摘要
通过对X光图像所产生畸变的分析,提出了一种基于泛化能力的神经网络系统整体校正方法。该方法只需确定网络输入输出和约束条件,无需考虑中间过多的不确定因素,即可建立空间点与图像点的映射关系,然后采用双线性变换进行灰度插值。实验表明,该方法校正效果好,能够满足对图像进行分析处理的要求。
By the analysis of Xray image distortion,this paper presents a neural network system correction method based on global optimization.This method requires the input,output and restriction conditions without considering the overmany uncertain factors,and can set up the mapping relations between space points and image points, then adopts a bilinear interpolation method for graylevel determination.Experimental results indicate that this method has good correction effect,and can meet the demands of image analysis and processing.
出处
《机械与电子》
2005年第3期38-41,共4页
Machinery & Electronics
关键词
X光图像
畸变校正
神经网络
Xray image
distortion correction
neural network
