草地视景图像质量优化仿真方法研究

Optimization of Grass Visual Image Quality and Its Simulation

草地视景图像的仿真质量关系到后期的处理,当草地视景图像层状纹理几何细节构成发生变化,并且跳变程度呈现高随机性时,采用传统方法进行草地视景图像跳变抵消。对视觉跳变随机程度不容易控制,测度不准确,导致视觉跳变抵消效果差、失真程度高的问题。为提高草地图像视景的准确性,提出采用高阶累积量纹理切片的草地视景仿真视觉跳变抵消算法,采用多频分离提升方法,实现对草叶递阶偏移视觉跳变程度的控制,融入模板叠加和相异特征匹配的虚化反混淆算法对数据进行处理,采用高阶累积量纹理切片进行计算,得到视觉跳变抵消迭代方程。仿真结果表明,采用改进算法进行大规模草地视景仿真,跳变纹理像素点明显减少,实现不同地形下的草地动态视景生成图像,绘制执行效率和生动逼真性等方面优于传统算法,可以满足较高的视觉要求。

A visual jump offset algorithm of grass visual simulation is presented based on high - order cumulants of the texture slice, the multi- frequency isolation and ascension methods are used to control the grass visual jump degree, the antialiasing algorithm is used to deal with the data by matching the hierarchical template superposition and different features, and the high - order cumulants of texture slice are used to obtain the visual jump offset iterative equation. The simulation results show that the improved algorithm can significantly decrease jump texture pixels, and has higher efficiency and vividness than the traditional algorithms.