针对在航空航天遥感领域中使用的CCD相机对景物进行成像时,由于像元尺寸的限制导致图像低分辨率的问题和传统的图像重构算法的不足,提出了小波神经网络映射图像重构过程非线性特性的方法。实验表明,用小波神经网络对低分辨的图像重构,可以得到高分辨率图像。通过仿真实验得到高分辨率和高信噪比的超分辨结果。 For the CCD camera used in the field of aerospace remote sensing to image the scene, due to the limitation of the pixel size, the low resolution of the image and the shortage of the traditional image reconstruction algorithm are proposed. The wavelet neural network mapping image reconstruction process Non-linear characteristics of the method. Experiments show that using wavelet neural network to reconstruct low resolution images, high resolution images can be obtained. Through the simulation experiment, the superresolution results of high resolution and high signal to noise ratio are obtained.