Abstract: Continual anomaly detection focuses on incrementally learning new classes while retaining historical memory. However, the spectral bias and high-frequency artifacts encountered in generative replay severely constrain the fine-grained segmentation of subtle anomalies. To address this, this study proposes DenoiseCAD, a noise-resistant framework based on a cascaded purification architecture, aiming to eliminate feature shifts caused by generative artifacts and prevent the model from capturing spurious features unrelated to defects. First, the study proposes a feature prototype-guided latent space correction mechanism. During the reverse diffusion process, it utilizes the feature prototypes of normal classes as semantic anchors and iteratively rectifies latent variables by calculating feature metric gradients, thereby suppressing distribution shift noise from the source. Second, a task-driven frequency filter is constructed based on parameter sensitivity experiments, implementing a multi-granular spectral joint constraint strategy tailored to data source characteristics to effectively block the propagation of high-frequency artifacts. Finally, anchor-based weight consolidation is implemented. Through isotropic parameter distance constraints, it prevents the model from overfitting to residual noise, thereby establishing a full-pipeline denoising framework from source to terminal. This effectively balances the model's plasticity and stability, successfully alleviates the catastrophic forgetting dilemma, and provides a reliable new paradigm for complex intelligent industrial inspection scenarios. Extensive experiments demonstrate that DenoiseCAD achieves state-of-the-art performance on both the VisA and MVTec datasets. Notably, it yields significant improvements of 2.8% and 1.5% in P-AP over previous state-of-the-art methods.

摘要： 持续异常检测侧重于增量学习新类别的同时保持历史记忆。然而，生成式回放面临的频谱偏差与高频伪影严重制约了微小异常的精细分割。为此，该研究提出了DenoiseCAD，一种基于级联纯化体系的抗噪框架，为消除生成伪影导致的特征偏移，防止模型捕捉与缺陷无关的虚假特征。首先，该研究提出了一种基于特征原型引导的潜空间校正机制，在模型扩散反向过程中利用正常类别的特征原型作为语义锚点，通过计算特征度量梯度来迭代修正潜变量，从源头抑制分布偏移噪声。其次，基于参数敏感性实验构建任务驱动型频率滤波，实施针对数据来源特性的多粒度频谱联合约束策略，有效阻断高频伪影的传播。最后，实施基于锚点的权重固化，通过各向同性的参数距离约束，防止模型对残留噪声过拟合。至此构建了从源头到末端的全链路去噪框架，从而有效平衡了模型的可塑性与稳定性，缓解了灾难性遗忘难题，为复杂工业智能质检场景提供了可靠的新框架。实验表明，DenoiseCAD 在 VisA 和 MVTec 数据集上均取得 SOTA 性能，其像素级异常分割精度较现有最优方法分别提升了 2.8% 和 1.5%。