中文摘要：

基于小波变换与神经网络复合模型的图像清晰度识别方法具有较强的图像边缘特征提取、非线性处理、自适应学习和模式识别能力。提出一种通过神经网络模拟人眼的调焦机制，基于小波变换与神经网络复合模型，实现对图像清晰度评价的方法。利用二维离散小波变换对图像信号的特征进行提取，并对7个小波分量及原始图像做统计处理得到16个统计值，作为图像的特征量供后续的识别分析。构建了5层BP神经网络模型对图像清晰度进行识别，采用可变步长附加动量项的最速下降法调整网络权值。所设计的神经网络首先对由75幅图像组成的训练集合进行训练，再对102幅图像组成的测试集合进行实验验证。结果表明，这是一种相当有效的判别方法，取得了较高的识别率。

英文摘要：

The image definition identification method based on the composite model of wavelet transform and neural network is useful in image edge character extraction, nonlinear process, self-adapted study and pattern recognition. An evaluation method of image definition based on the focusing mechanism of simulating human＇s eyes by neural networks and on the composite model of wavelet transformation and neural network was suggested. The two-dimensional （2D） discrete wavelet transformation was used to extract image signal character, and 16 statistical values obtained from 7 wavelet components and an original image by statistical process were treated as image eigenvalues for the follow-up identification and analysis. 5 layers of BP neural network were constructed to perform image definition identification adopting a fastest descent method with an additional momentum item of variable step length to adjust network weight values. The designed neural networks firstly train the training set composed by 75 images and then perform experimental verification for the testing set composed by 102 images. The results show that this is a very effective distinguishing method and can achieve a high recognition rate.