Abstract: Vegetation cover classification using multispectral remote sensing imagery is a hot research area, in which various new methods emerge endlessly. On the basis of reading a large number of references, the authors summarized in this paper the status and progress of vegetation cover classification with multispectral remote sensing imagery , analyzed advantages and disadvantages, adaptation and application of each vegetation classification feature and method, pointed out current difficulties and challenge, and predicted future development trend. The analysis suggests that future vegetation cover classification of multispectral remote sensing imagery needs not only innovation of classifier in the aspects of improvement of automation, efficiency, learning rate, adaptation and robustness, but also feature mining of vegetation classification. For the purpose of enhancing such aspects as using feature reparability and fusing multisource data, the adoption of multi -temporal images and the tapping of more new features in vegetation classification seem to be future trends.