Change search
ReferencesLink to record
Permanent link

Direct link
Measuring the Speed of a Floorball Shot Using Trajectory Detection and Distance Estimation With a Smartphone Camera: Using OpenCV and Computer Vision on an iPhone to Detect the Speed of a Floorball Shot
KTH, School of Computer Science and Communication (CSC).
2016 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesisAlternative title
Mätning av hastigheten på ett innebandyskott genom detektering av projektilbana och avståndsbedömning med kameran i en smartphone (Swedish)
Abstract [en]

This thesis describes the possibilities of using smartphones and their cameras in combination with modern computer vision algorithms to track and measure the speed of a floorball. Previous research within the area is described and an explanation is given as to why an implementation using three-frame temporal differencing to detect objects in motion works best to detect and track the ball.

100 floorball shots were recorded and measured using a speedometer radar and two different smartphones, one running the application and the other recording each shot. The video recording for each shot was then used to manually create a baseline for speed comparison. A second experiment was later conducted to analyse the sensitivity and effect on the determined ball size in the floorball shot analysis. The results from the first experiment show that the speedometer radar results in average deviate by 12% from the speed baseline. The speedshooting application however has results that, on average, deviate from the speed baseline by 6%. Furthermore, the results show that a faulty ball size detection is the major cause of error in the speedshooting application.

The main conclusion that can be drawn from this is that it is possible to use a smartphone and computer vision methodologies to determine the speed of a floorball shot. In fact, it is even possible to do so with greater accuracy than the radar used in the experiments in this thesis. However, to prove the accuracy of the application for normal use, further testing needs to be conducted in new experiment conditions, for example by recording shots at higher speeds than those recorded in the experiments in this thesis. 

Place, publisher, year, edition, pages
2016.
Keyword [en]
Computer Vision, Floorball, Bontouch, Ball, Trajectory, Speed, Motion Detection, Distance Estimation, Smartphone, Camera, OpenCV, iPhone, iOS
National Category
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-190823OAI: oai:DiVA.org:kth-190823DiVA: diva2:953092
Available from: 2016-08-18 Created: 2016-08-16 Last updated: 2016-08-18Bibliographically approved

Open Access in DiVA

fulltext(11606 kB)12 downloads
File information
File name FULLTEXT01.pdfFile size 11606 kBChecksum SHA-512
d475b7d0041f4a9b55b5659859d79782328a7bfcfedec6b7b12c5e8e16c75662b227667263c1fa1f50453cdf230bdbd2b77ee993f13b348fe00134762ac305cc
Type fulltextMimetype application/pdf

By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 12 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 17 hits
ReferencesLink to record
Permanent link

Direct link