| 基于滑动分割算法的我国耕地熟制识别研究 |
| |
| 引用本文： | 刘爽,马欣,李玉娥,张平究.基于滑动分割算法的我国耕地熟制识别研究[J].资源科学,2014,36(9):1969-1976. |
| |
| 作者姓名： | 刘爽 马欣 李玉娥 张平究 |
| |
| 作者单位： | 安徽师范大学国土资源与旅游学院, 芜湖 241003;中国农业科学院农业环境与可持续发展研究所/农业部农业环境与气候变化重点实验室, 北京 100081;中国农业科学院农业环境与可持续发展研究所/农业部农业环境与气候变化重点实验室, 北京 100081;中国农业科学院农业环境与可持续发展研究所/农业部农业环境与气候变化重点实验室, 北京 100081;安徽师范大学国土资源与旅游学院, 芜湖 241003 |
| |
| 基金项目： | 国家科技支撑项目：“气候变化对农业生产影响与风险评估技术”（编号：2012BAC19B01）；中国清洁发展机制基金：“我国农业领域适应气候变化技术清单与研发部署研究”（编号：1113113） |
| |
| 摘 要： | 熟制时空格局的正确识别对评估粮食产量的变化及其原因和农业发展的科学决策都有非常重要的意义,卫星遥感监测是获取区域和全国尺度熟制格局的有效手段。本文在对启发式分割算法改进的基础上行成了基于作物生长周期的滑动分割算法,并首次运用到耕地熟制的识别提取。针对NDVI时间序列曲线特征,在不引入熟制分区和物候等信息,仅以土地利用为辅助数据的前提下识别了1982-2006年我国耕地熟制格局。结果表明,本方法监测结果与统计数据和前人监测结果均呈显著相关性,为正相关关系,相关系数分别为0.77和0.93,均通过0.001水平的显著性检验;熟制的分布为区域性和复杂性并存,其空间分布规律性显著但熟制区内复杂性较明显;一熟区总体变动不明显但区域范围在逐步缩小,两熟区整体向北、向西扩展,向北扩展趋势明显,三熟区缓慢扩大并出现零星区域的北移、西移。
|
| 关 键 词： | 耕地熟制 NDVI 滑动分割算法 识别 |
Farmland Cropping System Identification in China Based on a Sliding Segmentation Algorithm |
| |
| LIU Shuang,MA Xin,LI Yu''e and ZHANG Pingjiu.Farmland Cropping System Identification in China Based on a Sliding Segmentation Algorithm[J].Resources Science,2014,36(9):1969-1976. |
| |
| Authors: | LIU Shuang MA Xin LI Yu'e and ZHANG Pingjiu |
| |
| Institution: | College of Territorial Resources and Tourism, Anhui Normal University, Wuhu 241003, China;Institute of Environment and Sustainable Development in Agriculture, Chinese Academy of Agricultural Sciences; and Key Laboratory of Agro-Environment and Climate Change, Ministry of Agriculture, Beijing 100081, China;Institute of Environment and Sustainable Development in Agriculture, Chinese Academy of Agricultural Sciences; and Key Laboratory of Agro-Environment and Climate Change, Ministry of Agriculture, Beijing 100081, China;Institute of Environment and Sustainable Development in Agriculture, Chinese Academy of Agricultural Sciences; and Key Laboratory of Agro-Environment and Climate Change, Ministry of Agriculture, Beijing 100081, China;College of Territorial Resources and Tourism, Anhui Normal University, Wuhu 241003, China |
| |
| Abstract: | Multiple cropping is an important feature of China's farmland cropping system and an important way to improve food production. Correct recognition of cropping spatio-temporal patterns is important to assess changes, understand grain yield, and the development of agriculture scientific decision-making. Here we present an improved heuristic segmentation algorithm and sliding segmentation algorithm based on crop growth cycles, and apply this to the identification of farmland cropping system extraction. NDVI time series curve characteristics analysis for the 1982-2006 cropping pattern of cultivated land in China was used. We found that the statistical check between monitoring results, statistical data and previous monitoring were positively correlated;the correlation coefficients were 0.77 and 0.93. The distribution of cropping systems is regional and complex, the space distribution of the main cropping system is regular, but the cropping system shows complexity within the cropping area. Overall changes in single cropping systems are not obvious but the regional scope is gradually narrowing. Double cropping systems are moving northward and westward;and the triple cropping system is slowly expanding northward and westward. |
| |
| Keywords: | cropping system NDVI Sliding Segmentation Algorithm identify |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《资源科学》浏览原始摘要信息 |
| 点击此处可从《资源科学》下载免费的PDF全文 |
|
