Image denoising is crucial for reliable image analysis. Researchers from diverse fields have long worked on this, but we still need better solutions. This article focuses on efficiently preserving key image features like edges and structures during denoising. Jump regression analysis is commonly used to estimate true image intensity amid noise. One approach is adaptive smoothing, which uses various local neighborhood shapes and sizes based on empirical data, while another is local pixel clustering to reduce noise while maintaining important details. This manuscript combines both methods to propose an integrated denoising technique.

翻译：图像去噪对于可靠的图像分析至关重要。来自不同领域的研究人员长期致力于此，但我们仍需要更优的解决方案。本文重点关注在去噪过程中如何有效保留边缘与结构等关键图像特征。跳跃回归分析常用于在噪声中估计真实的图像强度。一种方法是基于经验数据、采用不同局部邻域形状与尺寸的自适应平滑；另一种方法是通过局部像素聚类在保持重要细节的同时降低噪声。本文融合这两种方法，提出了一种集成的去噪技术。