| 基于卫星遥感的输油管廊沿线的地物识别研究 |
| |
| 引用本文： | 吴志锋,郑宇恒,李双琴,孙啸,张涛,沈忱,邹妍.基于卫星遥感的输油管廊沿线的地物识别研究[J].地理空间信息,2022,20(2):45-50. |
| |
| 作者姓名： | 吴志锋 郑宇恒 李双琴 孙啸 张涛 沈忱 邹妍 |
| |
| 作者单位： | 中国石油天然气股份有限公司西南管道分公司,成都 610037 |
| |
| 摘 要： | 基于eCognition软件进行研究区域面向对象的信息提取,对研究区进行多尺度影像分割的实验,找到适合影像地物信息提取的尺度及参数、分类规则,并对研究区的影像数据进行验证,通过eCognition分类结果与传统分类和基于GBDT分类算法结果的对比,说明面向对象的特征提取算法更适用于几何信息和结构信息丰富的高分辨率图像。
|
| 关 键 词： | 遥感影像 多尺度分割 影像特征 地物识别 |
Study on Object Identification along Oil Pipeline Corridor Based on Satellite Remote Sensing |
| |
| WU Zhifeng,ZHENG Yuheng,LI Shuangqin,SUN Xiao,ZHANG Tao,SHEN Chen,ZOU Yan.Study on Object Identification along Oil Pipeline Corridor Based on Satellite Remote Sensing[J].Geospatial Information,2022,20(2):45-50. |
| |
| Authors: | WU Zhifeng ZHENG Yuheng LI Shuangqin SUN Xiao ZHANG Tao SHEN Chen ZOU Yan |
| |
| Institution: | (Southwest pipeline branch of China national petroleum corporation,Chengdu 610037,China) |
| |
| Abstract: | In this paper,we selected the GF-2 satellite remote sensing image data to study the object recognition along oil pipeline,and used the object-oriented eCognition software to extract study area information.Then,we carried out multi-scale image segmentation experiments in the study area,and found the appropriate scale,parameters and classification rules of image object information extraction.Finally,we verified the im-age data of study area.Comparing with the results of eCognition classification,traditional classification and GBDT classification algorithm,we found that the object-oriented feature extraction algorithm is more suitable for the high resolution images with rich geometric information and structural information. |
| |
| Keywords: | remote sensing image multi-scale segmentation image feature object recognition |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
