知识卡片

普通读书笔记卡 · 9165.c · 生成模型:GPT-6

尺度空间让特征选择适应尺寸

理解等级:中级

SIFT在不同尺度的高斯平滑图中构造差分,并跨空间和相邻尺度寻找局部极大或极小。构建金字塔需结合平滑与重采样,不能只抽取每隔几个像素就保证没有混叠;尺度选择支持不同成像尺寸下重复定位。

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

      2. Detect local minima and maxima in the DoG images across different scales (figure 9.5, left) and reject those with low contrast (figure 9.5, right). Figure 9.5 After scale space extrema are detected (left), the SIFT algorithm discards low contrast keypoints (center) and then filters out those located on edges (right). Source: Lukas Mach CC-BY 3.0. 3. Reject extrema that are along edges by looking at

    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:摘录中关于尺度空间让特征选择适应尺寸的定义、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:2. Detect local minima and maxima in the DoG images across different scales (figure 9.5, left) and reject those with low contrast (figure 9.5, right). Figure 9.5 After scale space extrema are detected (left), the SIFT algorithm discards low contrast keypoints (center) and then filters out those located on edges (right). Source: Lukas Mach CC-BY 3.0. 3. Reject extrema that are along edges by looking at