基于关联规则的裙带菜筏式养殖遥感识别方法

Wakame Raft Interpretation Method of Remote Sensing based on Association Rules

筏式养殖裙带菜监测是大连地区遥感监测的重要应用之一,寻求一种快速准确的识别方法从而及时掌握裙带菜养殖的分布范围、面积、数量等基础信息,对地方政府调控产业发展起到关键作用。引入数据挖掘技术,利用Landsat TM数据源,开展基于关联规则的裙带菜筏式养殖信息提取方法研究,并在大连金石滩附近海域进行实验论证。应用数据挖掘关联规则分析方法,找出筏式养殖区和海水分类的知识规则,并通过该规则构建分类树提取养殖信息,最后结合筏式养殖形态特征对得到的分类结果进行噪声去除后,得到最终的分类结果。结果表明:该方法的总体识别准确度可达80%,与最大似然分类结果相比识别准确度提高11.64%,该方法能够满足监测基本需求,具备一定的可行性。

The wakame raft cultivation monitoring is an important part of remote sensing monitoring in Dalian. It＇s very necessary for local government to find out the position, range, quantity of the raft cultiva tion and grasp industrial development information by seeking an accurate and fast interpretation method. The wakame raft cultivation interpretation method is studied, by using Landsat TM data source,in the view of data mining association rule analysis technique,in Dalian Jinshitan Beach sea area, and the raft cultivation interpretation workflow is proposed based on association rules. The wakame raft cultivation area and sea classification knowledge are identified by using data mining association rules analysis. The raft cultivation could be interpreted with the classification rules. In order to eliminate the noise spots, in the context of morphological characteristics and cultivation habits of the wakame floating raft, the final classification bina- ry results is obtained, The method of the overall recognition accuracy reaches 80 %, which is compared with visual interpretation result, the interpretation accuracy rate is improved by 11.64 %. The method can meet the basic requirements,which is feasible for monitoring.