改进L-K光流法检测运动目标

Improved L-K Optical Flow Method for Moving Target Detection

近年来,图像处理技术蓬勃发展,运动目标检测作为图像处理研究领域的一个关键部分,具有重要的研究意义和应用价值。由于特征点匹配算法在实际应用中受到很多客观因素的影响,采用图像光度代替传统的视觉里程计算法越来越为研究者们所重视。本文以光流算法为基础,深度分析和改进L-K光流法检测运动目标的问题。论文的研究工作主要如下:介绍传统光流法的概念及算法;深度学习改进的L-K光流算法;提出一种基于金字塔L-K光流法检测运动目标。

In recent years,the image processing technology has been developing rapidly.As a key part of the image processing research field,moving target detection is of great research significance and application value.Because the feature point matching algorithm is affected by many objective factors in the practical application,the researchers pay more attention to using the image luminosity instead of the traditional visual distance calculation method.Based on the optical flow algorithm,this paper deeply analyzes and improves the problem of detecting moving targets by L-K optical flow method.The main research work of this paper is as follows:Introducing the concept and algorithm of traditional optical flow method;Improving L-K optical flow algorithm by deep learning;Proposing a detecting moving targets method based on pyramid L-K optical flow.