LAUSR

Soil and water conservation is one of the key measures to improve the ecological environment of small watersheds and maintain the healthy life of the region. With the current method used to monitor the evolution of topographic and hydrological characteristics of a small watershed, the edge keeping index of the image and the signal-to-noise ratio of the image are low, the monitoring results are inaccurate, and there are some problems such as low edge keeping ability, poor denoising effect, and low monitoring accuracy. A monitoring method based on remote sensing and GIS for the evolution of topographic and hydrological characteristics of small watersheds is proposed. The hyperspectral data was transformed by the Principal Component Analysis (PCA). Group principal component images, use sparse representation method based on adaptive dictionary and dual-tree complex wavelet transform method to denoise principal component images with a small amount of information were the objectives of the paper; also the use the multi-scale wavelet transform to detect image edge; build a binary model of a pixel, extract vegetation index, and terrain factor based on a binary model of the pixel, and realize terrain and hydrological characteristics of the small watershed. The experimental results show that the proposed method has the high edge-preserving ability, good denoising effect, and high monitoring accuracy.