In order to explore the rapid and non-destructive estimation method of chlorophyll content of maize, this paper used multi spectral image technology to diagnose the content of maize chlorophyll at the seedling stage.In the field experiments, the canopy multispectral images of field corn seedling were acquired by 2-CCD multi spectral image acquisition system, and the multi spectral images of diffuse reflectance gray board were collected synchronously.In order to eliminate the influence of light and temperature on the quality of image acquisition, accurately different light conditions of corn canopy image data conversion for the foliar reflectance data and calibration tests using a four different gradations of color satisfy the Lambertian surface diffuse gray boards, corn leaf spectral reflectance and image gray value between the linear formula was established.And the correction formula of gray value of corn canopy image was established by the multi spectral images of the diffuse reflectance gray board in the field experiment.The R, B, NIR and G of the maize canopy were extracted from the canopy of Maize at seedling stage, and the average gray values of the four bands were normalized.The average gray value of NDVI, G, NIR,DVI four bands were obtained by the gray value correction formula, and the average reflectance of 4 bands of LS-SVM, B, RVI and R were calculated.The fitting model of the vegetation index and chlorophyll content index were calculated by using the least squares support vector machine (NDGI).The results show that the fitting correlation coefficient is 0.71, and the fitting results of the vegetation index NDVI, RVI and Maize Canopy Chlorophyll content index are.It is proved that this method is feasible to establish the model of reflectance inversion and correction by diffuse reflectance gray board, which is a fast and nondestructive method to provide support for the content of chlorophyll in maize seedling stage.