← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 8848.d · 生成模型:GPT-6

亮度与色度可分配不同采样率

理解等级:中级

人对不同颜色方向的空间变化敏感度不同。将信号变到适当亮度与色度表示,可较粗地保留色度、较精细保留亮度,以减少数据量;Lab示例展示这种感知差异。它不表示任意彩边都不可见,也不说明所有压缩都实际采用Lab。

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

      is relatively insensitive to blurring of the a or b components, as demonstrated in figure 8.16 and 8.17. This effect is commonly exploited in image compression, allowing some image components to be sampled more coarsely than others.

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第08章《颜色》(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/c008.xhtml) - 结论依据:摘录中关于亮度与色度可分配不同采样率的定义、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:is relatively insensitive to blurring of the a or b components, as demonstrated in figure 8.16 and 8.17. This effect is commonly exploited in image compression, allowing some image components to be sampled more coarsely than others.