Video analysis provides a great way for students to understand all kinds of motion that is just too darn hard to measure any other way.
In an ecology class in 1975, my lab group and I ran around a flower field, following a bee, sticking a flag in the ground every time the bee landed, timing how long the bee stayed on the flower, shouting out numbers for the recorders to write down, and returning later to measure the positions of the flags. We used slide rules and the “sum of least squares” to see if there was any correlation between the time the bee spent on the flower (as an indicator of food quantity) and the angle and distance the bee flew to the next flower. We thought perhaps that a bee would try to stay in the area (sharper angle, short distance) when it had found a good food source. After a week of work we found the dreaded “no correlation”. Imagine how different that activity would look today. We would place a video camera over the flower field and clickity click click, collect some data. Open some graphing software and the analysis would be done in minutes.
Video is so easy to collect these days that students can record any kind of motion they want and apply newly learned physics rules to understand it. One of my students video-taped herself jumping a barrier on her horse. She discovered that she moved in such a way that she always kept herself over the center of mass of the horse. We had a great discussion about why this might be so.
Sometimes my students get bad results. I have learned a few tricks that increase the likelihood that they will get good data.
We used Vernier’s Logger Pro. It has a great tutorial. That is how I learned how to do video analysis and I use the tutorial to teach my students. Try it, you won’t be sorry.
In Logger Pro choose, File -> Open. Among the folders that contain the science experiments, you will find a folder called Tutorials.
Select 12 Video Analysis.cmbl. Follow the instructions and don’t skip anything. And tell you students not to skip anything.
The cursor that is used to locate the objects in each video frame is cross-hairs. Students usually think it best to put the center of the cursor on the center of the object. But the cursor obscures the view of the object somewhat and the center of the object is a judgement call. I tell them instead to place the top of the cursor on the top of the object.
Students should enlarge the movie screen as much as possible when locating the object in each frame. Click the movie, grab a corner and stretch the movie. After the locations have been determined, the movie can be made smaller again.
Students need to use an some object in the video of known size to set the scale. In the sample basketball video, there is a two-meter stick on the floor. If the students themselves are in the video, their height can be used.
It is important to place the meterstick the same distance away for the camera as the action. In the sample video, notice that the meterstick and the ball are the same distance.
Some of my students were having trouble with data, so I decided to see how much difference it would really make if the meterstick were too far or too close. In picture from the video shown here, you can see there are three metersticks. The center one is in the correct position, the others are too close or too far. My hidden associate, Ian, tosses a ball. Click on the link below the picture to watch the video.
Since we know the acceleration due to gravity is 9.80 m/s/s, we have a way to check our accuracy. The actual analysis screen shots are included at the end of the post. Here is a summary of the results of the analysis:
As you can see, the position of the reference meterstick makes quite a difference. It is an important consideration.
You might be wondering why the data from the correctly positioned meterstick didn’t give us a value for the vertical acceleration that was closer it -9.80 m/s/s. When I used the basketball video from the tutorial I got a value of -9.607 m/s/s. The results were better because the video was better.
The video I used was shot under low light levels. The camera collects light for a longer period of time for each frame when light levels are low and as a result, the image is blurred. This makes it difficult to get really accurate positions. The object in my video is small and hard to see. The basketball is much larger and easier. Sometimes we paint a white or black spot on the object to make it easier to see.
So, here is my list of things to do in order to get good video analysis results:
I usually have my students arrange the video, data table, and graphs on the screen and take a screen shot. These files are uploaded to a Google folder that is shared with me. This is there way of “turning in their lab” without wasteful printing. If you want more information about how to do this, see the previous posts on video analysis and the NSTA STEM14 tab for information on using Google Drive.
As always, you helpful comments are welcome.
Here are the images of the data analysis in case you would like to see them. Click on the images to see them enlarged.