面向烟雾遮蔽环境下设备端感知的去雾前端输入自适应门控机制
Input-Adaptive Gating of a Dehazing Front-End for On-Device Perception in Smoke-Obscured Environments
- Monta Vista High School（蒙塔维斯塔高中）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对烟雾环境设备端感知的两阶段视觉管线，提出输入自适应门控机制，仅在雾霾超阈值时运行去雾器，提升边缘检测精度与帧率。
AI中文摘要：
两阶段视觉管线通常在任务网络前放置增强网络，假设更清晰的输入能产生更好的输出。我们在消防员辅助管线中评估这一假设，该管线中去雾器位于边缘检测器之前，边缘检测器可将充满烟雾的房间渲染为结构轮廓。两者均针对树莓派4设计，参数规模分别为35.5万和2.3万，通过TensorFlow Lite量化为UINT8格式。浮点型去雾器在保留的真实烟雾图像上达到18.60 dB的峰值信噪比（PSNR），未处理图像为13.60 dB，在相同数据上训练的AOD-Net为17.08 dB；边缘检测器在最优数据集尺度（ODS）下的F值为0.738，优于优化后的Canny算法的0.692。去雾操作可提升浓雾下的边缘提取效果，但在清晰和轻度雾霾帧中会降低性能，此时去雾器丢弃的细节多于雾霾所遮蔽的部分。因此，我们仅在暗通道雾霾估计值超过阈值时运行去雾器，该测试耗时10.1毫秒，可让管线在去雾阶段节省469.6毫秒。在四种雾霾水平上取平均，自适应门控机制的平均ODS为0.675，比固定决策（始终去雾为0.664，不去雾为0.630）更准确；它将树莓派上单帧平均时间从569毫秒降至321毫秒，在清晰帧上将帧率提高五倍，从1.8帧/秒升至9帧/秒。
英文摘要：
Two-stage vision pipelines often place an enhancement network before a task network, on the assumption that a cleaner input produces a better output. We evaluate this in a firefighter assistance pipeline, where a dehazer precedes an edge detector that renders smoke-filled rooms as structural outlines. Both were designed for a Raspberry Pi 4, at 355K and 23K parameters, and quantized to UINT8 via TensorFlow Lite. The float dehazer reaches 18.60 dB peak signal-to-noise ratio (PSNR) on held-out real smoke against 13.60 dB unprocessed and 17.08 dB for an AOD-Net trained on the same data, and the edge detector reaches an F-measure at optimal dataset scale (ODS) of 0.738, outperforming an optimized Canny's result of 0.692. Dehazing improves edge extraction under dense smoke but degrades it on clear and lightly hazed frames, where the dehazer discards more detail than the haze obscures. We therefore run the dehazer only when a dark channel haze estimate exceeds a threshold, a 10.1 ms test that lets the pipeline save 469.6 ms on the dehazing stage. Averaged over four haze levels, gating is more accurate than either fixed decision, at 0.675 mean ODS against 0.664 for always dehazing and 0.630 for never dehazing. It reduces the mean per-frame time on the Raspberry Pi from 569 ms to 321 ms, and on clear frames increases the frame rate fivefold, from 1.8 to 9 frames per second.
