<p style="margin-top:0px; margin-bottom:5px; padding-top:0px; padding-bottom:0px; border:0px; list-style:none; word-wrap:normal; word-break:normal; line-height:21px; color:rgb(70,70,70); font-family:simsun; font-size:14px"> <span style="background-color:rgb(255,255,255)"><span style="word-wrap:normal; word-break:normal; line-height:27px; font-size:18px"><span style="word-wrap:normal; word-break:normal; font-weight:bold">博主声明</span></span>：本文<span style="word-wrap:normal; word-break:normal; font-weight:bold">英文内容</span>全部来自于http://en.wikipedia.org/wiki/Lucas_Kanade_method,中文由博主据此进行翻译而来。</span></p>
<p style="margin-top:0px; margin-bottom:5px; padding-top:0px; padding-bottom:0px; border:0px; list-style:none; word-wrap:normal; word-break:normal; line-height:21px; color:rgb(70,70,70); font-family:simsun; font-size:14px"> <span style="background-color:rgb(255,255,255)"> <wbr></wbr> <wbr></wbr> <wbr></wbr> 在计算机视觉中，Lucas-Kanade方法是一种广泛使用的光流估算差分方法，它由Bruce D.Lucas和Takeo Kanade共同开发。它假定在所考虑的像素的局部邻域内，本质上光流是恒定的，由此利用最小二乘原则对邻域内所有像素求解基本光流方程。[In <a href="http://en.wikipedia.org/wiki/Computer_vision" rel="noopener noreferrer" style="text-decoration:none; color:rgb(88,122,100)" target="_blank" title="Computer vision">computer vision</a>, the <strong>Lucas–Kanade method</strong> is a widely used differential method for <a href="http://en.wikipedia.org/wiki/Optical_flow" rel="noopener noreferrer" style="text-decoration:none; color:rgb(88,122,100)" target="_blank" title="Optical flow">optical flow</a> estimation developed by <a href="http://en.wikipedia.org/w/index.php?title=Bruce_D._Lucas&action=edit&redlink=1" rel="noopener noreferrer" style="text-decoration:none; color:rgb(88,122,100)" target="_blank" title="Bruce D. Lucas (page does not exist)">Bruce D. Lucas</a> and <a href="http://en.wikipedia.org/wiki/Takeo_Kanade" rel="noopener noreferrer" style="text-decoration:none; color:rgb(88,122,100)" target="_blank" title="Takeo Kanade">Takeo Kanade</a>. It assumes that the flow is essentially constant in a local neighbourhood of the <a href="http://en.wikipedia.org/wiki/Pixel" rel="noopener noreferrer" style="text-decoration:none; color:rgb(88,122,100)" target="_blank" title="Pixel">pixel</a> under consideration, and solves the basic optical flow equations for all the pixels in that neighbourhood, by the <a href="http://en.wikipedia.org/wiki/Least_squares_method" rel="noopener noreferrer" style="text-decoration:none; color:rgb(88,122,100)" target="_blank" title="Least squares method">least squares criterion</a>.<sup><a href="http://en.wikipedia.org/wiki/Lucas%E2%80%93Kanade_method#cite_note-LKproc-0" rel="noopener noreferrer" style="text-decoration:none; color:rgb(88,122,100)" target="_blank"><span style="word-wrap:normal; word-break:normal; line-height:18px">[</span>1<span style="word-wrap:normal; word-break:normal; line-height:18px">]</span></a></sup><sup><a href="http://en.wikipedia.org/wiki/Lucas%E2%80%93Kanade_method#cite_note-Lthesis-1" rel="noopener noreferrer" style="text-decoration:none; color:rgb(88,122,100)" target="_blank"><span style="word-wrap:normal; word-break:normal; line-height:18px">[</span>2<span style="word-wrap:normal; word-break:normal; line-height:18px">]</span></a></sup></span></p>
<p style="margin-top:0px; margin-bottom:5px; padding-top:0px; padding-bottom:0px; border:0px; list-style:none; word-wrap:normal; word-break:normal; line-height:21px; color:rgb(70,70,70); font-family:simsun; font-size:14px"> <span style="background-color:rgb(255,255,255)"> <wbr></wbr> <wbr></wbr> 通过综合多个附近像素的信息，Lucas-Kanade方法通常可以解决光流方程固有的模糊性问题。与逐点方法相比，对于图像噪声它也不太敏感。另一方面，因为这是一种纯局部的方法，它不能够提供图像均匀区域内的光流信息。[By combining information from several nearby pixels, the Lucas-Kanade method can often resolve the inherent ambiguity of the optical flow equation. It is also less sensitive to image noise than point-wise methods. On the other hand, since it is a purely local method, it cannot provide flow information in the interior of uniform regions of the image.]</span></p>
<h2 style="margin:0px; padding:0px; border:0px; list-style:none; color:rgb(70,70,70); font-family:simsun; line-height:21px"> <span style="background-color:rgb(255,255,255)">概念[<span style="word-wrap:normal; word-break:normal; line-height:31px">Concept]</span></span></h2>
<p style="margin-top:0px; margin-bottom:5px; padding-top:0px; padding-bottom:0px; border:0px; list-style:none; word-wrap:normal; word-break:normal; line-height:21px; color:rgb(70,70,70); font-family:simsun; font-size:14px"> <span style="background-color:rgb(255,255,255)"> <wbr></wbr> <wbr></wbr> <wbr></wbr> Lucas-Kanade方法假定两个附近帧图像内容的位移很小而且在所考虑的点p的邻域内大体不变。因此可以假定在以点p为中心的窗口内的所有像素，光流方程可以适用。就是说，局部图像光流矢量（Vx,Vy）必须满足以下方程：[The Lucas-Kanade method assumes that the displacement of the image contents between two nearby instants (frames) is small and approximately cons |
|
转播
