遥感融合图像分类精度的研究

Study of Classification Accuracy of RS Fused Image

把不同空间分辨率的TM和IRS遥感图像进行融合,综合了不同传感器数据所提供的信息,增强了图像的清晰度,改善了解译效果。对遥感融合图像进行分类,分类精度达97.90%,效果优于TM图像(分类精度为89.39%)。

The fused remote sensing image could merge two optical image data of different resolutions-a high spatial resolution panchromatic image and a low spatial resolution multi-spectral image.It could synthesize information from dif-ferent remote sensor.The sign may be strengthened in the fused image.The classification Accuracy based on the multi-layer perception neural networks for the fused RS image is higher than TM image.The classification accuracy for the fused RS image is up to97.90%,the accuracy of TM image is89.39%.