| 摘要: |
| 水稻种植面积监测是当前农业土地变化科学的热点问题,但运用遥感技术对水稻种植面积精确实施监测一直是难点。中分辨率遥感影像能够满足我国大面积水稻作物监测,成为业务化运行的主要数据源。为此,该研究尝试以中分辨率 TM影像为数据源,结合神经网络和面向对象 (SVM)两种算法对对黑龙江省富锦市2010年两期不同时相影像分别进行水稻分类提取,并对分类结果进行滤波处理及混淆矩阵精度评定。结果表明:(1)在高纬度单季稻生长区,面向对象分类算法的精度显著高于神经网络的分类精度,水稻用户精度和生产者精度在6月份分别高0.55%、1.37%,在8月份分别高0.62%、2.34%;(2)对神经网络分类的结果进行 Majority滤波处理,在一定程度上可以改善水稻分类的精度,水稻用户精度和生产者精度在6月份分别提高0.14%、0.5%,在8月份分别提高1.56%、1.43%;(3)选取关键水稻物候期的遥感影像获取水稻种植面积的精度更高,返青期水稻提取精度要高于乳熟期,其中神经网络算法的水稻用户精度及生产者精度分别提高 2.67%、3.45%;面向对象算法的水稻用户精度及生产者精度分别提高 2.6%、2.48%。未来需要重点考虑建立全国水稻物候历信息、面向对象算法中自动化最优尺度分割方法来提高水稻分类的精度。 |
| 关键词: 水稻 神经网络 面向对象分类 TM影像 |
| DOI：10.7621/cjarrp.1005-9121.20140105 |
| 分类号: |
| 基金项目: |
|
| CONTRASTING TWO CLASSIFICATION METHODS IN MAPPING PADDY RCIE USING THE MID-RESOLUTION TM IMAGES |
|
Li Zhipeng1, Li Zhengguo1, Liu Zhenhuan2, Wu Wenbin1, Tan Jieyang1, Yang Peng1
|
|
1.Key Laboratory of Agri-informatics, Ministry of Agriculture/Institute of Agricultural Resources and Regional Planning, Chinese Academy of
Agricultural Sciences, Beijing100081;2.Geography and Planning School of Sun Yat-sen University, Guangzhou 510275
|
| Abstract: |
| Monitoring the rice area has become a hot issue of the present agricultural land change study. However, it's uneasy to accurately monitor rice area using remote sensing technology. Based on two phase mid-spatial resolution TM images, this paper monitored the rice area in Fujin County, Heilongjiang Province using the methods of object oriented classification and Artificial Neural Network (ANN). The conclusions were as follows: Firstly, the accuracy using the method of object-oriented classification was higher than that using ANN classification method. It was 055% and 137% higher for rice user and producer accuracy on June, and 062% and 234% on August. Secondly, after the process of Majority analysis, the rice classification accuracy would increase by 014% and 05% on June, and 156% and 143% on August for rice user and producer accuracy using the method of ANN. Thirdly, choosing the proper time images can get a higher accuracy rice area, and the accuracy obtained from returning green stage was better than that from the milk ripe stage, the rice user accuracy and producer accuracy can increase 267% and 345% using ANN method and can increase 26% and 248% using the method of object oriented classification. In future, it should improve the classification accuracy through building long time series of rice phonological calendar and using the method of object oriented automatized segmentation scale. |
| Key words: paddy rice ANN object-oriented classification TM image |
