首页|基于FLFR-SCP算法的井下图像去雾研究

基于FLFR-SCP算法的井下图像去雾研究

扫码查看
煤矿井下环境复杂,煤炭的掘进生产、除尘降尘等环节容易产生大量的粉尘颗粒和雾气悬浮在空气当中,导致井下监控系统无法获得较为清晰的视频图像。基于暗道先验的去雾算法无法完成井下图像的实时处理和尘雾图像的局部去雾。对暗道先验算法进行改进,改进透射率图的处理算法、获取大气光值算法、重新设计大气透射率函数和局部尘雾处理算法。实验表明,改进后算法能够较为准确地计算出大气光值和透射率,可以获得较为清晰的无雾图像,改进后图像的处理时间有较大幅度缩短,基本满足井下实时处理的要求。针对井下图像尘雾区域不均匀的问题,算法也有较好的处理效果。
DEHAZING OF DOWNHOLE IMAGE BASED ON FLFR-SCP ALGORITHM
The underground environment of coal mines is complex.During coal excavation production,dust removal and dust reduction,a large number of dust particles and mist are suspended in the air,resulting in the underground monitoring system unable to obtain clearer video images.The defogging algorithm based on the dark tunnel prior can not complete the real-time processing of underground images and the partial defogging of dust and fog images.The processing algorithm of transmittance map,the algorithm of obtaining atmospheric light value were improved,and the algorithm of atmospheric transmittance function and local dust fog processing were redesigned.Experiments show that the improved algorithm can calculate the atmospheric light value and transmittance more accurately,and can obtain clearer fog-free images.The improved image processing time is greatly shortened,which basically meets the requirements of downhole real-time processing.Aiming at the problem of uneven dust and fog in underground images,the algorithm also has a good processing effect.

Dark tunnel priorReal-time processingDownhole image defoggingBilateral filteringSmart mine

任国强、韩洪勇、李成江

展开 >

山东科技大学电气信息系 山东济南 250031

北京邮电大学人工智能学院 北京 100089

暗道先验 实时处理 井下图像去雾 双边滤波 智慧矿山

山东省高等学校科技计划项目山东省重点研发计划项目(攻关)

J18KA3282016GGX105013

2024

计算机应用与软件
上海市计算技术研究所 上海计算机软件技术开发中心

计算机应用与软件

CSTPCD北大核心
影响因子:0.615
ISSN:1000-386X
年,卷(期):2024.41(8)
  • 10