雾霾天气下运动目标检测技术研究  

Research on Moving Target Detection Technology in Fog and Haze Weather

作　　者：娄莉[1] 张艳艳 

机构地区：[1]西安石油大学计算机学院,陕西西安710065

出　　处：《软件导刊》2017年第11期133-136,共4页Software Guide

基　　金：陕西省工业科技攻关项目(2015GY026);陕西省工业科技攻关项目(2016GY123);陕西省教育厅计算机科学与技术(学术学位)研究生教育综合改革项目(2015-SZH-015)

摘　　要：雾霾天气的频繁出现使运动目标检测受到了很大影响。为提高雾霾天气下运动目标检测质量,研究了当前去雾算法和运动目标检测算法。针对雾霾天气下可见度低、运动物体模糊的情况,比较了当前的去雾算法,使用性能优良的基于暗通道的去雾算法对图像进行去雾,然后对处理过的图像序列进行光流法计算,完成运动目标检测。将基于暗通道的去雾算法和光流算法相结合,应用于雾霾天气下的运动目标检测,以提高检测质量。通过Matlab仿真表明,改进方法比直接通过光流算法检测的效果更好。Nowadays,with more and more haze in air,the detection of moving target has been greatly affected.In order to improve the quality of moving target detection in haze weather,this paper studied the defogging algorithm and moving target detection algorithm in haze weather.Aiming at the situation of low visibility and moving objects in haze weather,the existed algorithm is compared,the image is defogged by the dark-channel defogging algorithm with great performance,and then the processed image sequence is computed with optical flow method to complete the detection of moving targets.In this paper,the dark-channel defogging algorithm and the optical flow algorithm are applied to the detection of moving objects in haze weather to improve the detection quality.The experimental results show that the method used in this paper is better than that directly through the optical flow algorithm.

关 键 词：运动目标检测 暗通道去雾算法 Horn-Schunck光流法 雾霾 Matlab 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]