← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9166 · 生成模型:GPT-6

边缘响应强不一定定位稳定

理解等级:中级

沿边缘方向移动仍可能产生相似响应,所以强边缘点的位置可能沿线不确定。比较局部二阶变化可拒绝这类关键点,并去掉低对比候选;剩下的是适合重复定位的局部结构,不是要求保留所有视觉边界。

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

      3. Reject extrema that are along edges by looking at the second partial derivatives in image-space around each extremum (figure 9.5, right). Edges have a much larger principal curvature across them than along them.

    参考来源

    - 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第09章《特征提取》(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_9.xhtml) - 结论依据:摘录中关于边缘响应强不一定定位稳定的定义、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:3. Reject extrema that are along edges by looking at the second partial derivatives in image-space around each extremum (figure 9.5, right). Edges have a much larger principal curvature across them than along them.