摘要
当前视频图像目标样本点密度取值误差较大,导致视频动态帧目标定位精度偏低。提出基于关键姿态的视频动态帧目标定位方法。采用高斯密度估计法构建视频动态图像关键姿态背景模型。分析视频序列像素点噪声概率密度,提取动态个体目标关键姿态特征轮廓,修正样本点密度取值,实现关键姿态映射下视频动态帧目标定位。为验证提出方法的性能,设计了实验。实验结果表明,提出方法下图像样本噪声得以有效抑制、目标定位精度高,可为相关领域研究提供可靠依据。
At present,the value error of target sample density in the video image is large,leading to the low positioning accuracy of the dynamic frame.Therefore,a method to locate the dynamic frame in video based on key attitude was put forward.Gaussian density estimation method was used to construct the key attitude background model of a dynamic image.The probability density of noise of the pixel in the video sequence was analyzed,and then the characteristic contour of the key attitude of the dynamic target was extracted.Moreover,the value of sample density was corrected.Finally,the target location of the dynamic frame under key attitude mapping was realized.In order to verify the performance of the proposed method,an experiment was designed.Experimental results prove that the proposed method can effectively suppress the noise in image samples and improve the target positioning accuracy.In addition,this method can provide a reliable basis for research in related fields.
作者
范洁
谢鑫
陈战胜
FAN Jie;XIE Xin;CHEN Zhan-sheng(College of Applied Science and Technology,Beijing Union University,Beijing 100101,China)
出处
《计算机仿真》
北大核心
2022年第3期156-159,248,共5页
Computer Simulation
基金
OBE成果导向的职业教育项目化课程标准开发与建设项目(12205561110-083)。
关键词
关键姿态
动态帧
视频图像
目标定位
核密度估计
高密度指标
Key posture
Dynamic frame
Video image
Target positioning
Kernel density estimation
High-density index