摘要
该文提出一种基于Adaboost的SAR图像溢油检测算法,该算法以纹理特征角二阶矩(ASM)、熵(ENT)、协同性(HOM)、相异性(DIS)作为图像特征向量,采用决策树作为Adaboost弱分类器,对SAR图像进行分类检测。实验结果证明该算法的有效性和可行性。
This paper presents an algorithm using Adaboost to detect oil spill in SAR Image.It use texture features-ASM,ENT,HOM and DIS as feature vector of images.The weak classifier of Adaboost is Decision Tree.Experimental results show that the algorithm is effective and feasible.
出处
《电脑知识与技术(过刊)》
2011年第10X期7252-7254,共3页
Computer Knowledge and Technology
关键词
ADABOOST
灰度共生矩阵
分类
溢油
Adaboost
gray level co-occurrence matrix
classification
oil spill