A practically feasible transfer learning method for deep-temporal urban change monitoring

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27740%2F23%3A10252827" target="_blank" >RIV/61989100:27740/23:10252827 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1080/01431161.2023.2243021" target="_blank" >https://doi.org/10.1080/01431161.2023.2243021</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1080/01431161.2023.2243021" target="_blank" >10.1080/01431161.2023.2243021</a>

Jazyk výsledku

angličtina

Název v původním jazyce

A practically feasible transfer learning method for deep-temporal urban change monitoring

Popis výsledku v původním jazyce

Neural networks have shown their potential to monitor urban changes with deep-temporal remote sensing data, which simultaneously considers a large number of observations within a given window. However, training these networks with supervision is a challenge due to the low availability of third-party sources with sufficient spatio-temporal resolution to label each window individually. To remedy this problem, we developed a novel approach utilizing transfer learning (TL) on a set of deep-temporal windows. We demonstrate that labelling of multiple windows simultaneously can be practically viable, even with a low amount of high spatial resolution third-party data. The overall process provides a trade-off between labour resources and the ability to train a network on existing systems, despite its intensive memory requirements. As a demonstration, an existing previously trained (pre-trained) network was used to transfer knowledge to a new target location. We demonstrate our method with combined Sentinel 1 and 2 observations for the area of Liège (Belgium) for the time period spanning 2017-2020. This is underpinned by our use of common metrics in machine learning and remote sensing, and in our discussion of selected examples. Three independent transfers of the same pre-trained model and their combination were carried out, all of which showed an improvement in terms of these metrics.

Název v anglickém jazyce

A practically feasible transfer learning method for deep-temporal urban change monitoring

Popis výsledku anglicky

Neural networks have shown their potential to monitor urban changes with deep-temporal remote sensing data, which simultaneously considers a large number of observations within a given window. However, training these networks with supervision is a challenge due to the low availability of third-party sources with sufficient spatio-temporal resolution to label each window individually. To remedy this problem, we developed a novel approach utilizing transfer learning (TL) on a set of deep-temporal windows. We demonstrate that labelling of multiple windows simultaneously can be practically viable, even with a low amount of high spatial resolution third-party data. The overall process provides a trade-off between labour resources and the ability to train a network on existing systems, despite its intensive memory requirements. As a demonstration, an existing previously trained (pre-trained) network was used to transfer knowledge to a new target location. We demonstrate our method with combined Sentinel 1 and 2 observations for the area of Liège (Belgium) for the time period spanning 2017-2020. This is underpinned by our use of common metrics in machine learning and remote sensing, and in our discussion of selected examples. Three independent transfers of the same pre-trained model and their combination were carried out, all of which showed an improvement in terms of these metrics.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10200 - Computer and information sciences

Projekt

—

Návaznosti

—

Rok uplatnění

2023

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

International Journal of Remote Sensing

ISSN

0143-1161

e-ISSN

1366-5901

Svazek periodika

44

Číslo periodika v rámci svazku

17

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

35

Strana od-do

5172-5206

Kód UT WoS článku

001054398700001

EID výsledku v databázi Scopus

2-s2.0-85169299359