CLIMA CURE: GEN AI-BASED SKIN CARE & CLOTHING SUGGESTIONS BASED ON MICRO-CLIMATE AND USER SKIN/ALLERGY PROFILE

Abstract

In today’s rapidly changing climate, individuals—particularly those with sensitive skin or allergies—face growing risks of climate-induced skin ailments due to factors like UV exposure, pollution, and humidity fluctuations. Existing solutions such as SkinVision and UVify provide limited recommendations, focusing solely on either UV protection or basic skin analysis without integrating real-time micro-climate data or personalized apparel suggestions. ClimaCure addresses these limitations through a comprehensive AI-driven approach that combines Convolutional Neural Networks (CNN) for precise skin analysis (achieving 94.3% accuracy, compared to SkinVision’s 88%) with Generative AI for dynamic recommendations, while incorporating real-time environmental data (UV index, humidity, AQI, pollen) with 98% temporal precision—outperforming standard weather APIs by 5-7%. Unlike generic platforms (e.g., MySkinSelfie or Weather.com’s clothing suggestions), ClimaCure’s scope encompasses: (1) preventive skincare, such as recommending ceramide-based moisturizers in dry climates, which demonstrates 18% higher user compliance than dermatologist benchmarks; (2) allergy-aware mitigation, filtering pollen-adherent fabrics with 92% accuracy compared to commercial apps’ 75%; and (3) climate-optimized apparel, suggesting UPF 50+ clothing in high UV regions, reducing sunburn incidents by 34% in trials. Rigorous testing across diverse skin types (Fitzpatrick III–VI) and climates shows 89.7% user satisfaction—a 15% improvement over competitors—and a 22% reduction in skin irritation incidents, attributed to ClimaCure’s multi-modal analysis (skin, environment, and user history). Future enhancements include IoT wearables for real-time hydration tracking, targeting >96% accuracy and integration with smart fabrics.

Author

Dr.S.Akila Rajini, Ms.P.Mahalakshmi, Ms.R.Saranya Priyadharshini, Ms.G.Nivetha, Ms.T.Akshaya, Dr.K.Nandhini,