The invention discloses a remote sensing image segmentation method which comprises the following steps: carrying out noise reduction preprocessing on a collected remote sensing image to obtain a denoised remote sensing image; respectively carrying out gray processing and super-pixel segmentation on the remote sensing image after de-noising; extracting textural features and color features based onthe superpixel segmentation result; performing geometric feature extraction on the remote sensing image subjected to gray processing; fusing the texture features, the color features and the geometricfeatures to obtain image segmentation features; based on the image segmentation characteristics, performing image filtering by adopting mean shift to obtain a preliminary segmentation result; according to the image segmentation method, subjecting the preliminary segmentation result to region merging to obtain the final segmentation result, noise processing is fully considered, and a multi-featurefusion method is adopted to segment the image, so that the image segmentation effect can be fully ensured.