多自由度人体运动姿态多目标图像重构仿真  

Simulation of Multi Object Image Reconstruction of Human Motion Pose with Multi Degree of Freedom

作　　者：刘卓 黄国阳[1] LIU Zhuo;HUANG Guo-yang(School of Information Technology,Guilin University of Electronic Science and Technology,Guilin Guangxi 541000,China)

机构地区：[1]桂林电子科技大学信息科技学院,广西桂林541000

出　　处：《计算机仿真》2021年第7期194-197,208,共5页Computer Simulation

基　　金：独立学院高水平篮球队建设现状及发展对策研究—以桂北片区为例(XJ202022)。

摘　　要：当前方法在进行人体运动姿态多目标图像重构时的精准度低、效果差,影响了应用效果。提出多自由度人体运动姿态多目标图像重构方法。计算图内的目标函数、像素灰度值进行,构建可以抑制噪声的图像重构模型;利用基于生物自然选择和遗传机理的遗传算法对图像重构模型进行求解,通过随机增加数值对像素群体的灰度值进行初始化,使图像重构效果更加出色;引入图像重构模型对适应值进行调整提高图像重构的精准度,采用赌盘选择的方法实现染色体的杂交、变异,将优秀的染色体复制到下一代群体中。经过拥挤度的计算实现图像重构。实验结果表明,所提方法在实践中拥有更加出色的重构效果、实时性和更高准确性。The currently method has low accuracy and poor effect in multi-object image reconstruction of human motion posture.In this regard,this paper proposes a multi-objective image reconstruction method of multi degree of freedom human motion posture.Firstly,the objective function and the gray value of the pixel in the image were calculated to establish the image reconstruction model which suppresses the noise.Secondly,according to the biological natural selection and genetic algorithm,the image reconstruction model was solved.In order to improve the effect of image reconstruction,the gray value of pixel group was initialized by randomly increasing the value.Then,the image reconstruction model was introduced to adjust the fitness value and improve the accuracy of image reconstruction.According to the method of roulette selection,chromosome hybridization and mutation were completed.Excellent chromosomes were copied to the next generation,and finally,the crowding degree was calculated to complete the image reconstruction.The results show that this method has excellent reconstruction effect,real-time performance and high accuracy.

关 键 词：图像重构 遗传算法 人体运动姿态 图像重构模型 像素灰度值 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]