基于烟花算法的平面视觉图像分块像素点压缩模型  

Plane visual image block pixel compression modelbased on fireworks algorithm

作　　者：张欣[1] 李仁伟[1] ZHANG Xin;LI Renwei(North University of China,Taiyuan 030000,China)

机构地区：[1]中北大学,山西太原030000

出　　处：《现代电子技术》2023年第19期61-64,共4页Modern Electronics Technique

摘　　要：为进一步提升图像压缩质量,降低图像存储空间的同时减少图像信息损失,提出基于烟花算法的平面视觉图像分块像素点压缩模型。采用基于图像纹理的自适应分块方式,通过不断对比分块图像的纹理复杂度与设定复杂度阈值,对平面视觉图像进行分块处理;经块截断编码方法获取分块平面视觉图像最优公共位图后,形成向量,利用烟花算法寻优出最优向量值以及对应位图的6个量化值,完成平面视觉图像分块像素点压缩。实验表明:该模型压缩后平面视觉图像与原始图像相比,色彩亮度几乎一致,图像信息几乎没有缺损,压缩后的图像峰值信噪比高。In order to further improve the compression quality of image and reduce the loss of image information while reducing the storage space of image,a plane visual image block pixel compression model based on fireworks algorithm is proposed.An adaptive blocking method based on image texture is used to block the plane visual images by constantly comparing texture complexity of image blocks and setting complexity threshold.After obtaining the optimal common bitmap of the plane visual image blocks with the block truncation coding method,the vector is formed.The fireworks algorithm is used to find out the optimal vector value and 6 quantized values of the corresponding bitmap,so as to complete the block pixel compression of the plane visual image.The experimental results shows that,in comparison with the original image,the color brightness of the plane vision image compressed by this model is almost the same,its image information is almost intact,and its peak signal⁃to⁃noise ratio(SNR)is high.

关 键 词：图像压缩 图像分块 公共位图 烟花算法 平面视觉 图像纹理 图像信息 变异火花 

分 类 号：TN911.73-34[电子电信—通信与信息系统] TP391[电子电信—信息与通信工程]