一种采用MTF编程代码量化图像清晰度等级的方法

在photoshop中对三幅不同风格的数字图像进行两种增大反差的操作,提高了图像的清晰度.另外,本文的重点是使用了MTF这一检测指标,在计算机中进行编程,以MTF曲线和坐标轴所围面积S为数据,量化了图像清晰度等级.结果表明,S值越大,理论图像清晰度越高；增大反差保留的半径大小后,S值也相对增大,但增大量减小；叠加反差图层能更快的提高图像清晰度.因此,对于细节较多的图像采取增大半径的方法,而细节较少则采取叠加图层的方法,对于计算机图像处理有一定的指导意义.

A method by using MTF programme code to quantify image definition levels

In this text,two methods were used to increase contrast of three different styles digital images in photoshop,which improve the image definition.In addition,importantly,MTF was selected to detect image,we programme it and use area S that MTF curves suround axes as data to quantify image definition levels.The results shows that——when S becomes larger,image definition is higher theoretically.S is increasing with radius value,but expanding value is decreasing.Besides,repeating contrast layers can improve image definition faster.Therefore,reinforcing radius is better method for more details images,while,repeating contrast layers is suited for less details images,which has directive significance for computer image processing.