摘要
图像超分辨率的重建效果直接影响图像进一步识别和处理,针对超分辨率图像的复杂性以及传统图像重建方法精度低的难题,提出了一种支持向量机的超分辨率图像重建方法。首先计算图像块的稀疏特性,得到稀疏表示系数,然后将稀疏表示系数输入支持向量机进行训练,建立超分辨率图像重建模型,最后采用仿真对比实验对其性能进行分析。结果表明,本文方法能够保持图像的边缘细节信息,改善了超分辨率图像重建的效果,并且比其它超分辨率图像重建方法的性能更加优越。
Super-resolution image reconstruction results directly affect the image recognition and processing, aiming at the problem that the complexity of the super resolution image and the low accuracy of the traditional image reconstruction method, a new method based on support vector machine is proposed. Firstly, the sparse representation of image block is calculated, and the sparse representation coefficients are obtained, and secondly, the sparse representa- tion coefficients is input to support vector machine and training and super resolution image reconstruction model is es- tablished, lastly, the performance is analyzed by simulation comparison experiment. The results show that the method in this paper can preserve the edge details of the image, and improve the effect of super-resolution image reconstruction, and the performance of the method is better than other method.
出处
《激光杂志》
北大核心
2016年第1期138-141,共4页
Laser Journal
基金
四川教育厅基金项目(13ZB0213)
关键词
超分辨率
支持向量机
图像重建
峰值信噪比
super resolution
support vector machine
image reconstruction
peak signal to noise ratio