Fixed-Quality Compression of Remote Sensing Images With Neural Networks

Fixed-quality image compression is a coding paradigm where the tolerated introduced distortion is set by the user. This article proposes a novel fixed-quality compression method for remote sensing images. It is based on a neural architecture we have recently proposed for multirate satellite image co...

Descripción completa

Detalles Bibliográficos
Autores: Mijares i Verdú, Sebastià|||0000-0003-1038-6413, Chabert, Marie|||0000-0001-9869-4693, Oberlin, Thomas|||0000-0002-9680-4227, Serra-Sagristà, Joan|||0000-0003-4729-9292
Tipo de recurso: artículo
Fecha de publicación:2024
País:España
Institución:Universitat Autònoma de Barcelona
Repositorio:Dipòsit Digital de Documents de la UAB
Idioma:inglés
OAI Identifier:oai:ddd.uab.cat:325621
Acceso en línea:https://ddd.uab.cat/record/325621
https://dx.doi.org/urn:doi:10.1109/JSTARS.2024.3422215
Access Level:acceso abierto
Palabra clave:Data compression
Neural network applications
Neural networks
Optical data processing
Remote sensing
Descripción
Sumario:Fixed-quality image compression is a coding paradigm where the tolerated introduced distortion is set by the user. This article proposes a novel fixed-quality compression method for remote sensing images. It is based on a neural architecture we have recently proposed for multirate satellite image compression. In this article, we show how to efficiently estimate the reconstruction quality using an appropriate statistical model. The performance of our approach is assessed and compared against recent fixed-quality coding techniques and standards in terms of accuracy and rate-distortion, as well as with recent machine learning compression methods in rate-distortion, showing competitive results. In particular, the proposed method does not introduce artifacts even when coding neighboring areas at different qualities.