遥感影像多目标优化信息提取模式研究

根据运筹学中基于层次分析法的多目标决策原理和遥感分层分类法理论,建立遥感分类方案优选模型,构建了遥感影像多目标优化信息提取模式.实现在遥感动态监测过程,应用该模式提取遥感信息,从时间、成本、精度、技术等多个目标获得最佳综合效用.以闽江口湿地农业遥感动态监测为例,对该遥感信息提取模式进行初步应用研究.研究结果表明,遥感影像多目标优化信息提取模式,允分利用了目视解译、非监督分类、监督分类和基于知识分类不同分类方法对特定层次特定地物分离的优势,同时避免它们的不足,有效保证了信息提取的精度,同时提高整个遥感动态监测过程的效率.

Remote sensed information extracting model based on multiple-objective decision support theory: A case study of wetland agriculture

In this paper, an optimal model of selecting the classification method in the layered classification was established based on the Analytic Hierarchical Process (AHP); and information extraction model during the layering classification was developed. In the practice of monitoring the ground surfaces, this model applied to a large amount of image processing will facilitate to reach the best synthetic effectiveness of image classification in term of the satisfaction of multiple-goal requirements of time, costs, data accuracy, and techniques applied to imagery processing. This method was applied to the classification and dynamic monitoring of the wetland agriculture in Minjiang River Estuary. Research shows that the classification model takes advantage of the visual interpretation, unsupervised classification, supervised classification, and knowledge-based classification while avoiding their shortcomings. As a result, the classification increases the efficiency in the remote-sensed data process of dynamic monitoring ground surface features, at the same time, the accuracy of classification can be met.