Maria Inês Sousa Machado

Dr. Maria Inês Sousa Machado

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 January 2025 in International Journal of Data Science and Analytics
Verified
Journal Article
Open access
PDF file
JCJoão Canelas
João Canelas
LCLuana Clementino
Luana Clementino
ACAndré Cid
André Cid
Joana CastroJoana Castro
Joana Castro
MMMaria Inês Sousa Machado
Maria Inês Sousa Machado
Susana VieiraSusana Vieira
Susana Vieira
The identification and quantification of marine mammals is crucial for understanding their abundance, ecology and supporting their conservation efforts. Traditional methods for detecting cetaceans, however, are often labor-intensive and limited in their accuracy. To overcome these challenges, this work explores the use of convolutional neural networks (CNNs) as a tool for automating the detection of cetaceans through aerial images from unmanned aerial vehicles (UAVs). Additionally, the study proposes the use of Long-Short-Term-Memory (LSTM)-based models for video detection using a CNN-LSTM architecture. Models were trained on a selected dataset of dolphin examples acquired from 138 online videos with the aim of testing methods that hold potential for practical field monitoring. The approach was effectively validated on field data, suggesting that the method shows potential for further applications for operational settings. The results show that image-based detection methods are effective in the detection of dolphins from aerial UAV images, with the best-performing model, based on a ConvNext architecture, achieving high accuracy and f1-score values of 83.9% and 82.0%, respectively, within field observations conducted. However, video-based methods showed more difficulties in the detection task, as LSTM-based models struggled with generalization beyond their training environments, achieving a top accuracy of 68%. By reducing the labor required for cetacean detection, thus improving monitoring efficiency, this research provides a scalable approach that can support ongoing conservation efforts by enabling more robust data collection on cetacean populations.

10 December 2025 in International Journal of Data Science and Analytics
Verified
Journal Article
Open access
PDF file
LVLara Veylit
Lara Veylit
UBUte Brönner
Ute Brönner
KFKari Fischer
Kari Fischer
DKDana Margareta King
Dana Margareta King
SSSahar Stevenson-Jones
Sahar Stevenson-Jones
IØIda Beathe Øverjordet
Ida Beathe Øverjordet
Restoration of degraded marine and coastal areas is a priority set by European policy makers, as exemplified by the recent adoption of the European Nature Restoration Law (NRL) in June 2024. This legislation-driven approach is expected to expand the European marine restoration community rapidly as the amount of funding and projects grows to meet the targets outlined by the NRL However, it is difficult to assess the success of restoration activities which is vital to ensure the viability of upscaling. Furthermore, best practices are not well established in the marine realm and scientific networks are still in their early stages of development. Here, we outline our development of a digital, online toolbox for marine restoration in collaboration with the European restoration community. The toolbox aims to go beyond simply making data FAIR (Findable, Accessible, Interoperable, and Reusable) through the creation of science-based digital services that allow for use of information for decision-making for marine restoration. The toolbox is constructed in a modular way to fulfil the needs of a diverse range of users across Europe and includes a centralized space to find methodological approaches, relevant networks, funding opportunities, and resources. The digital tools under development will be openly accessible through the Blue-Cloud 2026 platform within the thematic virtual research environment for marine restoration, which allows for the scalable use and reuse of data and code by any interested user. The toolbox aims to provide restoration community members science-based methods from which they can leverage and accelerate their restoration projects. In the pursuit to democratize access to best practices and knowledge, we go beyond providing code or data in static repositories. Based on our experience, we encourage others that are developing toolboxes or knowledge bases for diverse user groups to consider taking advantage of publicly funded infrastructure to promote their work according to open science practices.

18 April 2026 in International Journal of Data Science and Analytics
Verified
Journal Article
Open access
PDF file
LVLara Veylit
Lara Veylit
UBUte Brönner
Ute Brönner
KFKari Fischer
Kari Fischer
DKDana Margareta King
Dana Margareta King
SSSahar Stevenson-Jones
Sahar Stevenson-Jones
IØIda Beathe Øverjordet
Ida Beathe Øverjordet