摘要
提出了一种基于2D LogGabor纹理特征分析的机场ROI(Region of Interest)自动分割方法。首先将原始图像分块并做傅里叶变换;然后采用基于上下文纹理信息的LogGabor算法,提取图像块的尺度特征;再通过直线拟合,利用尺度特征分割出图像的机场ROI。实验证明文中方法具有较好的检测效果。
This paper presents a method for automatic extraction of the airport ROI based on 2D LogGabor texture analysis.First,the original image is blocked and Fourier transformed;then feature information of the block is extracted by adopting contextual LogGabor algorithm,followed by the segmentation of the airport ROI using the feature information.Experiment shows that the proposed algorithm achieves better testing effect.
出处
《电子科技》
2010年第9期125-126,129,共3页
Electronic Science and Technology
