摘要
针对运动目标检测中背景模型的维护问题,提出了基于动态三元组(DTDG)的背景建模方法。该方法给出了动态三元组的概念,对每个像素维护一个动态三元组,根据像素的动态变化信息决定更新策略。实现了背景的自动更新,可以适应光照的突变、缓变和场景本身的变化。实验表明了该方法在复杂场景条件下的有效性。
To solve the maintenance of background model in moving objects detection, an approach to background modeling based on dynamic-three-dimension-group （DTDG） is proposed. A concept-dynamic-three-dimension-group is issued out. Any pixel is modeled by one dynamic-three-dimension-group. Updating decisions are made according to the pixel dynamic change information. The approach can update the background to adapt to slow or sudden changes in illumination and moved background objects automatically. The experimental results demonstrate the effectiveness of the proposed approach under complex scene.
出处
《计算机工程与设计》
CSCD
北大核心
2008年第4期894-895,共2页
Computer Engineering and Design
关键词
背景模型
计算机视觉
目标检测
背景差
图像分割
background model
computer vision
object detection
background subtraction
image segment
