Analysis on correlation between information capacity of NDVI and fractal dimension at different landscape zones

Multi-spectral information of remote sensing offers a variety of spatial and temporal scales vegetation indexes for the surface environmental of the earth. Normalized vegetation index (NDVI), which is known as the best instruction factor of vegetation, is most widely used now. Information capacity based on multi-dimensional histogram is an evaluation model of grayscale digital image quality. Its size is connected with the meaningful gray level of image. While the meaningful gray level of NDVI image corresponds to the space complexity of the regional land cover. This paper studied the scientific connotation of NDVI information capacity, and analyzed the relationship between the regional land cover characteristics, NDVI information capacity and fractal dimension of remote sensing image by establishing experimental plots; explored the spatial pattern of NDVI information capacity, taking the spatial variation characteristics of NDVI Information capacity in different regions as a starting point. The results show that the four indicators—NDVI, TM information capacity, NDVI information capacity and fractal dimension reflect the complexity of the surface environmental, vegetation and other information of different areas, and a new factor— NDVI information capacity can be used for the study of spatial structure of regional landscape, thus it is useful to deepen the application of remote sensing data.