Celebi M Emre, Iyatomi Hitoshi, Schaefer Gerald, Stoecker William V
Department of Computer Science, Louisiana State University, Technology Center 206, One University Place, Shreveport, LA 71115, USA.
Skin Res Technol. 2009 Aug;15(3):314-22. doi: 10.1111/j.1600-0846.2009.00357.x.
Dermoscopy is one of the major imaging modalities used in the diagnosis of melanoma and other pigmented skin lesions. Because of the difficulty and subjectivity of human interpretation, automated analysis of dermoscopy images has become an important research area. Border detection is often the first step in this analysis.
In this article, we present an approximate lesion localization method that serves as a preprocessing step for detecting borders in dermoscopy images. In this method, first the black frame around the image is removed using an iterative algorithm. The approximate location of the lesion is then determined using an ensemble of thresholding algorithms.
The method is tested on a set of 428 dermoscopy images. The localization error is quantified by a metric that uses dermatologist-determined borders as the ground truth.
The results demonstrate that the method presented here achieves both fast and accurate localization of lesions in dermoscopy images.
皮肤镜检查是用于诊断黑色素瘤和其他色素沉着性皮肤病变的主要成像方式之一。由于人工解读存在困难且具有主观性，皮肤镜图像的自动分析已成为一个重要的研究领域。边界检测通常是该分析的第一步。
在本文中，我们提出了一种近似病变定位方法，作为检测皮肤镜图像边界的预处理步骤。在该方法中，首先使用迭代算法去除图像周围的黑色边框。然后使用一组阈值算法确定病变的近似位置。
该方法在一组428张皮肤镜图像上进行了测试。定位误差通过一种将皮肤科医生确定的边界作为真实标准的指标进行量化。
结果表明，本文提出的方法在皮肤镜图像中实现了病变的快速准确定位。
