Abstract
Deep learning models like U-Net and its variants, have established state-of-the-art performance in edge detection tasks and are used by Generative AI services world-wide for their image generation models. However, their decision-making processes remain opaque, operating as "black boxes" that obscure the rationale behind specific boundary predictions. This lack of transparency is a critical barrier in safety-critical applications where verification is mandatory. To bridge the gap between high-performance deep learning and interpretable logic, we propose the Rule-Based Spatial Mixture-of-Experts U-Net (sMoE U-Net). Our architecture introduces two key innovations: (1) Spatially-Adaptive Mixture-of-Experts (sMoE) blocks integrated into the decoder skip connections, which dynamically gate between "Context" (smooth) and "Boundary" (sharp) experts based on local feature statistics; and (2) a Takagi-Sugeno-Kang (TSK) Fuzzy Head that replaces the standard classification layer. This fuzzy head fuses deep semantic features with heuristic edge signals using explicit IF-THEN rules. We evaluate our method on the BSDS500 benchmark, achieving an Optimal Dataset Scale (ODS) F-score of 0.7628, effectively matching purely deep baselines like HED (0.7688) while outperforming the standard U-Net (0.7437). Crucially, our model provides pixel-level explainability through "Rule Firing Maps" and "Strategy Maps," allowing users to visualize whether an edge was detected due to strong gradients, high semantic confidence, or specific logical rule combinations.
Abstract (translated)
深度学习模型如U-Net及其变体，在边缘检测任务中已达到业界领先水平，并被全球的生成式AI服务用于其图像生成模型。然而，这些模型的决策过程仍然是不透明的，“黑箱”操作使得特定边界预测背后的逻辑难以理解。这种缺乏透明度是安全关键应用中的一个重要障碍，在这些应用中必须进行验证。为了弥合高性能深度学习和可解释性逻辑之间的差距，我们提出了基于规则的空间混合专家U-Net（sMoE U-Net）。我们的架构引入了两个关键创新：(1) 将空间自适应混合专家(sMoE)块集成到解码器跳跃连接中，这些块根据局部特征统计动态地在“上下文”（平滑）和“边界”（尖锐）专家之间切换；以及 (2) 使用泰克吉-松永-康(TSK)模糊头替换标准分类层。该模糊头通过明确的IF-THEN规则融合深度语义特征与启发式边缘信号。 我们在BSDS500基准测试上评估了我们的方法，达到了最优数据规模(ODS) F分数为0.7628，在性能上接近纯深度基线模型如HED（0.7688），而优于标准U-Net（0.7437）。尤为重要的是，我们的模型通过“规则触发图”和“策略图”，在像素级提供了可解释性，使得用户能够可视化边缘是由于强梯度、高语义置信度还是特定逻辑规则组合被检测出来的。
URL
https://arxiv.org/abs/2602.05100
