基于多分类器组合的高分辨率遥感影像目标识别

Target Identification from High Resolution Remote Sensing Image Based on Multiple Classifier Combination

研究多分类器组合的形式及其在目标识别中的应用。设计和实现了基于多分类器组合从高分辨率遥感影像中提取目标的具体方法。以楼房的识别与提取为例,选用QuickBird高分辨率影像,利用投票法、加权求和、完全一致规则及或运算合并等多分类器组合方法进行目标识别,比较了多分类器系统与单一分类器的精度与性能,从理论和实验两方面分析了多分类器组合用于目标识别的精度影响因素,提出了相应的改进策略。实验表明,在采用合理的组合规则下,多分类器组合能使目标识别结果达到高识别率和高置信度。

Multiple classifier combination is used to target identification from high resolution remote sensing image. Taking building identification as an example and using QuickBird image as the data source, voting combination, weighted summation and overall consistency criteria are used to combine multiple classifiers, and then the accuracy and performance of multiple classifier system and individual classifiers are compared from theoretical and experimental aspects. The experiment results show that multiple classifier combination can improve the performance of target identification with good identification rate and high confidence degree if reasonable combination strategy is selected.