| 引用本文: | 刘克宝,刘述彬,陆忠军,宋茜,刘艳霞,张冬梅,吴文斌.利用高空间分辨率遥感数据的农作物种植结构提取[J].中国农业资源与区划,2014,35(1):21~26 |
| |
|
| 本文已被：浏览 4923次 下载 2431次 |
码上扫一扫！ |
|
|
| 利用高空间分辨率遥感数据的农作物种植结构提取 |
|
刘克宝1, 刘述彬1, 陆忠军2, 宋茜3, 刘艳霞4, 张冬梅5, 吴文斌6
|
|
1.黑龙江省农业科学院遥感技术中心，哈尔滨 150086;2.黑龙江省农业科学院遥感技术中心，哈尔滨 150087;3.黑龙江省农业科学院遥感技术中心，哈尔滨 150088/中国农业科学院农业资源与农业区划研究所, 北京 100081;4.黑龙江省农业科学院遥感技术中心，哈尔滨 150089;5.黑龙江省农业科学院遥感技术中心，哈尔滨 150090;6.中国农业科学院农业资源与农业区划研究所, 北京 100081
|
|
| 摘要: |
| 农作物种植结构是掌握粮食种植面积和产量的重要前提,也是进行作物结构调整与优化的依据。该研究以黑龙江肇东市为研究区域，以高空间分辨率RapidEye影像为遥感数据， 基于最大似然监督分类方法提取了肇东市2011年农作物种植结构空间分布，利用地面样方调查数据进行了线状及细小地物扣除系数计算，实现遥感提取的农作物种植面积的精细提取，然后从面积数量和空间位置两个方面对遥感提取的农作物种植结构进行了精度评价。研究结果表明，利用RapidEye数据提取的农作物种植面积数据总体精度为97.00%,位置精度为96.15%,高空间分辨率数据在农作物种植结构遥感提取中具有重要潜力,线状及细小地物扣除系数可以有效减小线状及细小地物对高分提取的农作物种植结构的精度。 |
| 关键词: 高空间分辨率 农作物种植结构 扣除系数 精度 |
| DOI：10.7621/cjarrp.1005-9121.20140104 |
| 分类号: |
| 基金项目: |
|
| EXTRACTION ON CROPPING STRUCTURE BASED ON HIGH SPATIAL RESOLUTION REMOTE SENSING DATA |
|
Liu Kebao,Liu Shubin,Lu Zhongjun,Song Qian,Liu Yanxia,Zhang Dongmei,Wu Wenbin
|
|
1.Remote Sensing TechniqueCenter of Heilongjiang Academy of Agricultural Sciences,Haerbin 150086;2.Remote Sensing TechniqueCenter of Heilongjiang Academy of Agricultural Sciences,Haerbin 150086/Institute of Agricultural Resources and Regional Planning,Chinese Academy of Agricultural Sciences,Beijing 100081;3.Institute of Agricultural Resources and Regional Planning,Chinese Academy of Agricultural Sciences,Beijing 100081
|
| Abstract: |
| Crop planting structure is important for crop yield and acreage estimation,and it is also the basis of structural adjustment and optimization.To explore the potentials of high spatial resolution data in crop structure extraction,taking Zhaodong City in Heilongjiang Province as an example,this paper used a conventional supervised classification to obtain a crop structure map from three
RapidEye images.A prior ground truth database was created with 10 sample plots(1km*1km)randomly distributed over the study area for two main purposes:first,to set up interpretation keys for image classification;second,to minimize the effects caused by linear features and small
ground objects,which would improve the accuracy of crop area estimation.Classification results showed that the final crop structure map had an overall acuracy of 97.00% and the position accuracy reaches 96.15%,which was much higher than the application of middle /low resolution remote sensing images.This experimental case study implied that the accuracy of crop structure mapping can be improved significantly by the combination of high spatial resolution images with detailed ground truth database. |
| Key words: high spatial resolution cropping structure accuracy |
