Computer vision techniques for early detection of skin cancer

This thesis investigates the problem of developing new computer vision techniques for early detection of skin cancer. The first part of this work presents a novel methodology to correct color reproduction in dermatological images when different cameras and/or dermoscopes are used. Next, the problem...

Descripción completa

Detalles Bibliográficos
Autor: Quintana Plana, Josep
Tipo de recurso: tesis doctoral
Estado:Versión publicada
Fecha de publicación:2012
País:España
Institución:CBUC, CESCA
Repositorio:TDR. Tesis Doctorales en Red
OAI Identifier:oai:www.tdx.cat:10803/82072
Acceso en línea:http://hdl.handle.net/10803/82072
Access Level:acceso abierto
Palabra clave:Skin cancer
Càncer de pell
Cáncer de pell
Melanoma
Computer vision
Visió per ordinador
Visión por ordenador
3D scanner
Color calibration
Mosaicing
68
Descripción
Sumario:This thesis investigates the problem of developing new computer vision techniques for early detection of skin cancer. The first part of this work presents a novel methodology to correct color reproduction in dermatological images when different cameras and/or dermoscopes are used. Next, the problem of automatic full body mapping is addressed by proposing a mosaicing method based on an on-the-shelf digital compact camera and a set of markers. This method increases the possibilities of total body photography by taking the low-resolution images of a whole body exploration and automatically combining them into a high-resolution photomosaic. The third contribution of this work consists of the development of a full body scanner for acquiring cutaneous images. On one hand, the scanner reduces the long time-consuming examinations done in dermoscopy explorations, and on the other hand, it increases the resolution of total body photography systems.