基于直方图相关的图像灰度校正  

Image gray calibration based on histogram correlation

作　　者：邓志鹏[1] 许丽敏[1] 杨杰[1] 全勇[1] 

机构地区：[1]上海交通大学图像处理和模式识别研究所,上海200030

出　　处：《红外与激光工程》2003年第2期167-169,共3页Infrared and Laser Engineering

摘　　要：在对遥感图像进行地貌分析或匹配时,图像的灰度校正是必不可少的环节,其结果直接影响后续处理。提出基于线性点运算的直方图相关的灰度校正算法,首先将待校正图与预期结果图进行直方图灰度级的一致化,求解斜率系数,然后通过直方图相关方法解出平移量,最后利用线性点运算完成图像的灰度校正。结果表明,算法对由光照条件引起的灰度变化的校正具有较高的稳健性。For analyzing land covers and matching of remote sensing images, gray calibration of image is an indispensable part, for the result will directly influence the later procession. An algorithm of image gray calibration based on histogram correlation with linear point operations is put forward in three steps. Firstly, solve the slope ratio by making histogram of image for calibrating and expecting consistency. Secondly, solve the amount of shift by histogram correlation. Thirdly, realize gray calibration by using linear point operations. The result shows that the method has better stability for gray changing caused by illumination. \;

关 键 词：灰度校正 直方图相关 点运算 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]