论文部分内容阅读
以2013年9月的高分一号(GF-1)遥感影像为数据源,以宁夏盐池县为例,采用面向对象的分类方法对城镇绿地进行专题信息提取,并从景观面积与优势度、景观多样性和景观分维数等方面对盐池城镇绿地景观格局进行分析。结果表明:GF-1遥感影像在盐池城镇绿地分类中具有较高的精度,分类效果好,其总体分类精度为90.80%,Kappa系数为0.796 1,完全可满足西部小城镇绿地景观信息的提取需求。盐池县的城镇绿地景观格局表现出3方面的特征:城镇绿地景观合理性欠佳,景观类型单一,不同类型的绿地面积差异大;绿地景观整体较破碎,最为突出的是附属绿地,斑块密度达到53.92个·hm~(-2);城区绿地分维数较低,缺乏自然绿地斑块,人为修造严重。
Taking the GF-1 remote sensing image of September 2013 as the data source and Yanchi County of Ningxia as an example, this paper uses the object-oriented classification method to extract thematic information from the urban green space. Based on the landscape area and dominance, Landscape diversity and landscape fractal dimension of Yanchi urban green space landscape pattern analysis. The results show that the classification accuracy of GF-1 remote sensing image is high, and the classification accuracy is good. The overall classification accuracy is 90.80% and the Kappa coefficient is 0.796 1, which can fully meet the demand of green space landscape information extraction in western towns . The landscape pattern of urban green land in Yanchi County shows three characteristics: the landscape of urban green land is not good, the landscape type is single, and the difference of green land area varies greatly; the landscape of green land is fragmented, the most prominent is the afforestation green space, the patch density Reaching 53.92 · hm -2. The fractal dimension of urban green space is low, the natural green patches are lacking, and man-made construction is serious.