Abstract: In the process of analyzing urban land resource utilization types, classification results are often obtained through edge segmentation of satellite remote sensing images, which contain low resolution geospatial data, resulting in lower Kappa coefficients in the final classification results. Therefore, a classification method for urban land resource utilization based on unmanned aerial vehicle (UAV) mapping image edge segmentation algorithm is proposed. Using UAV equipment to collect surveying and mapping images of urban land resource areas, wavelet domain threshold denoising, adaptive median filtering, and adaptive Wiener filtering are completed to remove Gaussian and pulse mixed noise in the images. Divide the UAV mapping image into multiple small blocks and perform grayscale nonlinear transformation processing to enhance the original image. For the preprocessed UAV surveying images, the window grayscale weighting algorithm is used to extract image edge features, and then a low rank reconstruction network is used to complete the surveying image edge segmentation processing. Design a classification and recognition algorithm with support vector machine (SVM) as the core, judge each land resource segmentation area, and obtain the final urban land resource utilization classification result. The experimental results show that the Kappa coefficient of the classification results obtained by this method exceeds 0.85, greatly improving the accuracy of land resource utilization classification.