JPEG图像编码中质量因子选取方法研究

Research on the selection of quality factor in JPEG image coding

受信道带宽和存储空间限制,工程中经常需要将原始图像压缩为特定长度的文件。当前的主流图像压缩标准JPEG只能以质量因子作为约束条件进行图像压缩编码,却不能准确控制压缩后的文件长度。本文基于统计和线性拟合思想,设计出一种简便的质量因子选择方法,可将目标图像压缩至期望长度。仿真实验证实:使用本方法选出的质量因子进行图像压缩,可保证输出文件长度与期望值的平均误差在10%以内,基本满足国家某专项任务需求,同时具有简便快捷的特点。

Due to the limitation of channel bandwidth and storage space, compressing original images into specific-length files is frequently used in practice. The current image compression standard JPEG compresses the image only in terms of quality factor as a constraint, but cannot accurately control the length of the compressed file. Based on the idea of statistical and linear fitting, a simple quality factor selection method is proposed in this paper, by which the target image can be compressed to expected length. The simulations verify that the new method can guarantee the average error between the length of the output file and the expected length error of less than 10%, which satisfies the demand of a national special project, In addition, the new method is easy to implementation.