摘要
提出了一种利用遗传算法从高分辨率SAR图象中提取道路网络的方法。高分辨率SAR图象中目标背景复杂,同时由于受相干斑噪声的影响,很难直接从原始图象数据中提取道路特征。首先利用模糊C均值对滤波后的图象进行聚类,将道路类象素从图象中分离出来;根据聚类结果及道路特征建立数学模型,利用遗传算法搜索全局最优道路。实验结果表明该方法可以很好地从SAR图象中提取道路网络。
A method of using GA to extract roads from high-resolution SAR images is presented.As the background of objectives in high-resolution SAR images is complicated and also affected by speckle noise,it is almost impossible to extract roads directly from original remote sensing images.We use fuzzy C means to classify the filtered images unsuper-visedly to extract road pixels from images,and construct a model according to the classified results and the feature of roads,then search globally optimized road by means of genetic algorithm.The experimental results show that the algo-rithm can effectively extract road networks from high-resolution SAR images.
出处
《计算机工程与应用》
CSCD
北大核心
2004年第14期94-96,共3页
Computer Engineering and Applications
关键词
模糊聚类
遗传算法
道路提取
SAR图象
fuzzy clustering,Genetic Algorithm,road extraction,SAR image
