Motion Blur Robust Wheat Pest Damage Detection with Dynamic Fuzzy Feature Fusion

TL;DR

This work tackles motion blur in field wheat pest detection by embedding a Dynamic Fuzzy Robust Convolution Pyramid (DFRCP) into YOLOv11, augmented with a Dynamic Robust Switch (DRS) and a CUDA-4D fuzzy tensor rotation kernel to enable end-to-end blur-robust detection on edge devices. The approach fuses multiscale features, adaptively injects fuzzy cues, and rotates/interpolates feature tensors in GPU-accelerated pipelines, achieving about 86.4% $mAP@0.5$ and 47 FPS on a dynamic-blur wheat dataset, outperforming the baseline by roughly 12.7% $mAP@0.5$ while maintaining real-time performance. Key innovations include the DFCP for robust cross-scale fusion, the DRS for content-adaptive fusion weighting, and end-to-end CUDA-based deblurring that reduces latency and energy consumption. The proposed framework demonstrates strong generalization to wind-induced blur and related challenging conditions, enabling practical, lightweight edge deployment for agricultural monitoring and automation.

Abstract

Motion blur caused by camera shake produces ghosting artifacts that substantially degrade edge side object detection. Existing approaches either suppress blur as noise and lose discriminative structure, or apply full image restoration that increases latency and limits deployment on resource constrained devices. We propose DFRCP, a Dynamic Fuzzy Robust Convolutional Pyramid, as a plug in upgrade to YOLOv11 for blur robust detection. DFRCP enhances the YOLOv11 feature pyramid by combining large scale and medium scale features while preserving native representations, and by introducing Dynamic Robust Switch units that adaptively inject fuzzy features to strengthen global perception under jitter. Fuzzy features are synthesized by rotating and nonlinearly interpolating multiscale features, then merged through a transparency convolution that learns a content adaptive trade off between original and fuzzy cues. We further develop a CUDA parallel rotation and interpolation kernel that avoids boundary overflow and delivers more than 400 times speedup, making the design practical for edge deployment. We train with paired supervision on a private wheat pest damage dataset of about 3,500 images, augmented threefold using two blur regimes, uniform image wide motion blur and bounding box confined rotational blur. On blurred test sets, YOLOv11 with DFRCP achieves about 10.4 percent higher accuracy than the YOLOv11 baseline with only a modest training time overhead, reducing the need for manual filtering after data collection.

Figures (3)

• Figure 1: Overall design drawing.
• Figure 2: Dynamic robust switch (DRS).
• Figure 3: Processing rotaion conditions of pixels close to image edge.