This research proposes a new efficient encryption technique for multiple-color medical images. Using a random permutation number and mathematical formulas, the technique first separates every image into Red (R), Green (G), and Blue (B) Channels. These channels are then mixed with channels from other images. By drastically lowering the correlation between neighboring pixels, this inter-channel mixing improves the resistance of statistical analysis. Because of the increased uncertainty and diffusion introduced by this scrambling process, the encryption is highly resistant to statistical and differential attacks. For channel diffusion and pixel permutations, independent encryption keys are generated by logistic, Henon, and Chebyshev chaotic maps. The computed numerical results show strong performance, including an average entropy of 7.9994, key space of 2^512 , NPCR of 99.61
Video auditing can provide vital insights into complex industrial operations where quality, safety, and compliance are critical. However, there is a need to develop explainable artificial intelligence frameworks capable of recognising workflow steps and producing interpretable reports from long, untrimmed video sequences. The key module enabling reliable auditing is action segmentation, which remains challenging due to ambiguous step boundaries, fine-grained action similarities, and uncertainties caused by environmental variability, multi-object interactions, and class imbalance. To address these challenges, we introduce a framework which integrates fuzzy logic systems, deep learning models, and large language models through retrieval-augmented generation (RAG). The fuzzy logic component operates at two stages: generating soft training labels at action boundaries and providing per-window decision confidence with linguistic interpretability at inference, while a hybrid VideoMAE–TimeSformer architecture with MS-TCN++ sequence modelling captures spatiotemporal dependencies for accurate step prediction. The large language model component enriches predictions with external knowledge and generates structured, human-readable audit reports aligned with operational standards through retrieval-augmented generation. Several experiments on real-world maintenance videos were conducted to compare the proposed system with conventional deep learning pipelines. The results demonstrate that the proposed fuzzy logic system enhances the interpretability of audit outcomes, providing human-understandable confidence reasoning at action boundaries and generating explainable audit reports aligned with operational standards.
The rapid integration of Artificial Intelligence (AI) in global finance has outpaced the development of practical legal, ethical, and systemic governance frameworks, creating critical adoption challenges. Addressing this gap, this study examines the limitations of AI adoption in fintech through a holistic resilience perspective. Using the SETEH (Systemic–Ethical–Technical–Economic–Human) framework, the research analyzes 40 fintech case studies to assess how regulatory compliance, ethical governance, systemic risk, human psychological factors, and IT infrastructure jointly influence sustainable AI integration. The findings reveal that regulatory adaptability and robust oversight are essential to mitigate decision opacity, market concentration, and cyber risks. Ethical governance enhances transparency, reduces stakeholder uncertainty, and alleviates human resistance such as distrust and job insecurity. The study demonstrates that AI adoption failures often stem from overlooked interdependencies across legal, technical, and human dimensions. By offering an integrated analytical model grounded in empirical case evidence, this research contributes a structured framework for strengthening resilient, ethical, and legally compliant AI adoption in global financial systems.
Communicating complex environmental phenomena requires forms of visualization that go beyond analytical accuracy to support engagement, interpretation, and reflection. Eco-visualization addresses this challenge by combining data representation with aesthetic and narrative elements, but its creation often demands technical skills. This paper presents a Human–AI co-creative pipeline for the design of eco-visualizations, in which AI acts as an active collaborator throughout the visual creation process. The approach integrates generative image creation, AI-assisted segmentation, and data-driven color mapping into a structured workflow that supports iterative interaction between human and AI. We implement this pipeline as a web-based system and evaluate it through a user study with a mixed-method approach. Results show that while the system is perceived as engaging and conceptually innovative, users experience notable usability challenges and cognitive load during interaction. These findings highlight key tensions in Human–AI co-creation, particularly between control and ease of use. Overall, this work contributes a concrete design approach to Human–AI collaboration in eco-visualization and provides empirical insights that inform the development of more accessible and effective co-creative tools for communicating complex environmental data.
Mobile network data represent one of the most powerful emerging sources for large-scale mobility analysis. This paper proposes a complete data-driven pipeline to understand, detect, and predict population movement patterns within a region using anonymized origin–destination data provided by mobile phone operators. The dataset, covering 60 consecutive days in 2019, reports millions of daily flows between census areas together with demographic information. We conduct extensive exploratory analysis, develop a Z-score anomaly detection method capable of capturing event-driven disruptions, and design a regression model achieving a regional-scale MAPE around 10
