摘要
针对传统的图像重构算法的不足,提出一种基于小波神经网络的图像重建快速学习算法。运用小波神经网络对图像重构进行了仿真研究。实验表明,对于不同的误差模型,小波神经网络采用不同的基函数可以很好地对非线性系统进行逼近,收敛速度快,近似精度高,而且网络规模比较小,计算量少。对计算机视觉和图像处理具有良好的应用价值。
The limitation of the conventional Lambertian reflectance model of the image rebuild is addressed and a new Wavelet Neural Network (WNN)-based reneconce model is proposed.The new neural learning algorithm is to optimize a proper renectance mode land to recover the object surface by a simple Shape-Form. Shading(SFS) . Variational method with this WNN-based mode land fuzzy method model. An example is also given to prove that the SFS technique's robust be most objects,even when the lighting conditions are uncertain. The simulation result shows the training speed of NN can be improved greatly.The method is general and can be applied exten sively.
出处
《计算机应用与软件》
CSCD
北大核心
2003年第2期48-51,共4页
Computer Applications and Software
