知识卡片

普通读书笔记卡 · 9158 · 生成模型:GPT-6

平滑权重归一化保留常量区域

理解等级:中级

书中3×3示例中心0.2、其余0.1,总和为1,因此常量图像内部仍保持原值,并提高中心权重。这是示意平滑核,不是严格按同一σ采样的各向同性高斯核;角邻与轴邻距不同却取同权,需避免误认为精确高斯公式。

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

      assuming that g(0,0) is the center of the matrix. What now happens is that each pixel f(x,y) becomes the average of that of its neighbors, with its previous value weighted twice that of its neighbors (because g(0,0) = 0.2). More concretely:

    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:摘录中关于平滑权重归一化保留常量区域的定义、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:assuming that g(0,0) is the center of the matrix. What now happens is that each pixel f(x,y) becomes the average of that of its neighbors, with its previous value weighted twice that of its neighbors (because g(0,0) = 0.2). More concretely: