衍射光学可变图像去云雾算法的研究与应用

发布时间：2018-01-28 19:08

本文关键词： 衍射光学图像去雾雾罩层　出处：《激光杂志》2017年03期 　论文类型：期刊论文

【摘要】：衍射光学图像研究领域中针对衍射光学图像去云雾处理的研究,一直被作为衍射光学图像科学研究领域中重要的一项研究项目。近年来,在不断的应用实践中发现,传统衍射光学图像去云雾方法中采用算法存在光线可变度逻辑运算差、图像特征处理检索速度慢、衍射光学可变图像去云雾处理效果差等问题。针对传统衍射光学图像去云雾处理方法存在的问题,提出衍射光学可变图像去云雾算法设计,采用三维一体化设计对传统算法进行改进、优化。分别采用衍射光线性动态分布算法(XIVNW)、像素特征捕捉技术(BJNEF)、雾罩层提取剥离技术对传统算法进行改进升级。通过实验证明,提出的衍射光学可变图像去云雾算法设计具有衍射光学图像去云雾效果好、资源开销少、处理速度快等特点。
[Abstract]:In the field of diffractive optical image research, the research on decloud processing of diffractive optical image has been regarded as an important research project in the field of diffraction optical image science in recent years. In the continuous application practice, it is found that in the traditional diffraction optical image de-cloud method, the algorithm has poor logic operation of variable degree of light, and the retrieval speed of image feature processing is slow. Aiming at the problems existing in the traditional dehaze processing method of diffractive optical images, the design of the algorithm for dehaze removal from diffractive optical variable images is presented. Three dimensional integrated design is adopted to improve and optimize the traditional algorithm. The diffractive light linear dynamic distribution algorithm (XIVNWN) and pixel feature capture technique (BJNEF) are adopted respectively. Through experiments, it is proved that the design of the diffractive optical variable image de-fog algorithm has good effect of diffractive optical image removing cloud and mist, and the cost of resources is less. Fast processing speed and other characteristics.
【作者单位】：聊城大学东昌学院;天津中德应用技术大学电气与能源学院;
【基金】：天津市应用基础与前沿技术研究计划重点项目(15JCZDJC39000) 国家科技支撑计划项目(2015BAK06B04)
【分类号】：TP391.41;O436.1
【正文快照】： 随着光学研究领域的不断深入发展,以及光学图像清晰度不断提升,衍射光学图像被诸多领域所采用。对推动我国航天航空事业发展、拉动国民经济有着重要作用。衍射光学图像研究领域中,云雾是自然界中最常见的可见性干扰源中的一种。含有云雾的衍射光学图像对比度低,实体景物成像效

【相似文献】