Shi Donghao, Zhao Cunbin, Shao Jianwen, Feng Minjie, Luo Lei, Ouyang Bing, Huang Jiamin
Zhejiang Key Laboratory of Digital Precision Measurement Technology Research, Hangzhou, China.
Advanced Manufacturing Metrology Research Center, Zhejiang Institute of Quality Sciences, Hangzhou, China.
Front Neurorobot. 2025 Jun 10;19:1588565. doi: 10.3389/fnbot.2025.1588565. eCollection 2025.
Small object detection is a critical task in applications like autonomous driving and ship black smoke detection. While Deformable DETR has advanced small object detection, it faces limitations due to its reliance on CNNs for feature extraction, which restricts global context understanding and results in suboptimal feature representation. Additionally, it struggles with detecting small objects that occupy only a few pixels due to significant size disparities. To overcome these challenges, we propose the Context-Aware Enhanced Feature Refinement Deformable DETR, an improved Deformable DETR network. Our approach introduces Mask Attention in the backbone to improve feature extraction while effectively suppressing irrelevant background information. Furthermore, we propose a Context-Aware Enhanced Feature Refinement Encoder to address the issue of small objects with limited pixel representation. Experimental results demonstrate that our method outperforms the baseline, achieving a 2.1% improvement in mAP.
小目标检测是自动驾驶和船舶黑烟检测等应用中的一项关键任务。虽然可变形DETR在小目标检测方面取得了进展，但由于其依赖卷积神经网络（CNNs）进行特征提取，它面临着局限性，这限制了对全局上下文的理解，并导致特征表示不够理想。此外，由于尺寸差异显著，它在检测仅占据几个像素的小目标时也存在困难。为了克服这些挑战，我们提出了上下文感知增强特征细化可变形DETR，这是一种改进的可变形DETR网络。我们的方法在主干中引入了掩码注意力，以改进特征提取，同时有效抑制无关的背景信息。此外，我们还提出了一种上下文感知增强特征细化编码器，以解决像素表示有限的小目标问题。实验结果表明，我们的方法优于基线，平均精度均值（mAP）提高了2.1%。
