Abstract: As remote sensing images that only have three visible light bands: red (R), green (G) and blue (B) cannot use normalized difference vegetation index (NDVI) method to extract vegetation land cover information, a new method based on hue saturation and lightness color model is proposed. The image is transformed from RGB color space to HSL color space by using the self-developed system, and normalized hue lightness vegetation index (NHLVI) is constructed. By analyzing the characteristics of vegetation and non-vegetation information in HSL color space, and the characteristics of NHLVI, H, S, L, R, G, B, the vegetation information is extracted by cooperating with NHLVI and S indexes, and the non-vegetation information in the result is eliminated by using B component feature, so as to realize vegetation information extraction and improve the extraction accuracy. The results show that this method could improve the accuracy of visible light vegetation extraction and the applicability of the method by adding S index to the existing NHLVI index method.

Key words: vegetation extraction, NHLVI, hue, saturation, lightness