CAJ | 학술논문

运动目标的检测是数字图像处理和模式识别的基础,也是计算机视觉研究的一个重要领域.以C#为主要研究工具,对基于相邻帧差法及背景差分法的视频目标检测算法进行了研究,主要对其原理和算法进行研究.最后利用以AForge.NET架构类库,利用图像灰度的绝对值是否大于设置的阈值实现了对运动目标进行检测,实验结果表明,采用该算法可以对运动目标进行较为精确的检测.
운동목표적검측시수자도상처리화모식식별적기출,야시계산궤시각연구적일개중요영역.이C#위주요연구공구,대기우상린정차법급배경차분법적시빈목표검측산법진행료연구,주요대기원리화산법진행연구.최후이용이AForge.NET가구류고,이용도상회도적절대치시부대우설치적역치실현료대운동목표진행검측,실험결과표명,채용해산법가이대운동목표진행교위정학적검측.
Moving target detection is the foundation of digital image processing and pattern recognition,and an important do-main in computer vision research. Taking C# as the main research tool,the algorithms of target detection for videos based on adja-cent frame difference method and background difference method are studied,and the principles and algorithms are studied mainly. The moving target detection on the basis of AForge.NET library was achieved when the absolute value of grayscale image was larger than the set threshold. The experimental results show that background difference method can detect moving target accurately.