Kim J, Lee O. Correlation analysis between texture features and elasticity of skin hyperspectral images in the near-infrared band.
Skin Res Technol 2024;
30:e13654. [PMID:
38504440 PMCID:
PMC10951415 DOI:
10.1111/srt.13654]
[Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/06/2024] [Accepted: 02/19/2024] [Indexed: 03/21/2024]
Abstract
BACKGROUND/PURPOSE
Skin elasticity was used to evaluate healthy and diseased skin. Correlation analysis between image texture characteristics and skin elasticity was performed to study the feasibility of assessing skin elasticity using a non-contact method.
MATERIALS AND METHODS
Skin images in the near-infrared band were acquired using a hyperspectral camera, and skin elasticity was obtained using a skin elastimeter. Texture features of the mean, standard deviation, entropy, contrast, correlation, homogeneity, and energy were extracted from the acquired skin images, and a correlation analysis with skin elasticity was performed.
RESULTS
The texture features, and skin elasticity of skin images in the near-infrared band had the highest correlation on the side of eye and under of arm, and the mean and correlation were features of texture suitable for distinguishing skin elasticity according to the body part.
CONCLUSION
In this study, we performed elasticity and correlation analyses for various body parts using the texture characteristics of skin hyperspectral images in the near-infrared band, confirming a significant correlation in some body parts. It is expected that this will be used as a cornerstone of skin elasticity evaluation research using non-contact methods.
Collapse