基于最优分割的矩匹配图像非均匀性校正方法
CSTR:
作者:
作者单位:

(1.重庆光电技术研究所,重庆 400060;2.西安电子科技大学集成电路学部,西安 710126;3.电子科技大学集成电路科学与工程学院,成都 611731;4.集成电路与微系统全国重点实验室,重庆 400060)

作者简介:

李金(1982—),男,高级工程师,主要从事图像传感器测试技术研究;李明(1984—),男,正高级工程师,主要从事集成式图像传感器、光电探测器等研究;傅婧(1993—),女,博士,高级工程师,主要从事半导体器件及集成电路辐射效应机理与抗辐射加固技术。

通讯作者:

中图分类号:

TP391.41

基金项目:

国家自然科学基金青年科学基金项目(12305311);重庆市自科基金面上项目(CSTB2024NSCQ-MSX1189).


Image Non-Uniformity Correction Method Based on Optimal Segmentation Moment Matching
Author:
Affiliation:

(1. Chongqing Optoelectronics Research Institute,Chongqing 400060,CHN;2. Faculty of Integrated Circuits,Xidian University,Xi’an 710126,CHN;3. School of Integrated Circuit Science and Engineering,University of Electronic Science and Technology of China,Chengdu 611731,CHN;4. National Key Laboratory of Integrated Circuits and Microsystems,Chongqing 400060,CHN)

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    针对图像传感器像元固有缺陷引发的条纹噪声严重影响成像质量、传统定标法需反复标定、全局矩匹配易过度平滑图像导致细节丢失等问题,同时为了在有限的硬件资源平台上实现图像的实时处理,提出一种基于最优分割的矩匹配非均匀性校正方法。首先,对输入图像进行列特征预处理;随后,通过 Fisher算法实现图像的保序分块;接着,对得到的初始列块进行相似块合并优化,采用查并集算法将差值小于设定阈值的列块合并;最后,对每个列块执行矩匹配校正,得到输出图像。实验结果表明,该算法将平场图像的响应非均匀性(PRNU)由 4. 86%降至 0. 80%、暗信号非均匀性(DSNU)由 1. 73%降至 0. 77%;针对叠加列条纹的标准细节图像,校正后峰值信噪比(PSNR)由 22. 48 dB提升至 33. 21 dB,较两点法(31. 55 dB)和全局矩匹配(26. 17 dB)分别提升 1.66 dB和 7.04 dB;同时给出分割段数 k与两个阈值的实验论证与敏感性分析结果,并在真实传感器采集图像上完成了验证。所提方法可有效降低光响应非均匀性,对原始图像的细节具有良好的保真能力,且算法易于移植、便于硬件实现,能够满足小面积、低功耗的应用需求。

    Abstract:

    To address severe stripe noise from inherent pixel defects in image sensors and the limitations of conventional correction approaches,including frequent recalibration and detail loss induced by over-smoothing in global moment-matching, this paper presents a non-uniformity correction method based on optimal-segmentation moment matching for real-time processing under limited hardware resources. Initially,the input image is preprocessed via column feature analysis, followed by order-preserving blocking using the Fisher algorithm. Similar column blocks with differences below the threshold are merged using the union-find algorithm,and moment-matching correction is executed on each block to generate the corrected output. The experimental results show that the proposed method reduces the PRNU of the flat-field image from 4. 86% to 0. 80% and the DSNU from 1. 73% to 0. 77%;for a standard detail image with column stripes,the PSNR is improved from 22. 48 dB to 33. 21 dB,which is 1. 66 dB and 7. 04 dB higher than that of the two-point method(31. 55 dB)and global moment matching(26. 17 dB),respectively. The selection of segment number k and the two thresholds is further justified by experimental verification and sensitivity analysis,and the method is verified on the images captured by a real image sensor. It effectively suppresses photo-response non-uniformity while preserving image details,and features favorable portability and hardware-friendliness for small-area low-power scenarios.

    参考文献
    相似文献
    引证文献
引用本文

李金,孙铭阳,李明,傅婧,刘戈扬,倪飘,蒋君贤,任思伟.基于最优分割的矩匹配图像非均匀性校正方法[J].半导体光电,2026,47(5):975-982. LI Jin, SUN Mingyang, LI Ming, FU Jing, LIU Geyang, NI Piao, JIANG Junxian, REN Siwei. Image Non-Uniformity Correction Method Based on Optimal Segmentation Moment Matching[J].,2026,47(5):975-982.

复制
分享
相关视频

文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2026-09-04
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2026-10-23
  • 出版日期:
文章二维码