计算机科学 ›› 2014, Vol. 41 ›› Issue (6): 295-298.doi: 10.11896/j.issn.1002-137X.2014.06.059

• 图形图像与模式识别 • 上一篇    下一篇

约束优化进化的夜间图像时频复合加权提取

刘淑琴,彭进业   

  1. 西北大学信息科学与技术学院 西安 710127;西北工业大学电子信息学院 西安710072
  • 出版日期:2018-11-14 发布日期:2018-11-14
  • 基金资助:
    本文受国家自然科学基金(61075014)资助

Extraction of Nighttime Images with Time Frequency Weighted and Constrained Optimization Evolutionary Algorithm

LIU Shu-qin and PENG Jin-ye   

  • Online:2018-11-14 Published:2018-11-14

摘要: 研究了一种约束优化进化的夜间图像时频复合加权提取技术。对多帧夜间图像的时域与频域同时加权处理是一种深度检索夜间图像信息的先进技术。夜间图像质量低,传统的夜间图像信息检索技术采用单独的时域检索方法,无法对多帧夜间图像的频谱特性进行分析和信息检索。为此,提出一种约束优化进化的夜间图像时频复合加权提取技术。首先在频域和时域同时对多帧原始图像进行处理,提取多帧图像之间的相关信息;然后做加权处理,形成新的图像特征;在此基础上,通过约束优化进化算法对图像信息检索的结果不断进行循环优化,最终达到较好的效果。采用一组夜间图像进行了实验测试,结果显示,采用约束优化进化的夜间图像时频复合加权提取,多帧图像之间的时域和频域特征得到了很好的加权利用,最终实现了图像信息的深度提取。该方法在图像检索领域具有很好的应用价值。

关键词: 约束优化进化,图像信息提取,时频复合加权 中图法分类号TP399文献标识码A

Abstract: The extraction of nighttime images with time frequency weighted and constrained optimization evolutionary algorithm was studied.The treat ment of multi-frame nighttime images in time domain and frequency domain is an important method for information extraction.In traditional processing method,the low quality nighttime images are treated with a separate time-domain retrieval method,so the whole domain information cannot be used.The extraction of nighttime images with time frequency weighted and constrained optimization evolutionary algorithm was proposed.Firstly,the nighttime images were processed in the time domain and frequency domain simultaneously.Then the multi-frame images were weighted,and the new feature was formed.On this basis,the effect of the images was cyclically improved through constrained optimization evolutionary algorithm,and ultimately a better result was achieved.A team of nighttime images was used to test the ability.The result shows that the information of time domain and frequency domain can be used well,and the image information is extracted with good performance.The algorithm has good value for image information extraction.

Key words: Constrained optimization evolutionary,Image information extraction,Time frequency weighted

[1] 李敏,李俊.基于人类视觉系统特性的图像质量评价算法[J].科技通报,2013,29(2):160-162
[2] 雷亮,汪同庆,杨波.图像关联规则挖掘研究[J].计算机应用研究,2009(6):2374-2376
[3] 李艳玲,黄春艳,赵娟.基于灰色关联度的图像自适应中之滤波算法[J].计算机仿真,2010,27(1):238-240
[4] Cai Z,Wang Y.A multiobjective optimization based evolutionary algorithm for constrained optimization[J].IEEE Trans.on Evolutionary Computation,2006,10(6):658-675
[5] Runarsson T P,Yao X.Search biases in constrained evolutionary optimization[J].IEEE Trans.on Systems,Man,Cybernetics (C),2005,35(2):233-243
[6] 舒风笛,王敏,毋国庆.图像数据关联规则挖掘[J].小型微型计算机系统,2001(11):1353-1356
[7] 杜辉.基于小波变换的彩色图像中快速人脸检测算法[J].科技通报,2012,8(12):89-90
[8] Yu J X,Yao X,Choi C,et al.Materialized view selection as constrained evolutionary optimization[J].IEEE Trans.on Systems,Man,and Cybernetics (C),2003,3(4):458-467
[9] 杨赛,赵春霞.图像分类中的概率乘积核函数[J].中国图象图形学报,2013,8(4):45-47
[10] 雷庆,李绍滋.动作识别中局部时空特征的运动表示方法研究[J].计算机工程与应用,2011,6(34):7-10
[11] 何友,刘永,孟祥伟.杂波图CFAR平面技术在均匀背景中的性能[J].电子学报,2010,7(3):119-120,3
[12] 朱志刚,徐光祐,杨波.自动交通监测系统的二维时空图象方法[J].中国图象图形学报,2011,1(2):101-107
[13] 王扬扬,李一波,姬晓飞.人体动作的超兴趣点特征表述及识别[J].中国图象图形学报,2013,8(7):805-812
[14] Ge Ji,Wang Yao-nan,Zhang Hui,et al.Research on Pixel Probability Statistics based Background Modeling Algorithm Applied in Liquid Foreign Particle Inspection Machine[J].IJACT(J),2013,5(1):468-476
[15] Hua Zhen,Li Ye-wei,Li Jin-jiang.Image Salient Region Extraction Algorithm Based on Improved Visual Attention Model[J].JCIT (J),2011,6(5):280-290
[16] 沈垣,王汉全,毛建国.数字图像相关方法的大变形初值估计[J].重庆理工大学学报:自然科学版,2013,7(11):86-90

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!