One of the main areas of study in the field of remote sensing is the semantic segmentation of high-resolution isolated sense images that precisely categorize objects. Existing deep learning-based code segmentation techniques require a large quantity of annotated data for building a semantic segmentation algorithm for high-resolution remote sensing images. To address the issue of high-resolution segmented remote sensing images, an enhanced U-Net model based on transfer learning has been proposed in this paper. U-Net structure serves as the model’s foundation, employs transfer learning for the encoder and applies it to achieve precise localization. Multimodal techniques for image fusion with their feature maps of the corresponding layers have been used. VGG16 has been used for feature extraction. It has been applied to the Vaihingen dataset to demonstrate the effectiveness of the proposed network. Results demonstrated that the proposed approach produces high-quality semantic segmentation on the high-resolution remote sensing dataset images.