摘要
用高分影像数据,基于支持向量机的分类方法,提取研究区泥石流迹地与居民地信息,进行样本精度验证后,利用多环缓冲区、核密度估计、叠置分析等空间分析和统计手段,探讨泥石流迹地与居民地的空间关联关系.结果表明,高分数据制图的总体精度达到90.69%, Kappa系数为0.86;大部分泥石流迹地与居民地空间分布呈负相关,约10.85%的居民地与泥石流活动的危险区空间相交,活动区域内的人民生命财产安全存在较大隐患,同时居民密集地区的人类生产生活活动会加剧生态脆弱区的泥石流活动.
With high-resolution image data based on the method of support vector machine classification,information on debris flows of populated places in the study area was extracted, and then the accuracy of the sample verified. Combining a spatial analysis and spaces such as multi-ring buffer and kernel density estimation with an overlay analysis the spatial relationship between debris flows and residential areas was studies. The results showed that the overall accuracy of high-scoring image data mapping reached90.69%, and the Kappa coefficient was 0.86, which could serve for a post-study. Most of the debris flow sites were negatively correlated with the spatial distribution of residential areas, but about 10.85% of the residential areas intersected with the dangerous areas of debris flow activities. There was a big hidden danger with the safety of people’s lives and property in the active areas, meanwhile human production in densely populated areas had aggravates mudslide activities in ecologically fragile areas.
作者
张良洁
甘淑
袁希平
苏文豪
Zhang Liang-jie;Gan Shu;Yuan Xi-ping;Su Wen-hao(Faculty of Land Resource Engineering,Kunming University of Science and Technology,Kunming 650093,China;Surveying and Mapping Geo-informatics Technology Research Center on Plateau Mountains of Yunnan Higher Education,Kunming 650093,China)
出处
《兰州大学学报（自然科学版）》
CAS
CSCD
北大核心
2021年第1期122-128,共7页
Journal of Lanzhou University(Natural Sciences)
基金
国家自然科学基金项目(41561083,41861054)。
