應用顏色濾波器在即時影像中追蹤籃球

A Real-time Basketball Tracking System by Colour Filter.

邱柏翔(Po-Hsiang, Chiu)

指導教授 ： 涂世雄

中原大學/電機資訊學院/電機工程研究所/碩士(2018年)

https://doi.org/10.6840/cycu201800544

摘要

摘要:
本論文提出了一個簡單的籃球追蹤方法，利用顏色濾波器，依照各種不同顏色的籃球作為參考值，來達到追蹤的效果，希望對球賽的數據統計有所幫助。
本研究分為三個部分。第一部分我們將即時影像中得到的籃球框出並用顏色濾波器得到它的HSV參數值來進行追蹤。第二部分為影片球員的各種攻防精彩片段的擷取，如灌籃、封阻、高難度投球、抄截和失誤等等，搭配籃球的追蹤，透過片段的擷取收集出一個球員精彩片段的數據統計輔助系統。第三部分為實驗結果，我們將上述的這些片段建構成資料庫並分類，接著使用者就可以依照此分類來觀看擁有即時籃球追蹤的各種精彩攻防畫面的呈現。
本論文研究貢獻如下：
1. 省時性：比起利用整個影片回放的功能，這個系統可以讓使用者更快速的獲取比賽內的精彩片段，省去回放尋找的時間。
2. 便利性：這個系統可以記錄所有的精彩片段，讓使用者可以任意選取不同的片段，方便作為一個即時性的剪輯。
3. 準確性：利用顏色濾波器來追蹤籃球的效果在即時影像中表現的非常準確。

關鍵字

顏色濾波器 ； 影像 ； 追蹤 ； 球

並列摘要

Abstract:
In this thesis, we propose a real-time basketball tracking system by colour filter. According to different colors of basketballs, we use the colour filter to track the ball in the real-time to reach the purpose of establishing the statistics data and a highlight auxiliary system of a game.
There are three parts in this thesis. In the first part, we will circle the basketball and use colour filter to obtain its HSV parameters to track the ball. The second part is the capturing of highlights, like dunks, blocks, difficult shots, steals and turnovers, etc.. With ball tracking and frame capturing, we build the data statistics and auxiliary system of players’ highlights. The last part is the presentation of the experimental results by OpenCV.
The contributions in this thesis are as follows.
1. Time-saving: Comparing with the use of the video replay to watch the highlights, this system allows users to obtain the highlights films much faster.
2. Convenience: This system can capture all the highlights, allowing users can select the different film fragments arbitrarily and even can make the immediacy clips conveniently.
3. Accuracy: The simulation results tells us that our method can capture the dynamics of ball correctly by using the colour filter for tracking the basketball in the real-time.

並列關鍵字

colour filter ； tracking

參考文獻

References:

[1]. Misha Denil; Loris Bazzani; Hugo Larochelle; Nando de Freitas, “Learning Where to Attend with Deep Architectures for Image Tracking”, pp.2151-2184, Neural Computation (Volume: 24, Issue: 8, Aug. 2012).

[2]. Kuo-Hsien Hsia; Shao-Fan Lien; Juhng-Perng Su; Wei-Yi Ciou, “ANFIS-based aided stabilization control and image tracking for mobile robot”, 2012 International conference on Fuzzy Theory and Its Applications (iFUZZY2012), 16-18 Nov, 2012.

[3]. Thomas Kinsman; Peter Bajorski; Jeff B. Pelz, “Hierarchical image clustering for analyzing eye tracking videos”, 2010 Western New York Image Processing Workshop, 5 Nov, 2010.

[4]. R.N. Anderton; R. Appleby; J.R. Borrill; D.G. Gleed; S. Price; N.A. Salmon; G.N. Sinclair; A.H. Lettington, “Prospects of imaging applications [military]”, IEE Colloquium on Terahertz Technology and Its Applications (Digest No: 1997/151), 10 April, 1997.