摘要
在复杂多变的环境下,大范围动态图像序列中会存在大量噪声和干扰向量,导致传统方法较难正确检测出图像帧间运动,容易产生错误结果。针对现实环境中干扰因素的特性,给出一种大范围动态图像序列帧间运动检测方法,构建检测模型,通过分析动态图像序列中干扰向量的波动性和周期性特征,确定检测帧当前状态,分类为波动性干扰窗口、运动窗口以及背景窗口,利用阈值分割与形态学处理,将所有运动窗口合并确定运动帧,完成帧间运动检测。通过仿真证明,上述方法能够有效降低动态图像序列中干扰向量,极大提升帧间运动检测的准确性和鲁棒性。
In complex environment,large-scale dynamic image sequence has a large number of noise and interference vectors.In traditional methods,it is difficult to detect inter-frame motion correctly,leading to erroneous results.Due to the characteristics of interference factors in real environment,this paper presented a method to detect inter-frame motion in large-scale dynamic image sequence.First of all,a detection model was constructed.Then,the current state of detection frame was determined by analyzing the fluctuation and periodicity of interference vectors in dynamic image sequence,including the fluctuation interference window,motion window and background window.In addition,threshold segmentation and morphology were used to combine all motion windows,so as to determine the motion frames.Thus,the inter-frame motion detection was completed.Simulation results prove that the proposed method can effectively reduce the interference vectors in dynamic image sequences and greatly improve the accuracy and robustness of inter-frame motion detection.
作者
童钰
TONG Yu(Department of Computer Science and Technology,Hubei Normal University,Huangshi Hubei 435000,China)
出处
《计算机仿真》
北大核心
2019年第12期391-395,共5页
Computer Simulation
关键词
图像帧间
运动检测
帧间差法
形态学处理
阈值
Interframe
Motion detection
Interframe difference method
Morphological processing
Threshold
