深度学习轻量化侦察图像压缩网络
Deep learning Based Lightweight Reconnaissance Image Compression Network
谌宇 1谌德荣 1崇魁奇 1王泽鹏 1张凯2
作者信息
- 1. 北京理工大学,北京 100081
- 2. 北京航宇天穹科技有限公司,北京 100043
- 折叠
摘要
为了满足小型侦察平台对低复杂度图像编码算法的应用需求,提出基于深度学习轻量化侦察图像压缩网络.轻量化侦察图像压缩网络编码端利用三个卷积模块直接将图像映射为服从均匀分布的二进制码流,得到压缩数据;在卷积模块中采用深度可分离卷积、分组卷积+通道重排等方式降低了编码端参数量和计算量.轻量化侦察图像压缩网络解码端采用转置卷积和残差连接等方式提高特征提取能力,进而提高解码图像质量.对分辨率为 128×128 实际采集图像的测试结果表明,与JPGE2000 算法相比,基于深度学习轻量化侦察图像压缩网络PSNR提高了 3.85 dB,编码时间降低了 91%,实现了图像的轻量化编码压缩.
Abstract
In order to meet the application requirement of low complexity image coding algorithm for small re-connaissance platforms,a lightweight reconnaissance image compression network based on deep learning was proposed.At the coding end of the lightweight reconnaissance image compression network,three convolution modules were used to map the image directly to the binary code stream conforming to uniform distribution,and the compressed data was obtained.In the convolutional module,depth-separable convolution,group convolution plus channel shuffle were adopted to reduce the number of coding end parameters and the amount of computa-tion.The decoder of lightweight reconnaissance image compression network applied to transposition convolution and residual connection to improve the ability of feature extraction and the quality of decoded image.Test results of 128×128 images showed that compared with JPGE2000,the PSNR of lightweight reconnaissance image com-pression network based on deep learning was increased by 3.85 dB,the coding time was reduced by 91%,and the lightweight coding compression of image was realized.
关键词
侦察图像压缩/深度可分离卷积/分组卷积/通道重排Key words
reconnaissance image compression/deep separable convolution/group convolution/channel shuffle引用本文复制引用
出版年
2024
NETL
NSTL
万方数据