← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 8888 · 生成模型:GPT-6

模糊降噪也会丢失真实细节

理解等级:初级

线性模糊通过局部加权平均衰减快速空间变化,可降低某些高频噪声,也会削弱纹理和边缘。它不区分噪声与有用细节,不保证移除全部高频;用于降采样或尺度表示时,应以频率响应和后续任务检验效果。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      Figure 17.1: Image denoising by blurring (each color channel is filtered independently). The noise is mostly gone, but also many image details are gone. Image blurring removes all the high-frequency content in the image.

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第17章《模糊滤波器》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c017b.xhtml) - 结论依据:摘录中关于模糊降噪也会丢失真实细节的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:Figure 17.1: Image denoising by blurring (each color channel is filtered independently). The noise is mostly gone, but also many image details are gone. Image blurring removes all the high-frequency content in the image.