Methodology and Deterministic modelling

For the methodology, I have chosen to carry it out on the numerical analysis. When setting up the Go Pro, it was placed upon a tripod for support and stability which enabled the camera to remain stable mid shot. The camera lens was in parallel to the front of the tripod leg. This was because when analysed the angle of shots, there was something to determine the prime angle of the camera which had the whole action in shot. In order to get an accurate reading, we set up the camera in a sports hall against a plain background. This was due to the fact that the athlete would have markers on their joints so the background wouldn’t interfere with the action. Also it would be more of an accurate representation as the heights and angles can be analysed clearly compared to a still background rather than moving or busy backgrounds. 

In order to assess if the athlete was in the shot, we had to use scalers which would work out specific distances. This was done by placing a meter stick in the field of interest which then the measurements from that can be taken and placed in the kinoeva software to workout angles and distances. Once the scalers were in place, we had to place anatomical markers on the joints that were in use. In this case, joint markers were places on the toe, ankle, knee, hip, wrist, elbow and shoulder. This was done so that the specific joints used in the process were visible when analysing in kinoeva the angle of elbows/release and heights of release. If there were no markers to mark the joints, this would be more difficult to analyse as you wouldn’t know where to measure it from. To enable that the readings were correct, we recorded the same skill 3 times. This gave us an idea on which video is the clearest and most accurate to use.

In order to analyse the videos taken, we then imported them into kinovea to analyse the height of release, angle of release and angle of elbow. In order to analyse videos, I had to go through a number of steps which would prevent particular errors and improve the reliability and accuracy of the data imported. I imported my chosen video out of the 3 and also the scaling video. This was done so then we could work out the scaling factor of 0.5m. From this, we were able to create the angles and distances to a correct measurement. Once that scale was set, we removed it as this could have due to perspective error. By getting the correct scaling factor of 0.5m rather than measuring it from the video itself, it made our measurements more accurate.

Whilst in the process of videoing and analysing these videos we had to take into account parallax and perspective error. Parallax Error can be described as when a direction or the position of the same object differentiates when looking in different distances and positions, which would lead to getting a different reading of angles. In order to reduce this error, the athlete should be positioned well in shot and in the middle plane of view which then can minimise the results being incorrect. Also, another method to reduce this error can be to repeat the same skill more than once so then the internal validity of the results can increase to be more accurate. Error was also reduced by placing markers on the athlete so the movements were clear against the plain background. This was so that all the attention was focussed on the movements produced. Perspective error is known as when the correct height, width, depth and position is right when it comes to looking at three-dimensional objects and/or a two-dimensional surface. In order to reduce error for this we had to make sure the person didn’t move out of shot otherwise the focus of the lens would be shifted. We also needed to make sure that we maximise camera-subject distance whilst maximising the size of subject within field of view. This can be done by placing the camera at a perpendicular angle facing the athlete but with a distance away suitable. By reducing perspective error, we are keeping the results accurate; if the athlete moved closer or further away when completing the action, the measurements would change. This why we would place is far away and zoom in.

What is deterministic modelling?

It is a combination of all the biomechanical factors that determine an action or a movement. It starts off with primary levels and then broken down further into secondary, tertiary, quaternary and quinary. To start off with, you would look at the primary aspects like time, it would then be broken down to something like average speed or time taken, tertiary level would include stride lengths and stride frequency, and finally the quaternary factors can be take-off, and flight and landing distance as well as stride time. Finally, quinary factors include air resistance, angle/height of take-off and speed of take-off.

Why and how would this model be used?

It is used so then the breakdown of factors can be easier for the athlete and coach to plan a skill in finer details but also know what to work on and what strengths and weaknesses the athlete displays.It could include looking at the qualitative analysis, analysing the rhythm and posture of the motion, or by quantitatively looking into the sprint, watching the speed and distance covered. It would be used in training sessions so each component can be looked at and improved if needed. The coach would then analyse the relationship of skill performed to the motion produced.