基於CamShift演算法於有亮度變化環境的多人物追蹤

CamShift Based Multiple Human Tracking in Light Changing Environment

賴昱圻(Yu-Chi Lai)

指導教授 ： 張志永

國立交通大學/電機學院/電機與控制工程系所/碩士(2015年)

https://doi.org/10.6842/NCTU.2015.00617

摘要

本篇論文實現一套亮度變化環境中的自動化人物追蹤系統，此系統著重於人臉偵測和上半身追蹤。藉由人臉偵測和上身衣服顏色追蹤的技術，可以即時掌控環境內每位成員的位置。
相對於人臉的位置有助於上身位置的獲得，因此我們以擷取的人臉位置做為相對位置的參考點，以得到上身的位置。接著，我們使用HSV色彩空間中的色相和飽和度，建立上身衣服顏色的直方圖模型，接著使用CamShift演算法進行上身的色彩追蹤。我們更使用了亮度偵測的技術，使得提出的追蹤系統能適應亮度變化的環境，若亮度變化夠大，則我們對上身色彩重新建模，以達到適應環境亮度變化。

關鍵字

追蹤 ； 人物追蹤

並列摘要

In this thesis, we implement an automatic human tracking system that combines face detection and upper outer garment tracking in the light changing environment. Face detection can detect each person in the screen; upper outer garment tracking can track each person’s location by upper outer garment’s color.
We use Haar cascade classifier and skin color detection for human face detection, which can locate the human face region correctly. Then, relative position with respect to a face founded will be used to locate one’s body region, and in the sequel hue-saturation color histogram model for upper garment is generated. We use CamShift algorithm to track person’s position. For better accuracy, we check the light change for each image frame and rebuild color model to adapt to the large enough light change in the environment.

並列關鍵字

Cam shift ； Mean shift ； tracking ； human tracking

參考文獻

[2] I. Haritaoglu, D. Harwood, and L. S. Davis, “W : Real-time surveillance of people and their activities,” IEEE Trans. Pattern Anal. Machine Intell., vol. 22, no. 8, pp. 809–830, Aug. 2000.

[3] P. Viola and M. Jones, “Robust Real-Time Face Detection,” Int. Journal Computer Vision, vol.57, no. 2, pp.137–154, Mar. 2004.

[6] John G. Allen, Richard Y. D. Xu and J. S. Jin, “Object Tracking Using CamShift Algorithm and Multiple Quantized Feature Spaces,” School of Information Technologies University of Sydney, 2006.

[7] P. Hidayatullah and H. Konik, “CAMSHIFT Improvement on Multi-Hue and Multi-Object Tracking,” in Proc. of IEEE International Conference on Electrical Engineering and Informatics, Paris, pp. 1–6, July 2011.

[11] P. Li, “An Adaptive Binning Color Model for Mean Shift Tracking,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 18, no. 9, pp. 1293–1299, Sep 2008.