To address the challenges of difficult target detection and degraded detection performance in remote sensing images caused by adverse weather conditions such as rain, snow, and fog, this study proposes a remote sensing target detection model integrated with image enhancement, namely you only look once-image detail enhancement (YOLO-IDE), based on the YOLOv11 architecture. Firstly, inspired by the GFFA dehazing network, an image detail enhancement (IDE) module is designed to mitigate the impact of degraded features on the detection task. The enhancement module is integrated with the detection network into an end-to-end architecture to improve target detection performance. Meanwhile, the original C3k2 module in the YOLOv11 backbone is replaced with the CSP efficient dual layer aggregation networks (CSP-EDLAN) module. By adopting dual convolution techniques, the proposed module enhances feature representation while alleviating excessive computational consumption. Furthermore, the original Conv module is replaced with a receptive field dual attention convolution (RFDAConv) module, which expands the receptive field to suppress background noise and thereby improve target detection accuracy. The three components function in sequential coordination, first improving the quality of input images, then optimizing feature representation capability, and finally denoising deep-level features to refine outputs. Finally, experiments are conducted on both real-world and synthetic datasets under adverse weather conditions. The results show that, compared with the baseline YOLOv11 model, the proposed model achieves improvements of 2.4% and 2.3% in mAP@0.5, as well as 4% and 1.3% in mAP@0.5:0.95, respectively. These results verify the effectiveness and feasibility of the proposed method.