The tumors on the skin are characterized by a high mortality rate. Research is attempting the automatic early diagnosis of melanoma, a lethal form of skin cancer, from dermoscopic images. Automatic diagnostics provides a valid 'second opinion' to support physicians in deciding whether a skin lesion is a benign mole or a malignant melanoma. Determining effective detection methods to reduce the rate of error in diagnosis is a crucial challenge. Computer vision systems are characterized by several fundamental steps. Preprocessing is the first phase of detection and plays a fundamental role: the elimination of noise and irrelevant parts against the background of skin images to improve image quality. The purpose of this paper is to review the pre-processing approaches that can be used on skin cancer images. The current interest in the automatic analysis of images, is motivated by the possibility of being able to provide quantitative information on a lesion and to implement self diagnosis solutions.
Image pre-processing in computer vision systems for melanoma detection
Vocaturo E.;Zumpano E.;Veltri P.
2019-01-01
Abstract
The tumors on the skin are characterized by a high mortality rate. Research is attempting the automatic early diagnosis of melanoma, a lethal form of skin cancer, from dermoscopic images. Automatic diagnostics provides a valid 'second opinion' to support physicians in deciding whether a skin lesion is a benign mole or a malignant melanoma. Determining effective detection methods to reduce the rate of error in diagnosis is a crucial challenge. Computer vision systems are characterized by several fundamental steps. Preprocessing is the first phase of detection and plays a fundamental role: the elimination of noise and irrelevant parts against the background of skin images to improve image quality. The purpose of this paper is to review the pre-processing approaches that can be used on skin cancer images. The current interest in the automatic analysis of images, is motivated by the possibility of being able to provide quantitative information on a lesion and to implement self diagnosis solutions.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.