一种新的图像清晰度评价函数

New kind of clarity-evaluation-function of image

离焦模糊图像清晰度评价函数是采用数字图像处理技术实现自动调焦的一个关键,需要不断地提高评价函数的准确性和有效性。深入研究了各种图像梯度的分布情况后发现模糊图像小梯度像素数较大,而清晰图像大梯度的像素数则明显比模糊图像的多,因此可以给梯度加一个阈值,去掉梯度小的值保留梯度大的值,这样可以突出清晰图像的的优势,易于准确判断。首次提出了一种用图像梯度加阈值求和作为由于离焦产生的模糊图像的评价函数,建立了上述评价函数的数学模型,并给出了实验结果和分析。与以往的图像灰度方差、梯度和、小波变换等评价函数相比,给出的评价函数无偏性好、单峰性强,信噪比高,计算量小,在焦平面附近具有变化趋势明显和灵敏度高的特点。

The clarity-evaluation-function of out-of-focus blurred image is a key factor for auto-focusing by digital image processing and the accuracy and efficiency of the function need to be improved. It is found that the blurred image has large amount of pixels with low gradient while the focused image has obviously more pixels with high gradient than the blurred image after deep study of the gradients of various images. Therefore, we can threshold gradients in order to throw low gradients and remain high gradients. And then the advantages of focused image are obviously showed. A kind of clarity-evaluationfunction of out-of-focus blurred image is firstly proposed in the paper. It is the sum of Laplacian operator with threshold. The mathematical model is established individually, and the experimental results and analysis are also given. The clarity-evaluation-function of out-of-focus blurred image given here has the characteristics of non-deflection, single peak value, high sensitivity and SNR.