Abstract: To address the issues of small-target missed detection, false alarms from cloud/fog interference, and low computational efficiency in traditional wildfire detection for transmission line corridors, the YOLOv11_MDS detection model is proposed by integrating multi-scale convolutional attention (MSCA) and distribution shifting convolution(DSConv). The MSCA module is embedded in the backbone and neck to enhance multi-scale dynamic feature extraction of flame and smoke through collaborative depth strip convolution and channel attention. The DSConv with quantized dynamic shift mechanism is introduced to significantly reduce computational complexity while maintaining detection accuracy. Experiments show that the improved model achieves an mAP@0.5 of 88.21% on the self-built dataset, increasing by 2.93% compared with the original YOLOv11, with recall improved by 3.33% and frame rate enhanced to 42 FPS, notably improving the detection of small targets (pixel occupancy < 1%). Generalization tests demonstrate mAP improvements of 0.4% and 0.7% on benchmark datasets, effectively resolving false/missed detection in complex backgrounds. This study provides an engineering solution for real-time wildfire monitoring in transmission lines with balanced accuracy and efficiency.