← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9166.a · 生成模型:GPT-6

SIFT方向描述来自平滑图梯度

理解等级:中级

方向与幅值应由关键点尺度的平滑图梯度计算,幅值使用两个方向差分平方和的平方根,方向用可区分象限的角度。原书把DoG差分及“平方差”直接当幅值不够准确。Lowe原论文第5节给出依据。

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

      4. Assign a “magnitude” and “orientation” to each remaining extremum, now called a “keypoint.” The magnitude is the squared difference between the DoG filter response at the present pixel and the neighboring pixels. The orientation is the arctangent between the DoG differences in the y direction and the x direction. These calculations are made for all pixels in a fixed neighborhood around the initial keypoint

    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SIFT方向描述来自平滑图梯度的定义、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:4. Assign a “magnitude” and “orientation” to each remaining extremum, now called a “keypoint.” The magnitude is the squared difference between the DoG filter response at the present pixel and the neighboring pixels. The orientation is the arctangent between the DoG differences in the y direction and the x direction. These calculations are made for all pixels in a fixed neighborhood around the initial keypoint