论文部分内容阅读
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。
The reconstruction effect of super-resolution image directly affects the recognition and processing of the image. In view of the complexity of super-resolution image and the low precision of traditional image reconstruction method, a new SVM method is proposed. Firstly, the sparse features of image blocks are calculated to get the sparse representation coefficients. Then the sparse representation coefficients are input into the SVM for training, and the super-resolution image reconstruction model is established. Finally, the performance of the image is analyzed by simulation experiments. The results show that this method can preserve the edge details of the image, improve the effect of super-resolution image reconstruction, and outperform other super-resolution image reconstruction methods.