English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Nowadays, modern Earth Observation systems continuously collect massiveamounts of satellite information. The unprecedented possibility to acquire highresolution Satellite Image Time Series (SITS) data (series of images with highrevisit time period on the same geographical area) is opening new opportunitiesto monitor the different aspects of the Earth Surface but, at the same time, itis raising up new challenges in term of suitable methods to analyze and exploitsuch huge amount of rich and complex image data. One of the main taskassociated to SITS data analysis is related to land cover mapping wheresatellite data are exploited via learning methods to recover the Earth Surfacestatus aka the corresponding land cover classes. Due to operationalconstraints, the collected label information, on which machine learningstrategies are trained, is often limited in volume and obtained at coarsegranularity carrying out inexact and weak knowledge that can affect the wholeprocess. To cope with such issues, in the context of object-based SITS landcover mapping, we propose a new deep learning framework, named TASSEL(aTtentive weAkly Supervised Satellite image time sEries cLassifier), that isable to intelligently exploit the weak supervision provided by the coarsegranularity labels. Furthermore, our framework also produces an additionalside-information that supports the model interpretability with the aim to makethe black box gray. Such side-information allows to associate spatialinterpretation to the model decision via visual inspection.

Attentive Weakly Supervised land cover mapping for object-based satellite image time series data with spatial interpretation