改进蝙蝠算法在纸病图像增强中的应用

Application of Improved Bat Algorithm Optimization in Paper Defect Image Enhancement

为了辨识低照度条件下纸病图像的背景区域和纸病区域,提出一种基于蝙蝠算法优化的纸病图像增强方法。首先采用混沌映射和引力搜索算法分别更新个体响度、脉冲发射率和个体速度改进蝙蝠算法,然后将改进后的蝙蝠算法搜索伽马变换的最佳参数γ,利用该γ值调整图像灰度,提高图像对比度。结果表明,与基于传统直方图均衡算法的纸病增强算法、基于限制对比度自适应直方图均衡算法的纸病图像增强算法、基于标准蝙蝠算法优化的纸病图像增强算法、基于混沌蝙蝠算法优化的纸病图像增强算法相比,本文算法的对比度、灰度方差乘积函数、均值大小都有所提升,增强效果为所有对比算法中最优。

In order to identify the background area and the paper defect area of the paper defect image under low illumination conditions,a paper defect image enhancement method based on bat algorithm optimization is proposed.Firstly,chaotic mapping and gravitational search algorithms are used to update the individual loudness,pulse emission rate and individual velocity to improve the bat algorithm.Secondly,the improved bat algorithm is used to search for the best parameterγof the Gamma transformation,and use this value to adjust the image grayscale and improve the image contrast.The results show that compared with the paper defect enhancement algorithm based on the traditional histogram equalization algorithm,the paper defect image enhancement algorithm based on the limited contrast adaptive histogram equalization algorithm,the paper defect image enhancement algorithm based on the standard bat algorithm optimization,the paper defect image enhancement algorithm based on the chaotic bat algorithm optimization have been improved:the contrast,grayscale variance product function,and mean size of the algorithm,and the enhancement effect is the best among all contrast algorithms.