基于图像序列分析的全局直方图均衡

Global Histogram Equalization Based on Image Sequence Analysis

针对传统直方图均衡算法的不足,如增强过度,均值漂移和细节丢失等,本文提出了一种基于图像序列分析的全局直方图均衡算法。受到CLAHE的启发,本文设计了两种数学模型。图像渐变模型将输入的单张图像输出为质量渐变的图像序列;再根据序列图像的统计指标,经过混合优化模型计算出最优控制参数。为了提升主观感知效果本文提出了一种直方图后处理方法,最后根据修正后的直方图进行均衡化处理。实验结果表明:本文算法的结果图像具有较好的主观感受度和客观评价值,并且优于近年来提出的直方图均衡改进算法。

Aiming at the defects of Traditional Histogram Equalization(THE)algorithm,for instance,over-enhancement,mean shift and detail loss,we proposed a method called global histogram equalization based on image sequence analysis in this paper.Respired by CLAHE(Contrast Limited Adaptive Histogram Equalization),this paper designed two mathematical models.The image gradient model got quality gradient images after inputting image.Then the hybrid optimization model got the optimal control parameter according to the statistical indicators of sequence images;To improve the subjective perception effect,we proposed a histogram post-processing method.Finally,the equalization process is performed according to the modified histogram.The results showed that the proposed method had good perception with both subjective and objective evaluation and it outperforms the state-of-the-art histogram equalization-based methods.