The invention belongs to the field of agricultural remote sensing, and relates to a crop growth monitoring method based on a leaf area index (LAI) inverted through an inverse Gaussian process. The method comprises the following steps: obtaining consistent sentinel No.2 surface reflectivity, and extracting a spatial distribution map of crops; obtaining PROSAIL simulated crop spectrums under different backgrounds and converting the simulated crop spectrums into waveband spectrums; analyzing the principal components of the simulated surface reflectivity under a black soil background to obtain pure canopy reflection characteristics; analyzing the principal components of the actually measured soil spectral reflectivity to obtain pure soil reflection characteristics; linearly decomposing the simulated surface reflectivity containing soil information to obtain simulated pure canopy reflectivity; learning the mapping relationship between simulated pure canopy reflectivity and LAI in a simulative way through a Gaussian process to obtain a trained Gaussian process model; and linearly decomposing a sentinel No.2 image to obtain pure canopy reflectivity, selecting and inputting a simulated pure canopy reflectivity closest to the pure canopy reflectivity to the trained Gaussian process model to obtain LAI, and judging the current crop growth by taking the previous annual average LAI trajectory as a benchmark. The method of the invention avoids the problem of multi-parameter same effect and is of high precision.