节点文献
基于改进IBP算法的图像超分辨率重建
Super-resolution Image Reconstruction Based on Improved IBP Algorithm
【摘要】 针对传统IBP算法存在对图像细节获取能力差,重建图像清晰度不高的问题,提出一种改进方法。该方法利用小波包图像融合技术,获取所有低分辨率图像高频绝对值最大的值,与均值低频进行小波包重建,得到重建低分辨率图像。以该图像为参考图像,进行迭代反投影,获得高分辨率图像。改进后的算法增加了迭代反投影方法对高频信息的获取能力,提高了图像重建质量。
【Abstract】 Aiming at the problems that the conventional IBP algorithm has the poor ability to get the image details,and can not acquire the high definition degree of the image,an improved method is proposed.The method uses wavelet packet image fusion technology to obtain the values that have maximum high frequency absolute value of all the low resolution images,and then gets the reconstructed low resolution image by the wavelet packet reconstruction using the obtained value and the mean low-frequency value.Finally,the low resolution image is used as the reference image to carry out iterative back-projection and get high resolution images.It can been seen that the improved method enhances the ability of the iterative back-projection method to get high frequency information and improves the quality of the reconstructed images.
【Key words】 iterative back-projection; super-resolution; wavelet packet decomposition; image fusion;
- 【文献出处】 电子技术 ,Electronic Technology , 编辑部邮箱 ,2011年06期
- 【分类号】TP391.41
- 【被引频次】7
- 【下载频次】398