利用能量函数和主动轮廓模型提取道路边缘线  被引量:2

Extraction of Road Edge Lines from Remote Sensing Images Based on Energy Function and Snake Model

在线阅读下载全文

作  者:钱海明[1,2,3] 张印 王春林[1,2,3] 孙金彦[1,2,3] 黄祚继 QIAN Haiming;ZHANG Yin;WANG Chunlin;SUN Jinyan;HUANG Zuoji(Anhui&Huaihe River Institute of Hydraulic Research,Hefei 230088,China;Key Laboratory of Water Conservancy and Water Resources in Anhui Province,Hefei 230088,China;Anhui Dayu Hydraulic Conservancy Engineering Technology Co.Ltd,Hefei 230088,China;China Electric Power Research Institute,Beijing 10000,China)

机构地区:[1]安徽省(水利部淮河水利委员会)水利科学研究院,合肥230088 [2]水利水资源安徽省重点实验室,合肥230088 [3]安徽省大禹水利工程科技有限公司,合肥230088 [4]中国电力科学研究院,北京100000

出  处:《遥感信息》2018年第4期16-21,共6页Remote Sensing Information

摘  要:针对高分辨率遥感影像道路边缘线交叉口断裂问题,提出一种利用能量函数和主动轮廓型提取遥感道路边缘线方法。在利用概率增进树算法和形态学运算提取道路条带,通过细化算法获取道路中心线基础上,沿道路中心线外推以检测初始道路双边缘线;鉴于道路边缘线存在局部曲率变化过大的问题,利用梯度特征和光谱信息构建能量函数,设计路径代价,结合Dijkstra算法优化初始道路双边缘线。通过构建Snake模型,利用离散动态规划算法解算模型,连接交叉口断裂处,解决交叉口断裂问题,进而得到最终道路双边缘线。实验结果表明,该方法能够较好提取连续、平滑的道路双边缘线,准确率达88%。In this paper,the strategy to extract double road edge lines from acquired road stripe image is explored.The workflow is as follows:the road stripes are immediately acquired by probabilistic boosting tree algorithm and morphology,road centerlines are detected by thinning algorithm,and the initial road edge lines can be acquired along the road centerlines;the output contained local curvature of centerlines change much.In order to solve this problem,the energy function of edge line is constructed by gradient feature and spectral information,and Dijkstra algorithm is used to optimize the initial road edge lines.The Snake model is constructed to solve the fracture problem of intersection,and the discrete dynamic programming algorithm is used to solve the model.Experimental results show that the strategy proposed in this paper can be used to extract the continuous and smooth road edge lines from high resolution remote sensing images,and the accuracy is 88%.

关 键 词:道路边缘线 交叉口断裂 能量函数 主动轮廓模型 动态规划算法 

分 类 号:TP751.1[自动化与计算机技术—检测技术与自动化装置]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象