It is well known that the protein content of rice grain is an indicator of taste of cooked rice in the countries who eat rice for staple. We examined that remote sensors can tell the difference of rice grain quality, investigating the relationships among leaf nitrogen content of the rice canopy, grain protein content, and normalized difference vegetation index (NDVI) values. NDVI is a widely used vegetation index which is dimensionless, radiometric measures of vegetation and an indicator of relative abundance and activity of green vegetation. We measured NDVI ([RNIR-RRED]/[RNIR+RRED]) and green NDVI ([RNIR-RGREEN]/[RNIR+RGREEN]) using two different active sensors (Greenseeker, NTech Inc. USA), leaf nitrogen content and grain protein content during the rice growing season at the experimental plots of NIAST in 2005. Greenseeker is a hand-held optical sensor unit which captures the light reflectance of plants using artificial light sources of near infrared and red or green wavelengths beamed from the very sensor. Four levels of nitrogen fertilizers (0, 7, 10, 13kg/10a) with the same amount of phosphorous and potassium content of the fertilizers were applied to the rice paddy fields with randomized complete black design. Leaf nitrogen content per unit ground area and grain protein content were analyzed with micro-Kjeldahl method. Correlation analysis between the green NDVI and leaf nitrogen content of the rice canopy were highly correlated with after heading stage (r=0.89**) rather than before heading stage (r=0.81**). Relationships of measured nitrogen content to estimated on was significantly correlated (r=0.92**). Leaf nitrogen content of rice canopy was significantly, positively correlated to grain protein content after heading stage (r=0.96**). The correlation coefficients between green NDVI and NDVI and grain protein content were 0.94 and 0.85, respectively, and green NDVI showed higher coefficient of determination than that of NDVI during the harvest stage (283 DOY). After heading stage, relationships between green NDVI and grain protein content showed the highest positive correlation in the harvest stage. The measured and estimated grain protein content showed good agreement (r=0.94**). Results from this study show that green NDVI values represented a significant positive correlation with grain protein content and can be used to estimate the grain protein during the harvest. Optical remote sensors (green NDVI/NDVI) appeared very effective to estimate leaf nitrogen content and grain protein content of the rice canopy.