一种可感知雾的轻量级视觉转换器架构，用于边缘设备上的实时海岸景观识别
A fog-aware lightweight vision transformer architecture for real-time coastal landscape recognition on edge devices.
摘要(中文译文)
智能景观识别是智慧滨海旅游的核心，然而将视觉Transformer部署到资源受限的边缘硬件上仍面临困难——自注意力机制的计算量随token数量呈二次方增长，而海洋大气雾霾在实验室察觉之前就已悄然侵蚀识别精度。我们提出一种混合轻量级架构：在浅层阶段用深度可分离卷积token混合器替代标准自注意力，在深层阶段改用基于核的线性注意力，并采用自适应通道缩减策略，在经验上冗余度最高的位置裁剪特征维度。受Koschmieder大气散射模型启发，我们在卷积阶段与注意力阶段之间嵌入了一个雾感知特征校准模块，使去雾学习在网络内部完成，而非作为单独预处理步骤。训练通过三阶段流水线进行，将渐进式结构化剪枝与来自Swin-Base教师的温度调度知识蒸馏交替进行；该复合目标将模型压缩至4.8 M参数和0.91 GFLOPs，且未出现精度骤降。在覆盖浙江省六个海岸区域、八个滨海场景类别的新采集17,565张图像数据集上，我们的模型达到92.6% ± 0.4%的Top-1准确率（三次独立运行的均值 ± 标准差），在参数量低于6 M的所有基线中表现最佳，包括近期发布的MobileViT-v2、FastViT、TinyViT、EfficientViT和RepViT。在搭载TensorRT INT8优化的NVIDIA Jetson Orin Nano上，系统在7.4 W平均功耗范围内实现18.3 ms延迟（54.6 FPS）；在混合合成与真实拍摄的200 m能见度浓雾子集上，准确率保持率为85.7%，比最强基线高6.8个百分点。在福建、海南和Places365滨海子集上的跨数据集零样本测试，以及在Jetson Nano 4 GB和Coral Edge TPU上的部署基准测试表明，该设计能够合理迁移至原始采集地点之外——尽管并非轻而易举。
Abstract
Intelligent landscape recognition sits at the heart of smart coastal tourism, yet pushing vision Transformers onto resource-constrained edge hardware remains awkward-the self-attention mechanism scales quadratically with token count, and marine atmospheric haze quietly erodes recognition accuracy long before it is noticed in the laboratory. We propose a hybrid lightweight architecture that swaps standard self-attention in the early stages for depthwise separable convolution token mixers, switches to kernel-based linear attention in the deeper stages, and follows an adaptive channel reduction policy that trims feature dimensions where redundancy is empirically highest. A Fog-Aware Feature Calibration module, motivated by the Koschmieder atmospheric scattering model, is embedded between the convolutional and attention stages so that learned dehazing happens inside the network rather than as a separate preprocessing pass. Training proceeds through a three-phase pipeline that interleaves progressive structured pruning with temperature-scheduled knowledge distillation from a Swin-Base teacher; the composite objective shrinks the model to 4.8 M parameters and 0.91 GFLOPs without abrupt accuracy collapse. On a newly collected 17,565-image dataset spanning eight coastal scene categories across six shoreline regions of Zhejiang Province, our model reaches 92.6% ± 0.4% Top-1 accuracy (mean ± std over three independent runs), the best result among all baselines below 6 M parameters, including the recently released MobileViT-v2, FastViT, TinyViT, EfficientViT and RepViT. On an NVIDIA Jetson Orin Nano with TensorRT INT8 optimisation the system delivers 18.3 ms latency (54.6 FPS) within a 7.4 W average power envelope, and a 200 m visibility heavy-fog subset that mixes synthetic and real captures shows an 85.7% accuracy retention rate-6.8 percentage points above the strongest baseline. Cross-dataset zero-shot tests on Fujian, Hainan and the Places365 coastal subset, together with deployment benchmarks on Jetson Nano 4 GB and Coral Edge TPU, suggest the design transfers reasonably-though not effortlessly-beyond the original collection sites.
如何引用
AMA
Yanyan He, Dekun Liu, Yibo Hu, Qiang Li. A fog-aware lightweight vision transformer architecture for real-time coastal landscape recognition on edge devices.. Scientific reports. 2026; doi:10.1038/s41598-026-59312-1.
APA
Yanyan He, Dekun Liu, Yibo Hu, Qiang Li (2026). A fog-aware lightweight vision transformer architecture for real-time coastal landscape recognition on edge devices.. Scientific reports. https://doi.org/10.1038/s41598-026-59312-1
GB/T 7714
Yanyan He, Dekun Liu, Yibo Hu, Qiang Li. A fog-aware lightweight vision transformer architecture for real-time coastal landscape recognition on edge devices.[J]. Scientific reports, 2026 doi:10.1038/s41598-026-59312-1.
在 PubMed增强版 阅读全文
翻译全文 · AI 文献问答 · PDF 原文下载 · 引用导出