基于结构相似性粗定位与背景差分细分割的运动目标检测方法
DOI:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

TP391

基金项目:

国家自然科学基金(61971339);陕西省科技厅重点研发计划(2019GY-113);


Moving target detection method based on rough positioning of structure similarity and fine segmentation of background difference
Author:
Affiliation:

Fund Project:

National Natural Science Foundation of China (61971339); Key R&D Program of Shaanxi Provincial Department of Science and Technology (2019GY-113);

  • 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:

    针对抖动相机和静止相机下的运动目标检测问题,本文提出基于结构相似性粗定位与背景差分细分割的运动目标检测方法。首先使用动态模式分解法根据视频序列提取彩色背景图像为粗定位提供基础,提出在小范围内利用相关法对SIFT算子检测到的当前帧图像和彩色背景图像的特征点进行匹配,通过匹配点对的偏移量估计当前帧图像的偏移程度,以达到消除图像抖动的目的;然后利用结构相似性对目标区域粗定位,减少复杂背景的干扰;再对各通道下粗定位彩色背景图像及校正后的当前帧图像背景差分并对其结果进行与操作;最后通过形态学处理得到完整的运动目标。实验结果表明:本文方法不仅有效改善了相机的抖动问题,而且在抖动相机和静止相机两种情况下的检测率有所提高,与GMM等三种算法相比查全率和准确率分别提高1.6%、3.5%和3%以上。

    Abstract:

    Aiming at the problem of moving object detection in dither camera and still camera, This paper proposes a moving target detection method based on rough location of structure similarity and fine segmentation of background difference. First, the dynamic mode decomposition method is used to extract the color background image from the video sequence to provide the basis for coarse positioning. In a small range, the correlation method is used to match the feature points of the current frame image and the color background image detected by the SIFT operator, Estimate the offset degree of the current frame image by matching the offset of the point pair, in order to achieve the purpose of eliminating image jitter. Then use the structural similarity to coarsely locate the target area to reduce the interference of complex background; Then perform the AND operation on the color background image of the coarse positioning under each channel and the corrected background difference of the current frame image; Finally, the complete movement target is obtained through morphological processing. The method in this paper not only effectively improves the problem of camera shake, but also improves the detection rate in both cases of shaking camera and still camera. Compared with the three algorithms such as GMM, the recall rate, recall rate and accuracy rate are increased by at least 1.6%, 3.5% and 3%.

    参考文献
    相似文献
    引证文献
引用本文

蒙晓宇,朱磊,张博,等. 基于结构相似性粗定位与背景差分细分割的运动目标检测方法[J]. 科学技术与工程, 2021, 21(36): 15563-15570.
Meng Xiaoyu, Zhu Lei, Zhang Bo, et al. Moving target detection method based on rough positioning of structure similarity and fine segmentation of background difference[J]. Science Technology and Engineering,2021,21(36):15563-15570.

复制
分享
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:2021-05-10
  • 最后修改日期:2021-12-08
  • 录用日期:2021-08-20
  • 在线发布日期: 2022-01-04
  • 出版日期:
×
律回春渐,新元肇启|《科学技术与工程》编辑部恭祝新岁!
亟待确认版面费归属稿件,敬请作者关注