“Leaf Area Index (LAI) is a key parameter that characterizes crop canopy structure and growth. Accurate and timely monitoring of LAI using remote sensing technology is of great significance for field water and fertilizer management, food security, and agricultural production potential assessment. The red edge, as a sensitive spectral band indicating changes in leaf physiology and canopy structure, has been configured by various medium to high resolution (10-30m) satellite sensors and widely used in crop parameter inversion, providing a new opportunity to further improve the accuracy of crop LAI inversion. However, there are significant differences in the methods of applying the red edge band to LAI inversion in existing research, and due to the different research areas, it is still unclear how to effectively utilize the red edge band to improve the accuracy of LAI inversion. Based on this, this study adopts a hybrid method combining PROSAIL model and machine learning algorithm as the inversion strategy. Using Sentinel-2 images containing three red edge bands and ground LAI measurement data of major grain crops (rice, wheat, and corn) provided by the National Ecosystem Observation and Research Network, a crop LAI inversion algorithm coupled with red edge band features was constructed by optimizing the machine learning model and band combination, and systematic evaluation was carried out in different scenarios. The results showed that the multi-layer perceptron (MLPR) had the best fitting effect on LAI and multi band reflectance, and the introduction of red edge bands could effectively improve the accuracy of LAI inversion. Among them, the joint introduction of red edge 1 (RE1) and red edge 3 (RE3) had the best inversion effect (R2=0.784, RMSE=0.826). Compared with the Z1 combination without red edge bands (Green+Red+NIR+SWIR1+SWIR2), R2 increased by 4.9% and RMSE decreased by 15.6%. At the same time, introducing the red edge band not only reduces the systematic bias in LAI inversion, but also effectively mitigates the impact of saturation effects in the high-value range of LAI (4”