Emphasized Visualization of Hidden Age Spots Using Deep Learning of Young Skin

Autor:	Makoto Hasegawa, Rui Matsuo
Rok vydání:	2021
Předmět:	business.product_category Smart phone Artificial neural network Computer science business.industry Deep learning ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Younger people Visualization Computer vision Artificial intelligence business Young person Digital camera
Zdroj:	2021 36th International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC).
DOI:	10.1109/itc-cscc52171.2021.9501434
Popis:	An emphasized visualization of hidden age spots on the skin of older individuals using deep learning is discussed. Young person's UV skin images of younger people were captured using a medical dermoscopy digital camera. These UV images were then learned using neural networks. Even with a general smart phone camera, it is possible to generate an image that approximates a medical UV image through a trained neural network. The skin of a younger person has some moles but few age spots. Therefore, deep learning cannot be used to display hidden age spots that were not seen when the persons were young. By computing the difference image between a generated image and a generated image using images from the older generations, it is possible to extract hidden age spots.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::06e760b6b6c0e9bbc1983bdf5d936810 https://doi.org/10.1109/itc-cscc52171.2021.9501434 Zobrazit plný text záznamu