一种基于模糊融合的遥感图像目标识别方法

数据融合利用多传感器的信息,克服了单一传感器信息不完整、不精确、不确定的缺点,因此广泛应用于目标识别中,该文提出了一种基于模糊融合的遥感图像目标识别的新方法。首先在单源图像上提取可疑目标,然后根据目标在不同类型图像上的成像特点,选择合适的目标特征,充分考虑到各特征的重要程度,把模糊隶属度函数和模糊密度结合起来,最后利用特征层模糊融合对目标的身份进行判定。此方法应用在实际目标的识别中,取得了很好的效果。

An Approach of Object Recognition on Remote Sensing Images Based on Fuzzy Fusion

Data Fusion uses information of many sensors to overcome the imperfection,imprecision,uncertain of only one sensor.It is now widely used in object recognition.This paper advances a new approach of object recognition for remote sensing image based on fuzzy fusion.We firstly extract suspicious objects from each image,then choose proper feature according to the image type and consider the importance of every feature on the recognition of certain object,use subjection function and fuzzy density together on fusion.At last feature level fuzzy fusion results are used to identify objects.Ideal results are obtained when this approach is applied to the practice.