28 September 2021 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Land cover (LC) mapping is essential for monitoring the environment and understanding the effects of human activities on it. Recent studies demonstrated successful applications of specific deep learning models to small-scale LC mapping tasks (e.g., wetland mapping). However, it is not readily clear which of the existing state-of-the-art models for natural images are the best candidates to be taken for the particular remote sensing task and data. In this article, we answer that question for mapping the fundamental LC classes using the satellite imaging radar data. We took ESA Sentinel-1 C-band SAR images acquired during the whole summer season of 2018 in Finland, which are representative of the land cover in the country. CORINE LC map was used as a reference, and the models were trained to distinguish between the five major CORINE-based classes. We selected seven among the state-of-the-art semantic segmentation models so that they cover a diverse set of approaches: U-Net, DeepLabV3+, PSPNet, BiSeNet, SegNet, FC-DenseNet, and FRRN-B, and further fine-tuned them. Upon evaluation and benchmarking, all the models demonstrated solid performance with overall accuracy between 87.9% and 93.1%, with good to a very good agreement (Kappa statistic between 0.75 and 0.86). The two best models were fully convolutional DenseNets (FC-DenseNet) and SegNet (encoder-decoder-skip), with the latter having a much shorter inference time. Overall, our results indicate that the semantic segmentation models are suitable for efficient wide-area mapping using satellite SAR imagery and provide baseline accuracy against which the newly proposed models should be evaluated.
16 May 2017 in IEEE Transactions on Circuits and Systems I: Regular Papers
Real networks in our surrounding are usually complex and composite by nature and they consist of many interwoven layers. The commutation of agents (nodes) across layers in these composite multiplex networks heavily influences the underlying dynamical processes, such as information, idea and disease spreading, synchronization, consensus, etc. In order to understand how the agents' dynamics and the compositeness of multiplex networks influence the spreading dynamics, we develop a susceptible-infected-susceptible-based model on the top of these networks, which is integrated with the transition of agents across layers. Moreover, we analytically obtain a critical infection rate for which an epidemic dies out in a multiplex network, and latter show that this rate can be higher compared with the isolated networks. Finally, using numerical simulations we confirm the epidemic threshold and we show some interesting insights into the epidemic onset and the spreading dynamics in several real and generic multiplex networks.
16 October 2024 in Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society
In the evolving landscape of AI regulation, it is crucial for companies to conduct impact assessments and document their compliance through comprehensive reports. However, current reports lack grounding in regulations and often focus on specific aspects like privacy in relation to AI systems, without addressing the real-world uses of these systems. Moreover, there is no systematic effort to design and evaluate these reports with both AI practitioners and AI compliance experts. To address this gap, we conducted an iterative co-design process with 14 AI practitioners and 6 AI compliance experts and proposed a template for impact assessment reports grounded in the EU AI Act, NIST's AI Risk Management Framework, and ISO 42001 AI Management System. We evaluated the template by producing an impact assessment report for an AI-based meeting companion at a major tech company. A user study with 8 AI practitioners from the same company and 5 AI compliance experts from industry and academia revealed that our template effectively provides necessary information for impact assessments and documents the broad impacts of AI systems. Participants envisioned using the template not only at the pre-deployment stage for compliance but also as a tool to guide the design stage of AI uses.
21 November 2025 in Nature Cities
Urban parks are important for public health, but the role of specific spaces, such as playgrounds or lakes, and elements, such as benches or sports equipment, in supporting well-being is not well understood. Here, based on expert input and a review of the literature, we defined six types of health-related activity: physical, mindfulness, nature appreciation, environmental, social and cultural. We built a lexicon that links each activity to specific elements and spaces within parks present in OpenStreetMap. Using these data, we scored 23,477 parks across 35 cities worldwide on the basis of their ability to support these activities. We found clear patterns: parks in North America focus more on physical activity, while those in Europe offer more chances to enjoy nature. Parks near city centers support health-promoting activities better than those farther out. Suburban parks in many cities lack the spaces and equipment needed for nature-based, social and cultural activities. We also found large gaps in park quality between cities. Tokyo and Paris provide more equal access, while Copenhagen and Rio de Janeiro show sharp contrasts. These results can help cities create fairer parks that better support public health.
24 September 2024 in Proceedings of the ACM on Human-Computer Interaction
Integrating Artificial Intelligence (AI) into mobile and wearables offers numerous benefits at individual, societal, and environmental levels. Yet, it also spotlights concerns over emerging risks. Traditional assessments of risks and benefits have been sporadic, and often require costly expert analysis. We developed a semi-automatic method that leverages Large Language Models (LLMs) to identify AI uses in mobile and wearables, classify their risks based on the EU AI Act, and determine their benefits that align with globally recognized long-term sustainable development goals; a manual validation of our method by two experts in mobile and wearable technologies, a legal and compliance expert, and a cohort of nine individuals with legal backgrounds who were recruited from Prolific, confirmed its accuracy to be over 85%. We uncovered that specific applications of mobile computing hold significant potential in improving well-being, safety, and social equality. However, these promising uses are linked to risks involving sensitive data, vulnerable groups, and automated decision-making. To avoid rejecting these risky yet impactful mobile and wearable uses, we propose a risk assessment checklist for the Mobile HCI community.
2 September 2024 in IEEE Internet Computing
Impact assessment reports for high-risk AI systems will be legally required but challenging to complete, especially for smaller companies. That is because the current process is complex, costly, and relies on guidebooks with limited assistance. We propose AI Design , a semi-automatic framework for populating these reports. It consists of two components: (A) StakeLinker , an interactive tool combining various stakeholders’ perspectives; and (B) FillGen , an LLM-based tool processing stakeholders’ perspectives and producing the report to be reviewed by regulatory experts. We conducted two user studies: the first with 13 AI practitioners who confirmed StakeLinker’s effectiveness in gathering comprehensive input for impact assessment; the second with 8 additional practitioners who successfully evaluated a report for a crime analysis system pre-populated by FillGen . For generalizability purposes, we provide reports for two other AI systems.