摘要
随着城镇化进程的加快,不透水面逐渐取代自然地表景观成为重要土地利用类型。利用遥感技术可量化不透水面的时空变化。本研究通过Google Earth Engine云平台,利用Landsat8影像,参考刘良云土地利用产品数据,采用ROC曲线作为精度评价指标,定量对比分析PII,RCI,RRI,NDBI等8种指数在长春市不透水面提取中的效果。研究表明:RRI和UI指数识别精度分别为85%和81%,能够有效提取长春市不透水面信息,并为城镇规划、生态资源管理和环境可持续发展提供科学参考。
Rapid economic development has driven the process of urbanization,making impervious surfaces replace natural landscapes and become a crucial land use type.Remote sensing technology can quantify the spatiotemporal variations of impervious surfaces.This study,utilizing the Google Earth Engine cloud platform and Landsat8 imagery and referencing Liu Liangyun's land use product data,employed the ROC curve as the accuracy evaluation metric to quantitatively compare and analyze the effectiveness of eight indices(PII,RCI,RRI,NDBI,etc.)in extracting impervious surfaces in Changchun city.The study revealed that the RRI and UI indices have recognition accuracies of 85%and 81%,respectively,effectively extracting impervious surface information in Changchun city and providing scientific references for urban planning,ecological resource management,and environmental sustainability.
作者
孙柯楠
马玥
SUN Ke-nan;MA Yue(School of geomatics and prospecting engineering Jilin Jianzhu university 130118,China)
出处
《吉林建筑大学学报》
CAS
2024年第5期59-65,共7页
Journal of Jilin Jianzhu University
基金
吉林省教育厅科学技术研究项目(JJKH20210269KJ).
