摘要近年来，监控设施已经在我们生活中得到广泛应用。它采用图像处理、模式识别和计算机视觉等技术。它通过使用智能视频分析模块，并借助计算机强大的数据处理能力过滤掉视频画面中无用的信息，分析有用信息，能够迅速判断出监控系统中的异常情况，并以最快和最佳的方式发出警报或触发其它动作。64837

本文在对监控设施研究方向做了一定研究，发现其中关于滞留物部分有很大的研究价值。本次实验系统主要基于帧间差分法和背景差分法这两种技术，先对视频进行场景检测，并针对不同场景运用不同算法检测出目标物。最终实现降低检测虚警率，相对提高遗留物体检测精确性的目的。

毕业论文关键词   差分 视频帧 滞留物检测  图像分割 

毕业设计说明书（论文）外文摘要

Title       The Detection and Analysis Technology of   Abnormal Retentate On the Pavement      

Abstract In recent years, the intelligent monitoring system has been widely used in all fields. It adopts the technology of image processing, pattern recognition and computer vision. Through the intelligent video analysis module, it use the aid of the computer's powerful data processing ability to filter out the useless information in video, and analyze the useful information. It can quickly judge the abnormal conditions in monitoring system, and alarm or trigger other actions with the fastest and best way.

This thesis has done some research on the research direction of the monitoring facilities, and found the retentate parts have great research value . The experiment system is mainly based on the frame difference method and background difference method. The system first detect different scenarios on the video, and use different algorithms to detect the target object according to different scenarios. In this way, we can lower the false alarm rate, improve the accuracy of relative left objects detection.

Keywords   Difference, Video Frame, Retentate Detection, Image Segmentation

目   次

1  绪论 1

1.1  概述 1

1.2  本文组织结构 2

2  相关工具 2

2.1  概述 2

2.2  MFC 2

2.3  Visual Studio 2010 3

2.4  OpenCV 3

3  运动目标检测 5

3.1  概述 5

3.2  运动目标检测 5

3.2.1  运动目标检测任务 5

3.2.2  运动目标检测流程 6

3.3  运动目标检测技术 7

3.3.1  帧间差分法 7

3.3.2  背景差分法 9

4  滞留物检测算法 10

4.1  概述 10

4.2  算法约束条件 11

4.2.1  系统环境约束 11

4.2.2  目标物体运动特性约束 12

4.3  算法场景设计 12

4.3.1  概述 OpenCV路面异常滞留物检测与分析技术研究:http://www.751com.cn/jisuanji/lunwen_72195.html