Rural decline threatens sustainable development, leading to lower life quality and environmental degradation. To revitalize regional areas and address youth unemployment problems, the Korean government established a Youth Village program to attract young people to rural towns and set up new businesses. A rural town, Mungyeong, was chosen as a case study to investigate its rural stay program, motivations, and outcomes. Based on the report review and preliminary interview with the program organizer, we identify the program’s strengths, weaknesses, and critical success factors. The issues are then examined through a cultural lens to identify cultural and regional differences that may act as enablers or barriers to social innovation. The paper contributes an in-depth understanding of local challenges to develop an appropriate social innovation strategy. We anticipate that Mungyeong’s lessons will apply to other regional social innovation projects.

Accurate detection and severity estimation of corrosion on metallic surfaces is essential for maintaining material integrity and ensuring operational safety in industrial systems. To address limitations in manual inspection methods, this study presents a two-stage deep learning pipeline tailored for high-resolution scanning electron microscopy images. The framework combines instance-level corrosion segmentation using the YOLOv8-seg architecture with subsequent severity classification performed by EfficientNet-B0 and ResNet18. In the segmentation stage, models are trained using both manually annotated and automatically generated binary masks, enabling robust instance mask prediction through prototype-based mask decoding. The classification stage assesses the severity of corrosion by analyzing localized regions based on morphological features, leveraging convolutional neural networks optimized for binary output. Experimental results demonstrate strong performance: the segmentation model trained on manual annotations achieves a mean Intersection-over-Union (mIoU) of 89.91, a mask mAP@50 of 98.6, and an ROC-AUC of 94.69. For severity classification, EfficientNet-B0 achieves an accuracy of 93.75% and an F1-score of 93.29, outperforming ResNet18. These findings underscore the potential of combining lightweight segmentation and classification models to enable precise and automated corrosion analysis, supporting integration into real-time inspection workflows and intelligent material diagnostics in industrial settings.

This paper presents a speculative framework for service design education, practice, and research in light of emerging technologies. It examines the evolution of service design through stages, from its foundations to a “thinking” era, and proposes a model for “Service Design 2.0” that considers industry-led, human-centred, and information driven approaches. The paper explores future trends in service design practice, emphasizing socio-technical systems thinking and learning systems. Design research is discussed in terms of evolving user-designer interactions, moving toward symbiotic relationships where designing and using become intertwined. Finally, the paper advocates for a balanced design education that integrates arts, engineering, computer science, and humanities, while also introducing Critical Design not merely as a counter-perspective, but as a vital mode of inquiry for questioning the ideological assumptions embedded in socio-technical systems. The study considers both affirmative-technological and critical-humanistic outlooks.

This research explores the factors that contribute to resilient local youth communities in South Korea, using the “Workday Club” as a case study. It examines how these communities combat the challenges posed by urbanization and migration to Seoul causing a regional decline and a depopulation crisis. The study emphasizes the importance of rekindling traditional Korean community values, like “Dure,” which fosters collaboration and social support. The research also analyses youth communities and entrepreneurship and identifies critical factors such as social engagement and entrepreneurial opportunities as key to building resilience. Ultimately, it provides insights and strategies for building thriving youth communities that blend social and business aspects, foster mentorship, and encourage adaptability in a changing world.

Accurate detection and severity estimation of corrosion on metallic surfaces is essential for maintaining material integrity and ensuring operational safety in industrial systems. To address limitations in manual inspection methods, this study presents a two-stage deep learning pipeline tailored for high-resolution scanning electron microscopy images. The framework combines instance-level corrosion segmentation using the YOLOv8-seg architecture with subsequent severity classification performed by EfficientNet-B0 and ResNet18. In the segmentation stage, models are trained using both manually annotated and automatically generated binary masks, enabling robust instance mask prediction through prototype-based mask decoding. The classification stage assesses the severity of corrosion by analyzing localized regions based on morphological features, leveraging convolutional neural networks optimized for binary output. The experimental results demonstrate strong performance: the segmentation model trained on manual annotations achieves a Mean Intersection over Union (mIoU) of 89.91, a mask mAP@50 of 98.6, and an ROC-AUC of 94.69. For severity classification, EfficientNet-B0 achieves an accuracy of 93.75% and an F1-score of 93.29, outperforming ResNet18. The proposed framework connects advanced SEM with state-of-the-art machine learning. It provides a scalable, annotation-efficient way to use intelligent and automated corrosion characterization in materials science and industrial applications.