知识卡片

普通读书笔记卡 · 9159 · 生成模型:GPT-6

高斯差分突出中间尺度结构

理解等级:中级

对两个不同尺度的平滑图相减得到DoG,可抑制共同的低频分量并衰减很高频成分,突出一定尺度的变化。尺度比控制响应范围,书中4到5倍是示例取值而非所有DoG都必须遵循的定义。

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

      An alternative method for detecting edges is the difference of Gaussians (DoG) method. The idea is to subtract two images that have each been filtered using a Gaussian kernel with different width. Both filters suppress high-frequency information, and their difference therefore leads to a band-pass filtered signal, from which both low and high frequencies have been removed. As such, a DoG filter acts as a

    参考来源

    - 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第08章《视觉》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)/_epub-src/OEBPS/xhtml/chapter_8.xhtml) - 结论依据:摘录中关于高斯差分突出中间尺度结构的定义、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:An alternative method for detecting edges is the difference of Gaussians (DoG) method. The idea is to subtract two images that have each been filtered using a Gaussian kernel with different width. Both filters suppress high-frequency information, and their difference therefore leads to a band-pass filtered signal, from which both low and high frequencies have been removed. As such, a DoG filter acts as a