对称局部能量模型:改进的局部能量模型

本文对局部能量模型进行改进,提出了一种新的基于相位信息的图像特征检测算法——对称局部能量模型(SLE)。该算法以log Gabor小波为基本工具,只用四个对称相位进行图像特征检测,检测结果符合人类视觉系统特性。SLE不但可以检测阶跃特征、线特征等亮度特征,还可以同时检测到马赫带现象和正弦网格中的特征,而基于亮度梯度的算法和局部能量模型都不能做到这一点。与局部能量模型相比,SLE还提高了抗噪能力和特征定位性能。

Symmetry Local Energy Model:The Improved Local Energy Model

This paper presents an improved Local Energy model called Symmetry Local Energy model (SLE), which is a novel image feature detector based on phase information. SLE extracts the image features with only four symmetric phases. And the results are consistent with the characteristics of human visual systems. Not only can SLE detect the luminance profile, such as step and impulse discontinuities, but also can detect the features in Mach bands and in sinusoidal grating at the same time. These cannot be done simply by looking for local maxima in intensity gradient or by looking for local energy maxima. Compared with LE, SLE also improves the anti-noise ability and localization accuracy.