[摘要] 提出了一种高效的、基于深度卷积神经网络(CNN)的图像去模糊算法。网络结构基于条件生成对抗网络,并使用堆叠的自编码器结构与跳跃相连接。相关的试验结果表明:该算法有良好的图像去模糊效果,并且能够大幅度地降低时间与内存开销。
[关键词] 图像去模糊;卷积神经网络;对抗生成网络
[Abstract] In this paper, an efficient deep convolutional neural network (CNN)-based image deblurring method is proposed. The network architecture is based on conditional generative adversarial network integrated with stacked encoder-decoder architecture and skip connections. Experiment results show that the proposed method achieves good image deburring performance and in the meanwhile reduce the testing time and required memory resource.
[Keywords] image deburring; CNN; generative adversarial network