In order to rapidly acquire winter wheat growing information in the field, the retrieval method of vegetation coverage index(VCI) was researched based on multi-spectral imaging technique and imaging processing technology. Firstly, a 2-CCD multi-spectral image monitoring system was used to acquire the canopy images. The system was based on a dichroic prism, allowing precise separation of the visible (RGB) and near-infrared (NIR) band. Secondly, after the image smoothing using adaptive smooth filtering algorithm, the canopy image of winter wheat was segmented. HSI color model and automated threshold method were used to segment the RGB and NIR image respectively. The hue threshold was ［ π/4 , 6π/5］. The segmented results of RGB and NIR were combined to improve the segmentation accuracy and the VCI was calculated. Thirdly, the image parameters were abstracted based on the original visible and NIR images including the average gray value of each channel( A R, A G, A B ) and near-infrared ( A NIR ),the vegetation indices (NDVI, NDGI, RVI, DVI) which were widely used in remote sensing, and the H average value of canopy. The correlation analysis results showed that the correlation coefficients between vegetation indices and VCI were above 0.90. As a result, the retrieving multiple linear regressions (MLR) model was built by using NDVI, NDGI, RVI and DVI with R 2 c =0.948 and R 2 v = 0.884. It was feasible to diagnose vegetation coverage in the field and indicate the growth status.