知识卡片

普通读书笔记卡 · 8894.d · 生成模型:GPT-6

高斯导数的正负两瓣

理解等级:中级

二维高斯的一阶 x 导数为 gₓ=−xg/σ²,具有沿 x 变化的正负两瓣,总积分为零。它以两侧局部加权亮度之差检测变化,而不是输出该处平均亮度。卷积改为相关会反转奇对称核的响应符号,解释方向前必须核对实现约定。

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

      If g is a Gaussian, then the first order derivative is (18.17) The next plots shows the 2D Gaussian with σ = 1, and its x-derivative: Figure 18.7: (a) 2D Gaussian with σ = 1, and (b) its x-derivative. Figure 18.8 shows an image filtered with three Gaussian derivatives with different widths, σ. Derivatives at different scales emphasize different aspects of the image. The fine-scale

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第18章《图像导数》(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/c018.xhtml) - 结论依据:摘录中关于高斯导数的正负两瓣的定义、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:If g is a Gaussian, then the first order derivative is (18.17) The next plots shows the 2D Gaussian with σ = 1, and its x-derivative: Figure 18.7: (a) 2D Gaussian with σ = 1, and (b) its x-derivative. Figure 18.8 shows an image filtered with three Gaussian derivatives with different widths, σ. Derivatives at different scales emphasize different aspects of the image. The fine-scale