Underwater object detection is critically significant in domains such as underwater resource exploration and marine environmental monitoring. However, due to the complex conditions of underwater environment, underwater imaging often suffers from issues such as blurred object boundaries, color distortion, and small object sizes, which limit the performance of traditional general-purpose object detection algorithms, even with the newly developed YOLOv11. To address these challenges, we propose an enhanced underwater small object detection framework based on edge enhancement, termed USEE-YOLO. To tackle the problem of blurred object boundaries in underwater images, we integrate edge enhancement and reassembly modules that reconstruct object boundary information at both the image and feature map levels. To rectify color distortion, USEE-YOLO adopts histogram-based image enhancement methods to optimize color representation. In addition, the framework incorporates a specialized detection head designed for small objects to improve recognition performance. Experiments conducted on the DUO and RUOD datasets show that USEE-YOLO improves \({\text{mAP}}_{50:95}\) by 2.2% and 1.5%, respectively, compared to YOLOv11, demonstrating its superior performance and robustness.