Non-contact measurement and monitoring of plant growth using computer vision has great significance in improving the level of intelligent control in greenhouses. A computer vision system which can be used to monitor the growth of cucumber seedlings in a near-real-time and non-destructive way was developed in an experimental greenhouse. Image analysis software which can be used to measure the top projected leaf area(TPLA) and shoot height of the seedlings was developed using VC++6.0. By continuously monitoring the change of TPLA of controlled and treated seedlings, it was found that the change rate of TPLA is a good indicator of fertilizing status for cucumber seedlings. Compared with the manual measured average shoot height, the image processing results yielded a correlation of R²=0.927. These results indicate that computer vision application in greenhouse plant growth monitoring is feasible and promising.