多视角监控视频中动态目标的时空信息提取方法

doi:10.11947/j.AGCS.2022.20200507

测绘学报 ›› 2022, Vol. 51 ›› Issue (3): 388-400.doi: 10.11947/j.AGCS.2022.20200507

多视角监控视频中动态目标的时空信息提取方法

李景文^1,2, 韦晶闪¹, 姜建武^1,2, 陆妍玲¹, 刘垒², 唐一飞³, 李旭²

1. 桂林理工大学测绘地理信息学院, 广西桂林 541004;
2. 广西空间信息与测绘重点实验室, 广西桂林 541004;
3. 广西壮族自治区森林资源与生态环境监测中心, 广西南宁 530000

收稿日期:2020-10-30 修回日期:2021-11-20 发布日期:2022-03-30
通讯作者: 姜建武 E-mail:fengbuxi@glut.edu.cn
作者简介:李景文(1971-),男,博士生导师,研究方向为GIS理论和应用。E-mail:lijw@glut.edu.cn
基金资助:
国家自然科学基金（41961063）；国家文化和旅游科技创新工程（2019-011）；广西自然科学基金（2019GXNSFGA245001）

Spatio-temporal information extraction method for dynamic targets in multi-perspective surveillance video

LI Jingwen^1,2, WEI Jingshan¹, JIANG Jianwu^1,2, LU Yanling¹, LIU Lei², TANG Yifei³, LI Xu²

1. School of Surveying and Mapping and Geographic Information, Guilin University of Technology, Guilin 541004, China;
2. Guangxi Key Laboratory of Spatial Information and Mapping, Guilin 541004, China;
3. Guangxi Zhuang Autonomous Region Forest Resources and Ecological Environment Monitoring Center, Nanning 530000, China

Received:2020-10-30 Revised:2021-11-20 Published:2022-03-30
Supported by:
The National Natural Science Foundation of China(No. 41961063);The National Culture and Tourism Science and Technology Innovation Project (No. 2019-011); Guangxi Natural Science Foundation of China (No. 2019GXNSFGA245001)

摘要/Abstract

摘要： 监控视频中动态目标的精准定位与跟踪作为计算机视觉领域中重要的研究方向，近年来已成为监控领域的研究热点。传统视频动态目标检测仅依赖图像特征数据，忽略了与地理坐标系精准匹配，特别是对于多个摄像机覆盖的区域，拍摄的角度不同，投影后形成的图像空间分辨率也不同，因此，难以满足智能监控在复杂的地理场景中全方位时空信息感知。本文提出一种多摄像头协同的视频监控图像与地理空间数据互映射模型构建方法来获取动态目标的轮廓和地理位置等时空信息，首先建立监控图像信息与地理空间数据的互映射关系，将观测角度不同、尺度不同和空间分辨率的监控图像置于同一坐标系下，并在此基础上通过融合Canny算子与背景减法来检测目标的边缘信息；然后采用质心偏移算法还原目标在该场景的实际位置，从而实现多角度下连续跟踪，提升地理场景的时空理解力和分析力，提高动态目标的精准定位与跟踪能力。

关键词: 多视角, 动态目标检测, 特征信息提取, 地理空间映射

Abstract: As an important research direction in the field of computer vision, the precise positioning and tracking of dynamic targets in surveillance video has become a research hotspot in the surveillance field in recent years.Traditional video dynamic target detection relies only on image feature data, ignoring accurate matching with geographic coordinate systems.Especially for the area covered by multiple cameras, the spatial resolution of the image formed after projection is different when the shooting angle is different.Therefore, it is difficult to satisfy intelligent monitoring in complex geographical scenes with all-round spatio-temporal information perception.In this paper, we propose a collaborative video surveillance image and geospatial data inter-mapping model construction method to obtain spatio-temporal information such as the outline and geographic location of dynamic targets.Firstly, the mutual mapping relationship between surveillance image information and geospatial data is established, and the surveillance images with different observation angles, scales and spatial resolutions are placed under the same coordinate system.And based on this, the edge information of the target is detected by fusing Canny operator and background subtraction, and then the actual position of the target in the scene is restored by using the center-of-mass offset algorithm.This enables continuous tracking of targets under multiple angles, improves the spatio-temporal understanding and analysis of geographic scenes, and enhances the precise positioning and tracking of dynamic targets.

Key words: multi-view, dynamic target detection, feature information extraction, geospatial mapping

中图分类号:

P208
TP391

李景文, 韦晶闪, 姜建武, 陆妍玲, 刘垒, 唐一飞, 李旭. 多视角监控视频中动态目标的时空信息提取方法[J]. 测绘学报, 2022, 51(3): 388-400.

LI Jingwen, WEI Jingshan, JIANG Jianwu, LU Yanling, LIU Lei, TANG Yifei, LI Xu. Spatio-temporal information extraction method for dynamic targets in multi-perspective surveillance video[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(3): 388-400.

参考文献

[1] 张旭, 郝向阳, 李建胜, 等. 监控视频中动态目标与地理空间信息的融合与可视化方法[J]. 测绘学报, 2019, 48(11):1415-1423. DOI:10.11947/j.AGCS.2019.20180572. ZHANG Xu, HAO Xiangyang, LI Jiansheng, et al. Fusion and visualization method of dynamic targets in surveillance video with geospatial information[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48(11):1415-1423. DOI:10.11947/j.AGCS.2019.20180572.
[2] 管皓, 薛向阳, 安志勇. 深度学习在视频目标跟踪中的应用进展与展望[J]. 自动化学报, 2016, 42(6):834-847. GUAN Hao, XUE Xiangyang, AN Zhiyong. Advances on application of deep learning for video object tracking[J]. Acta Automatica Sinica, 2016, 42(6):834-847.
[3] 李志欣, 施智平, 张灿龙, 等. 混合生成式和判别式模型的图像自动标注[J]. 中国图象图形学报, 2015, 20(5):687-699. LI Zhixin, SHI Zhiping, ZHANG Canlong, et al. Hybrid generative/discriminative model for automatic image annotation[J]. Journal of Image and Graphics, 2015, 20(5):687-699.
[4] 朱文青, 刘艳, 卞乐, 等. 基于生成式模型的目标跟踪方法综述[J]. 微处理机, 2017, 38(1):41-47. ZHU Wenqing, LIU Yan, BIAN Le, et al. Survey on object tracking method base on generative model[J]. Microprocessors, 2017, 38(1):41-47.
[5] 陈旭, 孟朝晖. 基于深度学习的目标视频跟踪算法综述[J]. 计算机系统应用, 2019, 28(1):1-9. CHEN Xu, MENG Zhaohui. Survey on video object tracking algorithms based on deep learning[J]. Computer Systems & Applications, 2019, 28(1):1-9.
[6] HORN B K P, SCHUNCK B G. Determining optical flow[J]. Artificial Intelligence, 1981, 17(1-3):185-203.
[7] MEI Xue, LING Haibin. Robust visual tracking and vehicle classification via sparse representation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(11):2259-2272.
[8] CRUZ-MOTA J, BOGDANOVA I, PAQUIER B, et al. Scale invariant feature transform on the sphere:theory and applications[J]. International Journal of Computer Vision, 2012, 98(2):217-241.
[9] 李玺, 查宇飞, 张天柱, 等. 深度学习的目标跟踪算法综述[J]. 中国图象图形学报, 2019, 24(12):2057-2080. LI Xi, ZHA Yufei, ZHANG Tianzhu, et al. Survey of visual object tracking algorithms based on deep learning[J]. Journal of Image and Graphics, 2019, 24(12):2057-2080.
[10] 欧阳谷, 钟必能, 白冰,等. 深度神经网络在目标跟踪算法中的应用与最新研究进展[J]. 小型微型计算机系统, 2018, 39(2):315-323. OUYANG Gu, ZHONG Bineng, BAI Bing, et al. Recent research advances and application of object tacking algorithm based on deep neural network[J]. Journal of Chinese Computer Systems, 2018, 39(2):315-323.
[11] HARE S, GOLODETZ S, SAFFARI A, et al. Struck:structured output tracking with kernels[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(10):2096-2109.
[12] MA Chao, HUANG Jiabin, YANG Xiaokang, et al. Hierarchical convolutional features for visual tracking[C]//Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago. Chile:IEEE, 2015:3074-3082.
[13] 王志军, 马凯. 基于成像角度的特征点质心提取精度研究[J]. 激光杂志, 2018, 39(4):90-93. WANG Zhijun, MA Kai. Study on the precision of feature point centroid extraction based on imaging angle[J]. Laser Journal, 2018, 39(4):90-93.
[14] 王敏, 赵金宇, 陈涛. 基于各向异性高斯曲面拟合的星点质心提取算法[J]. 光学学报, 2017, 37(5):226-235. WANG Min, ZHAO Jinyu, CHEN Tao. Center extraction method for star-map targets based on anisotropic Gaussian surface fitting[J]. Acta Optica Sinica, 2017, 37(5):226-235.
[15] 王宇岚, 孙韶媛, 刘致驿, 等. 基于多视角融合的夜间无人车三维目标检测[J]. 应用光学, 2020, 41(2):296-301. WANG Yulan, SUN Shaoyuan, LIU Zhiyi, et al. Nighttime three-dimensional target detection of driverless vehicles based on multi-view channel fusion network[J]. Journal of Applied Optics, 2020, 41(2):296-301.
[16] 胡玉兰, 石心蕊. 基于多摄像头协同的目标融合[J]. 电子世界, 2019(13):58-59. HU Yulan, SHI Xinrui. Object fusion based on multi-camera collaboration[J]. Electronics World, 2019(13):58-59.
[17] 张兴国. 地理场景协同的多摄像机目标跟踪研究[D]. 南京:南京师范大学, 2014. ZHANG Xingguo. Object tracking methods based on geographic scene and multi-camera[D]. Nanjing:Nanjing Normal University, 2014.
[18] 刘垒. 基于地理约束场景的动态目标检测及行为识别方法研究[D]. 桂林:桂林理工大学, 2020. LIU Lei. Research of dynamic target detection and behavior recognition method based on geographic constraint scene[D]. Guilin:Guilin University of Technology, 2020.
[19] ZHANG Z. A flexible new technique for camera calibration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(11):1330-1334.
[20] 李莉. OpenCV耦合改进张正友算法的相机标定算法[J]. 轻工机械, 2015, 33(4):60-63, 68. LI Li. Camera calibration algorithm based on OpenCV and improved Zhang zhengyou algorithm[J]. Light Industry Machinery, 2015, 33(4):60-63, 68.
[21] 刘艳, 李腾飞. 对张正友相机标定法的改进研究[J]. 光学技术, 2014, 40(6):565-570. LIU Yan, LI Tengfei. Reaserch of the improvement of Zhang zhengyou camera calibration method[J]. Optical Technique, 2014, 40(6):565-570.
[22] 李云功. 基于轮廓特征描述的目标识别算法研究[D]. 沈阳:沈阳理工大学, 2019. LI Yungong. Research on object recognition algorithm based on contour feature description[D]. Shenyang:Shenyang Ligong University, 2019.
[23] WANG Shengchun, WANG Hao, ZHOU Yunlai, et al. Automatic laser profile recognition and fast tracking for structured light measurement using deep learning and template matching[J]. Measurement, 2021, 169:108362.
[24] 华媛蕾, 刘万军. 改进混合高斯模型的运动目标检测算法[J]. 计算机应用, 2014, 34(2):580-584. HUA Yuanlei, LIU Wanjun. Moving object detection algorithm of improved Gaussian mixture model[J]. Journal of Computer Applications, 2014, 34(2):580-584.
[25] 徐亮, 魏锐. 基于Canny算子的图像边缘检测优化算法[J]. 科技通报, 2013, 29(7):127-131, 150. XU Liang, WEI Rui. An optimal algorithm of image edge detection based on canny[J]. Bulletin of Science and Technology, 2013, 29(7):127-131, 150.

多视角监控视频中动态目标的时空信息提取方法

Spatio-temporal information extraction method for dynamic targets in multi-perspective surveillance video

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价

[1]	方莉娜, 沈贵熙, 游志龙, 郭迎亚, 付化胜, 赵志远, 陈崇成. 融合点云和多视图的车载激光点云路侧多目标识别[J]. 测绘学报, 2021, 50(11): 1558-1573.
[2]	赵传, 郭海涛, 卢俊, 余东行, 张保明. 基于深度残差网络的机载LiDAR点云分类[J]. 测绘学报, 2020, 49(2): 202-213.
[3]	张旭, 郝向阳, 李建胜, 李朋月. 监控视频中动态目标与地理空间信息的融合与可视化方法[J]. 测绘学报, 2019, 48(11): 1415-1423.