 This paper proposes an improved small-object detection algorithm for high-altitude UAV photography. It replaces the original convolution layer with a specialized SPD convolution module, adds a coordinate attention mechanism to further improve detection accuracy, replaces the original upsampling method with transposed convolution to increase the receptive field range of the neck, and finally replaces the CIOU loss function with the alpha-IOU loss function to solve the problem of the slow convergence of gradients during training on small target images. The experimental results demonstrate that the proposed algorithm provides significantly improved results. Average precision equals 80.17 percent, accuracy equals 73.45 percent, and recall rate equals 76.97 percent, improvements by 14.96 percent, 6.24 percent, and 7.21 percent, respectively, compared with the original model, and also outperforms other detection algorithms. This article was authored by Yanlong Chang, Dongli, Yunlong Gao, and others. We are article.tv, links in the description below.