Pigment-Aware: Understanding Skin Tone Variations in Digital Image Analysis

Autor(es)

Marasco, Emanuela ; Cruz Junior, Luismar B.

• Tipo de documento: Artigo Científico
• Fonte: IEEE ACCESS; v. 13, p. 10-pg., 2025-01-01.

Resumo

Optical sensors, such as the RGB cameras embedded in smartphones, often fail to accurately capture the full spectrum of skin tones. As a result, individuals with darker skin may experience reduced performance in machine vision-based security systems. Insufficient attention to human diversity, including variations in skin tone, can contribute to biased training data and, subsequently, to disparities in AI systems, especially in biometric recognition. This paper highlights the need for more consistent and objective approaches to assessing skin tone, which are often treated subjectively or applied inconsistently. We address this issue by analyzing RGB finger photo data using colorimetric techniques to support the development of more inclusive machine vision systems. (AU)

• Processo FAPESP: 13/07276-1 - CEPOF - Centro de Pesquisa em Óptica e Fotônica
• Beneficiário: Vanderlei Salvador Bagnato
• Modalidade de apoio: Auxílio à Pesquisa - Centros de Pesquisa, Inovação e Difusão - CEPIDs
• Processo FAPESP: 14/50857-8 - INCT 2014 - de Óptica Básica e Aplicada às Ciências da Vida
• Beneficiário: Vanderlei Salvador Bagnato
• Modalidade de apoio: Auxílio à Pesquisa - Temático
• Processo FAPESP: 24/00206-2 - Avaliação e monitoramento da terapia fotodinâmica em câncer de pele do tipo não-melanoma in-vivo usando imagens espacialmente moduladas
• Beneficiário: Luismar Barbosa da Cruz Junior
• Modalidade de apoio: Bolsas no Brasil - Pós-Doutorado