摘要
由于太阳耀斑图像中地球云雾对斑迹有所干扰,故需要对太阳耀斑图像进行去云处理。针对多尺度Retinex理论与暗通道优先算法两种去云方法进行比较研究,详细介绍了两种方法的数学模型与具体算法,分别运用两种方法对太阳耀斑图像进行去云处理,并进行MATLAB仿真,根据仿真结果对两者去云效果进行评估与对比。通过图像与客观误差参数比较两种去云方法,发现暗通道优先算法在影像平均灰度、细节信息显示、图片信息量与相对清晰程度等方面均优于多尺度Retinex去云方法,其中图片清晰度提升了约16%,标准差与信息量均提升了约8%,平均灰度降低了约27%。
Due to the interference of the earth's cloud in the solar flare image,it is necessary to remove cloud from the solar flare imag-es. The two methods of MSR theory and dark channel prior algorithm are compared. The mathematical models and specific algorithms of the two methods are introduced in detail. Two methods are used to process the solar flare image and perform MATLAB simulation. Ac-cording to the simulation results,the cloud-removing effects of the two methodsare evaluated and compared. Comparing the two meth-ods of cloud removal by image and objective error parameters,it is found that the dark channel priority algorithm is better than the multi-scale Retinex cloud method in displaying the average gray level,detailed information display,picture information and relative clarity of the image. Among them,the picture clarity is improved by about 16%,the standard deviation and information are improved by about 8%,and the average gray level is reduced by about 27%.
作者
王凯
李婉卿
熊炜
邹宇
WANG Kai;LI Wan-qing;XIONG Wei;ZOU Yu(College of Electrical Engineering & New Energy,China Three Gorges University,Yichang 443000,China;Guangxi Power Grid Qinzhou Power Supply Bureau,Qinzhou 535000,China)
出处
《软件导刊》
2019年第6期163-166,170,共5页
Software Guide
基金
广西电网公司科技项目(0406002018030103SJ100011)
