液晶与显示, 2010, 25 (3): 454, 网络出版: 2010-08-31
基于高斯背景建模的目标检测技术
Object Detecting Technology Based on Gauss Background Modeling
摘要
研究了室内环境下的运动目标的检测技术。讨论了各种目标检测的算法,提出基于室内相对静止背景下的目标检测算法,利用高斯背景建模方法提取较为理想与“干净”的背景帧,并可进行背景的自适应更新,然后再利用所得的背景与当前视频帧进行相减,得到视频中运动目标的前景图像,经过平滑去噪处理得到较为理想的目标检测效果。
Abstract
The moving target detecting technology based on the in-door background was studied in this paper. Various kinds of object detecting algorithms were discussed, and the object detecting algorithm based on relatively immobile in-door background was put forward. The Gauss background modeling was used to distill perfect background frame which could be adaptive updated. Then the subtraction between the background frame and current frame was made to get the foreground of moving target in the video. Finally, the ideal target detecting result could be obtained after the noise was smoothed and eliminated.
康晓晶, 吴谨. 基于高斯背景建模的目标检测技术[J]. 液晶与显示, 2010, 25(3): 454. KANG Xiao-jing, WU Jin. Object Detecting Technology Based on Gauss Background Modeling[J]. Chinese Journal of Liquid Crystals and Displays, 2010, 25(3): 454.