Nighttime image dehazing faces a more complex degradation pattern than its daytime counterpart, as haze scattering couples with low illumination, non-uniform lighting, and strong light interference. Under limited supervision, this complexity aggravates domain drift and training instability, since target-domain samples are scarce while naively introducing external data may weaken adaptation due to distribution mismatch. This paper presents our solution to the NTIRE 2026 Night Time Image Dehazing Challenge, built as a unified framework that integrates domain-aligned data construction, stage-wise training, and inference-time enhancement. Specifically, a pre-trained CLIP visual encoder screens candidate external samples by similarity to construct training data closer to the target domain. NAFNet is then trained in two stages, first adapting to the target domain and then expanding to broader degradation patterns. At inference time, TLC, x8 self-ensemble, and weighted snapshot fusion are combined to improve output stability. Rather than relying on complex network redesign, the proposed framework offers a practical and effective pipeline for nighttime image dehazing.

翻译：夜间图像去雾面临比白天更复杂的退化模式，因为雾霾散射与低照度、非均匀照明及强光干扰相互耦合。在有限监督条件下，这种复杂性加剧了域漂移和训练不稳定问题——目标域样本稀缺，而简单引入外部数据可能因分布不匹配削弱自适应性。本文提出了针对NTIRE 2026夜间图像去雾挑战的解决方案，构建了一个集成域对齐数据构建、分阶段训练与推理增强的统一框架。具体而言，预训练的CLIP视觉编码器通过相似度筛选候选外部样本，构建更接近目标域的训练数据。随后采用两阶段训练NAFNet：首先适应目标域，再扩展至更广泛的退化模式。推理阶段结合TLC、8倍自集成与加权快照融合以提升输出稳定性。该框架不依赖复杂的网络重设计，为夜间图像去雾提供了一种实用高效的流水线方案。