One of the most important applications of remote sensing is to classify vegetation based on the vegetation. indices. Vegetation is classified according to the components such as the shape and the size of leaf, the amount of water inside the leaf.. LAI and NDVI, which are calculated from the satellite image, are two of the important indices for vegetation classification. This paper will compare and evaluate the precision of the vegetation classification with LAI and NDVI deriving from Landsat5 TM satellite image on the region of Thuong Tin district, Hanoi, Vietnam, where rice, corn and other agricultural vegetation are planted. The result shows that the value of LAI changes very little from the dry to rainy season, while the value of NDVI is clearly different. Furthermore, because the interested region is a plain region, NDVI is not affected by topography and shape. Consequently, in order to classify the vegetation in the interested region, NDVI should be used rather than LAI.