc.type Artículo es_ES
dc.identifier.doi 10.3390/rs15092451 es_ES
dc.relation.projectID info:eu-repo/grantAgreement/MICINN//RTC2019-007434-7/ es_ES
dc.rights.accessRights Abierto es_ES
dc.description.bibliographicCitation Berga, D.; Gallés, P.; Takáts, K.; Mohedano, E.; Riordan-Chen, L.; García-Moll, C.; Vilaseca, D.... (2023). QMRNet: Quality Metric Regression for EO Image Quality Assessment and Super-Resolution. Remote Sensing. 15(9). https://doi.org/10.3390/rs15092451 es_ES
dc.description.accrualMethod S es_ES
dc.relation.publisherversion https://doi.org/10.3390/rs15092451 es_ES
dc.type.version info:eu-repo/semantics/publishedVersion es_ES
dc.description.volume 15 es_ES
dc.description.issue 9 es_ES
dc.relation.pasarela S\504599 es_ES
dc.contributor.funder Ministerio de Ciencia e Innovación es_ES


Este ítem aparece en la(s) siguiente(s) colección(ones)

Mostrar el registro sencillo del ítem