使用行人偵測及追蹤之自動行人計數

Automatic Pedestrian Counting Using Pedestrian Detection and Tracking

姜禮玄(LI-HSUAN CHIANG)

指導教授 ： 張元翔

中原大學/電機資訊學院/資訊工程學系/碩士(2022年)

https://doi.org/10.6840/cycu202201395

摘要

人流計數已經是個重要的領域，得知人數對公共場所管理及決策是個重要的因素，在社會安全及控制管理方面發揮著十分重要的作用，我們提出一種基於頭部檢測及追蹤的人流計量方法，透過攝影機得到視訊影像，對觀察區進行人流計數，使用前景檢測取得運動中人的前景影像，並使用型態學方法優化，再利用基於 Haar-Like 特徵的 adaboost 分類器偵測頭部，並以 CSRT tracker 追蹤頭部，最後利用追蹤資訊完成計數工作,在複雜環境下準確率約為 85%實驗結果驗證了該方法的有效性。

關鍵字

運動偵測及追蹤 ； OpenCV ； 行人計數 ； 視訊監控

並列摘要

Pedestrian counting is already an important area, and the number of people is an important factor in the management and decision-making of public places, it plays a very important role in social security and control management. In this paper, we present a video surveillance system for automatic pedestrian counting, which includes pedestrian detection and pedestrian counting. The people flow is counted in the observation area, the foreground image of the moving person is obtained by foreground detection, and the morphological method is used to optimize, and then the adaboost classifier based on Haar-ike feature is used to detect the head, and use The CSRT tracker tracks the head ,and finally uses the tracking information to complete the counting work. Our results demonstrated a preliminary success with the accuracy of ~85%, indicating that our system may potentially be used in real scenarios.

並列關鍵字

Motion Detection and Tracking ； OpenCV ； Pedestrian Counting ； Video Surveillance

參考文獻

[1] N. Dalal and B. Triggs, "Histograms of oriented gradients for human detection," IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), pp.886-893, 2005.

[2] B. Wu and R. Nevatia, “Detection and tracking of multiple, partially occluded humans by Bayesian combination of edgelet based part detectors,” Int J Comput Vision, pp. 247-266, 2007.

[3] J. Wu, C. Geyer and J. M. Rehg, “Real-time human detection using contour cues,” IEEE International Conference on Robotics and Automationk, pp.860-867 2011.

[4] C. Gao, P. Li, Y. Zhang, L. Wang, “People counting based on head detection combining Adaboost and CNN in crowded surveillance environment,” Neurocomputing, Vol.208, pp.108-116, 2016.

[5] C. Papageorgiou and T. Poggio, “A trainable system for object detection,” International Journal of Computer Vision, pp. 15-33, 2000.