Abstract

This paper describes implementation of two methods for tracking and extraction of moving objects in successive frames. The motion of moving objects is detected as a temporal Laplacian image or a difference image from successive frames. On the basis of information on detected object motion, efficient algorithms of tracking of moving objects are considered for the cases that they overlap and unoverlap one another. A simple and fast algorithm for extracting moving objects is also considered, by which moving objects with gray levels alone are successively extracted in successive frames. Experiments based on simulated and real data are presented. The high speed digital image processing system developed, which is capable of processing moving objects in real time, plays an important role in implementing proposed methods.

© 1986 Optical Society of America

Detecting and tracking moving objects in long-distance imaging through turbulent medium

Eli Chen, Oren Haik, and Yitzhak Yitzhaky
Appl. Opt. 53(6) 1181-1190 (2014)

Single-pixel tracking of fast-moving object using geometric moment detection

Linbin Zha, Dongfeng Shi, Jian Huang, Kee Yuan, Wengweng Meng, Wei Yang, Runbo Jiang, Yafeng Chen, and Yingjian Wang
Opt. Express 29(19) 30327-30336 (2021)

Mask-based single-pixel tracking and imaging for moving objects

Yu Zhang, Hongjie Wang, Yongkai Yin, Wenjie Jiang, and Baoqing Sun
Opt. Express 31(20) 32554-32564 (2023)

Image-free real-time detection and tracking of fast moving object using a single-pixel detector

Zibang Zhang, Jiaquan Ye, Qiwen Deng, and Jingang Zhong
Opt. Express 27(24) 35394-35401 (2019)

Detection and tracking of small moving objects in image sequences by use of nonlinear spatiotemporal optical systems

Weiping Lu, Svetlana L. Lachinova, and Robert G. Harrison
Opt. Lett. 29(8) 824-826 (2004)