IDP-YOLOV9 is an enhanced object detection model designed to improve performance in severe weather conditions from a drone perspective. The model integrates a parallel architecture comprising the Image Dehazing and Enhancement Processing (IDP) module and an improved YOLOV9 detection module. The IDP module includes the Three-Weather Removal Algorithm (TRA) and Deep Learning-based Image Enhancement (DLIE) modules, which work together to filter multiple weather factors and enhance image quality. The improved YOLOV9 detection module incorporates a three-layer routing attention mechanism to enhance object detection. Experiments show that the IDP module significantly improves image quality by mitigating the impact of adverse weather conditions. Compared to traditional single-processing models, the method improves recognition accuracy on complex weather datasets by 6.8% in terms of mean average precision (mAP50). The model's primary contributions include designing a parallel optimization architecture for the TRA and DLIE modules, proposing an improved YOLOV9 detection network with a three-layer routing attention mechanism, and introducing a comprehensive loss function that dynamically adjusts parameters based on weather factor features. The model is trained end-to-end and demonstrates improved detection accuracy under various weather conditions. The IDP-YOLOV9 model is evaluated on different datasets and compared with existing methods, showing superior performance in image processing and object detection under severe weather conditions.IDP-YOLOV9 is an enhanced object detection model designed to improve performance in severe weather conditions from a drone perspective. The model integrates a parallel architecture comprising the Image Dehazing and Enhancement Processing (IDP) module and an improved YOLOV9 detection module. The IDP module includes the Three-Weather Removal Algorithm (TRA) and Deep Learning-based Image Enhancement (DLIE) modules, which work together to filter multiple weather factors and enhance image quality. The improved YOLOV9 detection module incorporates a three-layer routing attention mechanism to enhance object detection. Experiments show that the IDP module significantly improves image quality by mitigating the impact of adverse weather conditions. Compared to traditional single-processing models, the method improves recognition accuracy on complex weather datasets by 6.8% in terms of mean average precision (mAP50). The model's primary contributions include designing a parallel optimization architecture for the TRA and DLIE modules, proposing an improved YOLOV9 detection network with a three-layer routing attention mechanism, and introducing a comprehensive loss function that dynamically adjusts parameters based on weather factor features. The model is trained end-to-end and demonstrates improved detection accuracy under various weather conditions. The IDP-YOLOV9 model is evaluated on different datasets and compared with existing methods, showing superior performance in image processing and object detection under severe weather conditions.