摘要
针对图像重建误差较大的问题,提出了一种基于BP神经网络学习算法的图像压缩方法。分析BP网络模型结构和网络误差,设计了该算法实现的流程,并进行了算法测试,讨论了不同隐层神经元个数和子块大小对图像重建的影响。实验结果表明,该学习算法能较好地实现对图像的压缩,网络训练效果较好,误差较小。
To the problem of large image reconstruction error,an image compression method is proposed on BP neural network learning algorithm.By analyzing the structure and network errors of the BP network model,it is designed the flow of the algorithm,and tested with discussing the impact of the number of different hidden layer neurons and the size of sub-blocks on image reconstruction.The results show that the learning algorithm can compress well,the effect of network training is better,and the error is smaller.
作者
李亚文
何建强
LI Ya-wen;HE Jian-qiang(College of Electronic Information and Electrical Engineering,Shangluo University,Shangluo 726000,Shaanxi)
出处
《商洛学院学报》
2020年第4期1-9,共9页
Journal of Shangluo University
基金
陕西省教育厅专项科研计划项目(19JK0261)
商洛学院科研创新团队(19SXC03)。
