冶炼车间大气散射模型图像去烟尘算法  

Dust removal algorithm for atmospheric scattering model image of smelting workshop

作　　者：汤汶龙 龙永红[1] TANG Wenlong;LONG Yonghong(College of Railway Transportation,Hu’nan University of Technology,Zhuzhou,Hu’nan 412007)

机构地区：[1]湖南工业大学轨道交通学院,湖南株洲412007

出　　处：《电气技术》2023年第9期20-27,共8页Electrical Engineering

基　　金：湖南省自然科学基金(2023JJ50196)。

摘　　要：针对冶炼车间中大量烟尘、水雾等悬浮颗粒造成的图像降质等问题,本文提出冶炼车间大气散射模型图像去烟尘算法。为更好地估计真实大气光值,通过简单线性迭代聚类分割算法求取初始大气光值,并采用快速引导滤波对初始大气光值进行精细化处理,同时利用自适应伽马函数对大气光和原始烟尘图像进行修正,分别得到最终大气光和优化后的烟尘图像。通过优化的颜色衰减先验模型估计出透射率,最后根据大气散射模型恢复无烟尘图像。实验结果表明,该算法可降低图像中的烟尘浓度,减少图像细节损失,使方均误差平均下降66.2%,峰值信噪比平均提高30.5%,结构相似度平均提高48.6%。Aiming at the problem that the image is degraded by a large number of suspended particles such as soot and water mist generated in the smelting workshop,an dust removal algorithm for the atmospheric scattering model image of the smelting workshop is proposed.In order to better estimate the real atmospheric light value,the algorithm in this paper obtains the initial atmospheric light value by simple linear iterative clustering segmentation algorithm,and uses fast guided filtering to refine the initial atmospheric light value.At the same time,the adaptive gamma function is used to correct the atmospheric light and the original soot image,and the final atmospheric light and the optimized soot image are obtained respectively.The transmittance is estimated by the optimized color attenuation prior model.Finally,the smoke-free image is restored according to the atmospheric scattering model.The experimental results show that the algorithm can effectively reduce the smoke concentration in the image and reduce the loss of image details.The mean square error is reduced by 66.2%on average,the peak signal-to-noise ratio is increased by 30.5%on average,and the structural similarity is increased by 48.6%on average.

关 键 词：图像去烟尘 大气散射模型 简单线性迭代聚类 自适应伽马函数 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TF08[自动化与计算机技术—计算机科学与技术]