Fleming M G, Steger C, Zhang J, Gao J, Cognetta A B, Pollak I, Dyer C R
Department of Dermatology, Medical College of Wisconsin and Zablocki VA Hospital, Milwaukee, USA.
Comput Med Imaging Graph. 1998 Sep-Oct;22(5):375-89. doi: 10.1016/s0895-6111(98)00048-2.
Techniques were developed for automated detection and characterization of dermatoscopic structures, including the pigment network and brown globules. These techniques incorporate algorithms for grayscale shape extraction based on differential geometry developed by Steger, a snake algorithm, and a modification of the region competition strategy of Zhu and Yuille. A novel approach was developed for global segmentation of pigmented lesions, based on stabilized inverse diffusion equations. Procedures for detection of air bubbles and hairs in dermatoscopic images are also reported.
已开发出用于自动检测和表征皮肤镜结构的技术，包括色素网络和棕色小球。这些技术纳入了基于Steger开发的微分几何的灰度形状提取算法、一种蛇形算法以及对Zhu和Yuille的区域竞争策略的改进。基于稳定的逆扩散方程，开发了一种用于色素沉着病变全局分割的新方法。还报告了检测皮肤镜图像中气泡和毛发的程序。
