利用灰度差估计的条纹非均匀性校正方法  

Fringe Non-Uniformity Correction Method Based on Gray Difference Estimation

作　　者：徐泽林 路东明[1] 王利平[1] 顾国华[1] Xu Zelin;Lu Dongming;Wang Liping;Gu Guohua(School of Electronic Engineering and Optoelectronics Technology,Nanjing University of Science and Technology,Nanjing,Jiangsu 210094,China)

机构地区：[1]南京理工大学电子工程与光电技术学院,江苏南京210094

出　　处：《光学学报》2021年第5期20-26,共7页Acta Optica Sinica

摘　　要：条纹非均匀性是红外焦平面阵列成像系统中一种较为常见的固定图案噪声,其对成像质量会产生很大的影响。通常认为图像的非均匀性主要体现在像素间的灰度差,即非均匀性所引起的灰度差。为了解决这一问题,提出一种利用灰度差估计的条纹非均匀性校正方法。该方法根据条纹噪声的空间特性,对每一帧图像的非均匀性条纹的所在位置进行判断;根据条纹位置处相邻像素间的灰度差,对该处存在的非均匀性进行初步估计,并对下一帧的估计结果进行评价以进一步完善估计值。实验结果表明,所提方法能够显著减少条纹非均匀性,并且能够有效地保护图像的边缘信息。Stripe non-uniformity is a relatively common fixed pattern noise in infrared focal plane array imaging systems,which has a great impact on imaging quality.It is generally believed that the non-uniformity of an image is mainly reflected in the gray level difference between pixels,that is,the gray level difference caused by the non-uniformity.In order to solve this problem,a correction method of stripe non-uniformity based on gray difference estimation is proposed.According to the spatial characteristics of the strip noise,the position of the non-uniform strip in each frame is judged.According to the gray level difference between adjacent pixels at the position of the strip,the non-uniformity is preliminarily estimated,and the estimation results of the next frame are evaluated to further improve the estimated value.Experimental results show that the proposed method can significantly reduce fringe non-uniformity and can effectively protect the edge information of the image.

关 键 词：成像系统 图像处理 红外焦平面阵列 灰度差估计 条纹非均匀性校正 空间特性 

分 类 号：TN215[电子电信—物理电子学]