The quality of ecological environment can reflect the impact of human society sustainable development on human living environment. In this paper, based on Google Earth Engine (GEE) platform, we calculated the remote sensing ecological index of Shaanxi Province from 1999 to 2018. On this basis, we analyzed the correlation between each index factor and remote sensing ecological index (RSEI), and evaluated the spatial change of ecological environment quality in the study area. The conclusions were as follows:(1) In the model of RSEI, the RSEI was negatively correlated with both heat index and the surface dryness index, and positively correlated with the vegetation index and humidity index. The average value of surface dryness index accounted for the largest proportion of RSEI, while that of heat index was the smallest. (2) The overall ecological environment remote sensing index of Shaanxi Province showed a fluctuation upward trend of ‘up-down-up-down’, rising in 1999, falling in 2003, rising in 2005 and falling in 2012. (3) The quality of ecological environment in Shaanxi Province rose from north to south. The quality of Huanglong Mountain and Ziwuling Forest Park in northern Shaanxi was higher than that in Guanzhong Plain, and urban areas were lower than the surrounding areas. (4) From 1999 to 2003, the proportion of improved area and reduced area of ecological environment quality was close to 12%, and the reduced area was mainly in Guanzhong Plain. From 2003 to 2005, the area of reduced ecological environment quality accounted for 23.8%, which was mainly distributed in the Mu Us desert in northern Shaanxi. From 2005 to 2012, the proportion of ecological environment quality increased to 29.71%, mainly in northern Shaanxi. From 2012 to 2018, the reduction ratio was 20.91%, mainly in the Qinling and Guanzhong areas in southern Shaanxi. (5) In the past 20 years, the ecological quality of Guanzhong urban areas decreased significantly, while that of Northern Shaanxi increased significantly. In the slope diagram of regression equation, the scale in city area was mainly from -0.015 to -0.010. (6) The linear graph and the normal test distribution graph showed that the RSEI with each index fitted linear and normal distribution, therefore it could be used to study the Pearson linear relationship. (7) Both the linear fitting degree and the slope of the city were small. The surface dryness index, the vegetation index correlation degree and slope value were high in the arid area of Northern Shaanxi. Compared with the surface dryness index, the fitting degree of the vegetation index in the loess hilly region of Northern Shaanxi was slightly higher.