论文部分内容阅读
视频图像的高效无损压缩在海量的航空和遥感图像传输、珍贵的文物信息的保存等方面具有重要的应用价值,而目前的研究热点主要针对有损压缩,为此通过对现有的无损压缩方法的分析和研究,提出一种在稀疏编码与二进神经网络相结合的框架下建立新的图像无损压缩方法。首先借助二进神经网络中的线性可分结构系建立冗余字典,获得有效的稀疏分解基;再借助二进神经网络学习算法将图像映射为以线性可分结构系为神经元的二进制神经网络,在此基础上建立相应的模式匹配算法将每个神经元与冗余字典简历映射关系,通过稀疏系数建立原始图像的编码形式,进而实现了图像的无损压缩,并从理论上分析了该方法可以有效地提高压缩比,最后通过实验验证了该算法的有效性和通用性。
Efficient lossless compression of video images plays an important role in massive aviation and remote sensing image transmission and preservation of valuable cultural relics. At present, the research focuses mainly on lossy compression. To this end, the existing lossless compression methods This paper proposes a new image lossless compression method under the framework of sparse coding and binary neural network. Firstly, a redundant dictionary is constructed by using the linear separable structure in the binary neural network to obtain an effective sparse decomposition basis. Then, the binary neural network learning algorithm is used to map the image into a binary neural network with linear separable structure as a neuron , Based on which a corresponding pattern matching algorithm is established to map the relationship between each neuron and the redundant dictionary resume, and the original image encoding format is established by using sparse coefficients, and then the image is losslessly compressed, and the method is theoretically analyzed Can effectively improve the compression ratio, and finally verify the effectiveness and versatility of the algorithm by experiments.