基于改进YOLOv11n模型的输送带异物检测方法

打开文本图片集

中图分类号：TD634.1 文献标志码：A

DOI：10.13272/j.issn.1671-251x.2025040072

Abstract：Toaddress the problem of large differences in foreign object sizes，complex backgrounds，and poor detection performance of smalland slender targets in coal mine conveyor belts，a foreign object detection method based onan improved YOLOvlln model was proposed.The core improvements of the YOLOvlln model included three aspects： first，a Scale Sequence Feature Fusion （SSFF） module was introduced into the neck network toenhance the effective captureand fusion of informationat different scales through sequential scale interaction；second，aparalel C3K2PPA module was constructed，introducing a Parallelized Patch-Aware Attention （PPA）module in thespatial dimension to highlight keyregion representations and improverecal; third， a Band-Aware Contrast Fusion （BACF） module was designed at the lateral fusion layer of the same scale. By combining belt-directional priorsand high-passedge indicators，and replacing simple concatenation with pixelwise gating，the module suppressed periodic background noise along the belt direction and enhanced cros-branch differences without increasing the number of channels，thereby improving the model's discriminative capability and robustness under complex working conditions.The experimental results showed that the precision and recall of the improved YOLOv1ln model reached 0.914 and 0.892， respectively， with mAP@0.5 and mAP@0.5：0.95 （204号 values of 93.1% and 62.2% ， showing significant improvement over the original YOLOvlln and outperforming mainstream lightweight models such as YOLOv5s， YOLOv8n， and YOLOv10n in accuracy and robustness The inference speed of the model reached 96 frames per second，indicating high real-time performance and efficient execution in coal mine conveyor belt foreign object detection tasks. Heatmap analysis showed that the improved YOLOvl1n model effectively enhanced the target-area focusing capability，reduced redundant bounding boxes， and improved the detection accuracy of small targets.

Key Words： conveyor belt foreign object detection; YOLOvlln; Scale Sequence Feature Fusion： Parallelized Patch-Aware Attention; Band-Aware ContrastFusion

0引言

带式输送机作为煤流运输的核心装备，其运行状态直接关系到整个生产系统的稳定性与安全性[1-2]。（剩余10165字）