| 基于多源遥感影像的长春市城市建成区提取 |
| |
| 引用本文： | 王博帅,蒲东川,李婷婷,牛雪峰.基于多源遥感影像的长春市城市建成区提取[J].吉林大学学报(地球科学版),2021,51(4):1284-1294. |
| |
| 作者姓名： | 王博帅 蒲东川 李婷婷 牛雪峰 |
| |
| 作者单位： | 吉林大学地球探测科学与技术学院, 长春 130026 |
| |
| 基金项目： | 国家自然科学基金项目（41472243）；吉林大学研究生创新基金资助项目（101832020CX230） |
| |
| 摘 要： | 夜光遥感影像记录的城市灯光与人类活动密切相关，已广泛应用于城市信息提取。珞珈一号作为新一代夜光遥感数据源，比以往的夜光数据具有更高的空间分辨率和光谱分辨率，可以更清晰地表达城市建成区范围和内部结构。本文利用珞珈一号夜光遥感影像，通过人类居住指数（human settlement index， HSI）、植被覆盖和建筑共同校正的城市夜光指数（vegetation and build adjusted nighttime light urban index， VBANUI）及支持向量机（support vector machine， SVM）监督分类3种方法对长春市城市建成区进行提取，并与利用NPP/VIIRS（suomi national polar-orbiting partnership/visible infrared imaging radiometer suite）夜光遥感影像、采用同样方法得到的结果对比。结果显示：本文提出的VBANUI提高了传统植被覆盖校正的城市夜光指数（vegetation adjusted nighttime light urban index， VANUI）的提取精度，使用珞珈一号夜光遥感影像通过VBANUI提取的城市建成区结果最优，其Kappa系数为0.80，总体分类精度为90.74%；使用珞珈一号和NPP/VIIRS夜光遥感影像通过HSI按最佳阈值提取城市建成区的Kappa系数分别为0.75和0.72，总体分类精度分别为88.27%和86.54%；复合数据的SVM监督分类法中Landsat-NDBI、Landsat-NDBI-VIIRS、Landsat-NDBI-LJ和Landsat-NDBI-LJlog的Kappa系数分别为0.602、0.627、0.643和0.681，总体分类精度分别为81.11%、81.52%、82.25%和84.48%。研究结果表明：3种提取方法下，均为使用珞珈一号夜光遥感影像的结果优于使用NPP/VIIRS夜光遥感影像的结果，证明相比于NPP/VIIRS夜光遥感影像，珞珈一号夜光遥感影像更适用于城市尺度的建成区范围提取。
|
| 关 键 词： | 珞珈一号夜光遥感影像 NPP/VIIRS夜光遥感影像 城市建成区 VBANUI 人类居住指数 支持向量机 |
| 收稿时间： | 2020-06-28 |
Mapping of Urban Built-Up Area of Changchun City Based on Multi-Source Remote Sensing Images |
| |
| Wang Boshuai,Pu Dongchuan,Li Tingting,Niu Xuefeng.Mapping of Urban Built-Up Area of Changchun City Based on Multi-Source Remote Sensing Images[J].Journal of Jilin Unviersity:Earth Science Edition,2021,51(4):1284-1294. |
| |
| Authors: | Wang Boshuai Pu Dongchuan Li Tingting Niu Xuefeng |
| |
| Institution: | College of GeoExploration Science and Technology, Jilin University, Changchun 130026, China |
| |
| Abstract: | Nighttime remote sensing images have been widely used in the extraction of city information,as nighttime light data are closely related to human activities. As a new generation of noctilucent remote sensing data source, LJ1-01 has higher spatial and spectral resolutions, and can express the scope and internal structure of urban built-up areas more clearly. In this study, LJ1-01 nighttime light remote sensing images were used to extract the urban built-up area of Changchun City based on human settlement index (HSI),vegetation and build adjusted nighttime light urban index(VBANUI), and support vector machine(SVM) supervised classification. The results were compared with those obtained by suomi national polar orbiting partnership/visible infrared imaging radiometer suite (NPP/VIIRS) using the same method. It showed that the extraction accuracy of the traditional vegetation adjusted nighttime light urban index(VANUI) is improved by VBANUI proposed in this paper. Among them, the urban built-up area extracted by VBANUI using LJ1-01 has the best result, with the Kappa coefficient of 0.80 and the overall classification accuracy of 90.74%; The optimal Kappa coefficients of urban built-up areas extracted by HSI using LJ1-01 and NPP/VIIRS are 0.75 and 0.72, respectively, and the overall classification accuracy is 88.27% and 86.54%, respectively; The kappa coefficients of Landsat-NDBI composite data, Landsat-NDBI-VIIRS composite data, Landsat-NDBI-LJ composite data, and Landsat-NDBI-LJlog composite data in SVM supervised classification are 0.602, 0.627, 0.643, and 0.681, respectively, and their overall classification accuracy is 81.11%, 81.52%, 82.25%, and 84.48% respectively. The results show that the LJ1-01 nighttime image is the best among the three extraction methods. This study proves that the LJ1-01 night light remote sensing image is more suitable for urban level built-up area extraction than NPP/VIIRS night light remote sensing image. |
| |
| Keywords: | LJ1-01 nighttime light remote sensing image NPP/VIIRS nighttime light remote sensing image urban built-up area VBANUI human settlement index support vector machine |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《吉林大学学报(地球科学版)》浏览原始摘要信息 |
| 点击此处可从《吉林大学学报(地球科学版)》下载免费的PDF全文 |
|
